HydrogenAudio

Hydrogenaudio Forum => Scientific Discussion => Topic started by: William on 2005-12-29 12:45:48

Title: Why 24bit/48kHz/96kHz/
Post by: William on 2005-12-29 12:45:48
Yes, I have searched the forum.
Yes, maybe I am dumb.

But it seems I cannot find the answer.

Why do we need 24bit/48kHz/96kHz/192kHz if 16bit/44.1kHz is good enough? Are there any situations that 16bit/44.1kHz simply cannot satisfy? In other words, is there any real need for the higher bit depth and sampling rate?

Thanks for answering.
Title: Why 24bit/48kHz/96kHz/
Post by: Lyx on 2005-12-29 12:53:12
Quote
Why do we need 24bit/48kHz/96kHz/192kHz if 16bit/44.1kHz is good enough?
[a href="index.php?act=findpost&pid=353038"][{POST_SNAPBACK}][/a]

We dont need it. It's just virtual useless number-games to give people the incentive to buy new equipment and then re-buy all our music. There are some *technical* arguments for using 48khz instead of 44khz.... but the actual benefit for normal endusers is zero.
Title: Why 24bit/48kHz/96kHz/
Post by: William on 2005-12-29 13:09:39
I heard someone saying that increasing the sampling rate improves the SNR. I simply don't get it. It would be grateful if someone can enlighten me on this as well.

Thank you.
Title: Why 24bit/48kHz/96kHz/
Post by: Garf on 2005-12-29 13:13:22
Quote
I heard someone saying that increasing the sampling rate improves the SNR. I simply don't get it. It would be grateful if someone can enlighten me on this as well.

Thank you.
[a href="index.php?act=findpost&pid=353047"][{POST_SNAPBACK}][/a]


Normally, improving sampling rate improves bandwidth, not SNR. SNR is improved by increasing bit depth. Noise shaping allows trading bandwidth for effective SNR.

The question is why one would need a SNR of >96dB or a bandwidth of over 22kHz for end user playback.
Title: Why 24bit/48kHz/96kHz/
Post by: William on 2005-12-29 13:22:10
Quote
Normally, improving sampling rate improves bandwidth, not SNR. SNR is improved by increasing bit depth. Noise shaping allows trading bandwidth for effective SNR.

This is exactly what I am wondering...

What I learnt is that, in frequency domain, the higher the sampling rate, the larger the bandwidth, and the farther away between the base band and its images..

And I see nothing related to SNR by improving sampling rate.

Would you please give me more information on noise shaping?

Quote
The question is why one would need a SNR of >96dB or a bandwidth of over 22kHz for end user playback.

Then why DVD audio uses 96kHz? This is something I always wonder.

Thank you.
Title: Why 24bit/48kHz/96kHz/
Post by: bizangoin on 2005-12-29 13:32:17
I totally agree with you all. Increasing sample rate and bit depth is not perceptible, excepting for some animals.

Sony Super Audio Compact Disc (SACD) is the fine high end example of the no-use technology. Sony claims to consumers his high sample rate, bandwidth etc but do not claim that 0.00001% could hear the difference.

Moreover, even if you have a HD Player with 100 kHz bandwidth, you have to get the same characteristics for the whole audio elements (amplifiers, loudspeakers).

"The audio quality of a sound equipment set is equal to the audio quality of the worst device."
Title: Why 24bit/48kHz/96kHz/
Post by: Garf on 2005-12-29 13:45:53
Quote
Would you please give me more information on noise shaping?
[a href="index.php?act=findpost&pid=353055"][{POST_SNAPBACK}][/a]


Use the search function.
Title: Why 24bit/48kHz/96kHz/
Post by: William on 2005-12-29 13:59:01
I am sorry. Thanks.
Title: Why 24bit/48kHz/96kHz/
Post by: singaiya on 2005-12-29 17:44:20
Quote
Why do we need 24bit/48kHz/96kHz/192kHz if 16bit/44.1kHz is good enough? Are there any situations that 16bit/44.1kHz simply cannot satisfy? In other words, is there any real need for the higher bit depth and sampling rate?

Thanks for answering.
[a href="index.php?act=findpost&pid=353038"][{POST_SNAPBACK}][/a]


I have read that if you are recording music and performing several layers of edits and signal processing, that higher sampling and/or bit depth may be of actual value to the final mixdown. I don't do that kind of work, so that is hearsay pretty much. For simple audio playback you are correct, 16/44.1 is totally sufficient.
Title: Why 24bit/48kHz/96kHz/
Post by: rosshmusic on 2005-12-29 20:32:09
I agree that it makes no difference to end users...  though there are technical reasons why this would have its advantages...

most of it comes from the way hardware plays back music.... it has to do with the enforcements of peaks... each sample is essentially enforced as a peak in the reproduced sound wave... but this should NOT be the case... the sample may have been taken along any part of the wave (and most likely not at a peak)...

The affects of this are mostly negligable and hardware also works to minimize the affects... but it shouldn't have to... and higher sampling rates allow the hardware to start  with a closer reresentation to the original wave...

I'll look for some articles about this "phenom"... I actually looked them up a year or two ago cause an audiophile engineer I know was telling me about it and I didn't believe him...

in the end the LARGE majority people would never be able to tell... and I think most of the reason CD's sound like s**t to me is due to gawd awful compression they fell they need to put on the CD's 

Peace
Ross
Title: Why 24bit/48kHz/96kHz/
Post by: Garf on 2005-12-29 20:46:23
Quote
I agree that it makes no difference to end users...  though there are technical reasons why this would have its advantages...

most of it comes from the way hardware plays back music.... it has to do with the enforcements of peaks... each sample is essentially enforced as a peak in the reproduced sound wave... but this should NOT be the case... the sample may have been taken along any part of the wave (and most likely not at a peak)...

The affects of this are mostly negligable and hardware also works to minimize the affects... but it shouldn't have to... and higher sampling rates allow the hardware to start  with a closer reresentation to the original wave...

I'll look for some articles about this "phenom"... I actually looked them up a year or two ago cause an audiophile engineer I know was telling me about it and I didn't believe him...
[a href="index.php?act=findpost&pid=353155"][{POST_SNAPBACK}][/a]


This really doesn't make any sense at all.
Title: Why 24bit/48kHz/96kHz/
Post by: bug80 on 2005-12-29 21:16:47
Quote
I have read that if you are recording music and performing several layers of edits and signal processing, that higher sampling and/or bit depth may be of actual value to the final mixdown. I don't do that kind of work, so that is hearsay pretty much.

Exactly. In the digital domain you make rounding errors due to the quantization. If an engineer does all the processing in the 16 bit domain, these rounding errors will build up and therefore decrease the signal to noise ratio. If all the processing is done in the 24 bit domain and the audio is quantized to 16 bits per sample afterwards, these rounding errors will be negligible.

A higher sampling rate is not really necessary.

Quote
For simple audio playback you are correct, 16/44.1 is totally sufficient.[a href="index.php?act=findpost&pid=353114"][{POST_SNAPBACK}][/a]

I also think that's the case. I don't think many people are able to ABX 24 bit versus 16 bit audio, for example. Or 44.1 kHz versus 48 kHz.

edit: I can imagine that a higher sampling rate can be of use if multiple anti-aliasing filters exist in a recording chain. Each anti-aliasing filter will remove high frequencies, so multiple filters may remove too much energy from the high frequency bands. However, in most studios only one A/D converter with anti-aliasing filter is present in the chain (for example in the mixing desk).
Title: Why 24bit/48kHz/96kHz/
Post by: listen on 2005-12-29 21:24:21
I had a thread here: http://www.hydrogenaudio.org/forums/index....showtopic=17118 (http://www.hydrogenaudio.org/forums/index.php?showtopic=17118)

It is quite an indirect and meandering thread (experimentation, disbelief etc), but amongst all of it are some definite abx results.

The trouble is, there is no way I can say for certain that i am not simply hearing large errors introduced by my headphones.  And since headphones on the market do not seem to be designed for such high resolution there is not much i can do about that.
Title: Why 24bit/48kHz/96kHz/
Post by: Tool462 on 2005-12-29 21:50:26
ok, i'm a newb so don't give me some slack, but wouldn't that mean that CDs are the last audio format we'd need for the foreseeable future? I mean I know from previous posts that DVD Audio is useless, so anything better than CDs with only 2 channels is pretty useless.
Title: Why 24bit/48kHz/96kHz/
Post by: rosshmusic on 2005-12-29 21:55:14
Quote
Quote
I agree that it makes no difference to end users...  though there are technical reasons why this would have its advantages...

most of it comes from the way hardware plays back music.... it has to do with the enforcements of peaks... each sample is essentially enforced as a peak in the reproduced sound wave... but this should NOT be the case... the sample may have been taken along any part of the wave (and most likely not at a peak)...

The affects of this are mostly negligable and hardware also works to minimize the affects... but it shouldn't have to... and higher sampling rates allow the hardware to start  with a closer reresentation to the original wave...

I'll look for some articles about this "phenom"... I actually looked them up a year or two ago cause an audiophile engineer I know was telling me about it and I didn't believe him...
[{POST_SNAPBACK}][/a] (http://index.php?act=findpost&pid=353155")


This really doesn't make any sense at all.
[a href="index.php?act=findpost&pid=353156"][{POST_SNAPBACK}][/a]


heh... I thought the same thing when I read it back... I'm not the best to explain obviously... I didn't have time (and can't recall where) to find the resources we found talking about this... but I did find this one (which is not the one I remember but talks about the same thing)....
[a href="http://www.themusicpage.org/articles/SamplingTheory.html]http://www.themusicpage.org/articles/SamplingTheory.html[/url]

any better?
Title: Why 24bit/48kHz/96kHz/
Post by: Garf on 2005-12-29 22:58:19
Quote
http://www.themusicpage.org/articles/SamplingTheory.html (http://www.themusicpage.org/articles/SamplingTheory.html)

any better?
[a href="index.php?act=findpost&pid=353170"][{POST_SNAPBACK}][/a]


A paper where the author doesn't seem to correctly understand Nyquist is not an acceptable argument, no.
Title: Why 24bit/48kHz/96kHz/
Post by: Garf on 2005-12-29 23:01:19
Quote
ok, i'm a newb so don't give me some slack, but wouldn't that mean that CDs are the last audio format we'd need for the foreseeable future? I mean I know from previous posts that DVD Audio is useless, so anything better than CDs with only 2 channels is pretty useless.
[a href="index.php?act=findpost&pid=353167"][{POST_SNAPBACK}][/a]


Well, the advantage of DVDA and SACD is exactly that they are multichannel. And have DRM, which is obviously a big advantage to some people.
Title: Why 24bit/48kHz/96kHz/
Post by: rosshmusic on 2005-12-30 01:59:10
Quote
Quote
http://www.themusicpage.org/articles/SamplingTheory.html (http://www.themusicpage.org/articles/SamplingTheory.html)

any better?
[a href="index.php?act=findpost&pid=353170"][{POST_SNAPBACK}][/a]


A paper where the author doesn't seem to correctly understand Nyquist is not an acceptable argument, no.
[a href="index.php?act=findpost&pid=353188"][{POST_SNAPBACK}][/a]

where does he go askew .... ?
Title: Why 24bit/48kHz/96kHz/
Post by: Revliskciuq on 2005-12-30 02:11:51
I have to step out and disagree that there is no audible difference between 16-bit and 24-bit recordings. The decision to use 16/44.1 was arrived at, not because it was audio nirvana, but because it was the best comprimise possible with the digital to analogue conversion technology of the 80's.

To hear the benifits of 24-bit vs. 16-bit, you need a few things:

1) Capable and descriminating ears. The fact is that no everyone can hear the difference, and that's fine. However, even with a capable ear, the individual has to know what he's looking for. We're talking about very subtle points here, the difference is not going to be night and day.

2) Capable audio equipment. Just because you have a DVD-A or SACD player doesn't mean your system is capable of delivering an audible difference. You need very good equipment to make 24-bit listening worth your while.

3) Quality recording. It doesn't matter if the recording is 64-bits - if it was poorly recorded, and poorly produced, it's going to sound poor. Adding another 8 bits is not a magical fix-all.

Quote
Well, the advantage of DVDA and SACD is exactly that they are multichannel. And have DRM, which is obviously a big advantage to some people.


I disagree with this, especially with respect to SACD. Alot of hardcore audiophiles (the people who would be investing in expensive high bit recordings) feel very strongly that music should only bet two channels.

In SACD multichannel is optional, but 2-channel is required; in other words every SACD will have high-bit two channel audio, but not necessarily multichannel.
Title: Why 24bit/48kHz/96kHz/
Post by: William on 2005-12-30 03:03:09
OK, so here are some more technical questions I found...

1) The 44.1kHz has its historical reasons, but how about 48kHz, 96kHz, or even 192kHz?
How are these numbers chosen? Any technical and practical advantage over 44.1kHz? And why creative cards resamples to these sampling rates?

2) 16bits, 24bits, 32bits, etc, I think these values are chosen because they form complete bytes (1 byte = 8 bits). Or are there any other reasons for the numbers higher than 16bits? Again, any technical and practical advantages over 16bits, besides keeping accuracy and preventing errors from various quantization before final down-mixing?

What I only see is that, these higher numbers mean more data is sampled and stored with higher accuracy, and thus gives theoretically higher quality than 16bits/44.1kHz. But, oh well, they are only theoretical after all. Practically we may not hear a single difference...
Title: Why 24bit/48kHz/96kHz/
Post by: bug80 on 2005-12-30 12:54:18
Quote
where does he go askew .... ?
[a href="index.php?act=findpost&pid=353233"][{POST_SNAPBACK}][/a]

The Nyquist theorem states that frequencies below the Nyquist frequency can be sampled correctly, not at the Nyquist frequency, to begin with.
Title: Why 24bit/48kHz/96kHz/
Post by: Garf on 2005-12-30 12:59:37
Quote
I have to step out and disagree that there is no audible difference between 16-bit and 24-bit recordings. The decision to use 16/44.1 was arrived at, not because it was audio nirvana, but because it was the best comprimise possible with the digital to analogue conversion technology of the 80's.

To hear the benifits of 24-bit vs. 16-bit, you need a few things:

1) Capable and descriminating ears. The fact is that no everyone can hear the difference, and that's fine. However, even with a capable ear, the individual has to know what he's looking for. We're talking about very subtle points here, the difference is not going to be night and day.

2) Capable audio equipment. Just because you have a DVD-A or SACD player doesn't mean your system is capable of delivering an audible difference. You need very good equipment to make 24-bit listening worth your while.

3) Quality recording. It doesn't matter if the recording is 64-bits - if it was poorly recorded, and poorly produced, it's going to sound poor. Adding another 8 bits is not a magical fix-all.
[a href="index.php?act=findpost&pid=353234"][{POST_SNAPBACK}][/a]


Practise disagrees with your suppositions. Is there any proof that we can hear beyond 96dB SNR?

Practical experiments have shown most listeners already get into problems hearing to the 13th bit (78db SNR). If you think you (or another person) can do better, please let them take the MAD challenge with whatever equipment and ears you want, come back aftwards, and see if you would still tell me there is a practical reason to have more.

For the record, claiming you hear over 100dB SNR equals claiming you can hear the person next to you breathing while listening to a live rock concert. Still sounds reasonable?
Title: Why 24bit/48kHz/96kHz/
Post by: boombaard on 2005-12-30 13:24:17
so, on a slightly different note, would the 24/96 recordings be harder to saturate (eg. to master crappily and make it clip?)

if so, it might be preferable with all the 'new' music that comes out
Title: Why 24bit/48kHz/96kHz/
Post by: Garf on 2005-12-30 13:25:35
Quote
so, on a slightly different note, would the 24/96 recordings be harder to saturate (eg. to master crappily and make it clip?)

if so, it might be preferable with all the 'new' music that comes out
[a href="index.php?act=findpost&pid=353307"][{POST_SNAPBACK}][/a]


No, it's exactly the same as with CD.

People have already noted that many SACD and DVDA recordings are just as clipped and compressed as CD's.
Title: Why 24bit/48kHz/96kHz/
Post by: boombaard on 2005-12-30 13:37:59
Quote
Quote
so, on a slightly different note, would the 24/96 recordings be harder to saturate (eg. to master crappily and make it clip?)

if so, it might be preferable with all the 'new' music that comes out
[a href="index.php?act=findpost&pid=353307"][{POST_SNAPBACK}][/a]


No, it's exactly the same as with CD.

People have already noted that many SACD and DVDA recordings are just as clipped and compressed as CD's.
[a href="index.php?act=findpost&pid=353308"][{POST_SNAPBACK}][/a]


how odd.. since the SNR is that much higher, it can't be that they just master it for sacd/dvda and then use that same mastering for cd, since it would, if i'm not mistaken, be clipped much worse than on the 24/bla version..
i wonder what the point of their advertising with 'higher quality' is it's basically just higher volume out of the box
Title: Why 24bit/48kHz/96kHz/
Post by: Garf on 2005-12-30 14:13:30
Quote
how odd.. since the SNR is that much higher, it can't be that they just master it for sacd/dvda and then use that same mastering for cd, since it would, if i'm not mistaken, be clipped much worse than on the 24/bla version..
i wonder what the point of their advertising with 'higher quality' is it's basically just higher volume out of the box
[a href="index.php?act=findpost&pid=353312"][{POST_SNAPBACK}][/a]


Clipping has to do with maximum sample values, not with SNR.

The differences between CD and other formats are completely irrelevant for that. The clipping is not a technical problem. It's one of being AS LOUD AS POSSIBLE. Mastering engineers are REDUCING the dynamic range. That's yet another reason "better" formats are useless.
Title: Why 24bit/48kHz/96kHz/
Post by: boombaard on 2005-12-30 14:46:13
Quote
Quote
how odd.. since the SNR is that much higher, it can't be that they just master it for sacd/dvda and then use that same mastering for cd, since it would, if i'm not mistaken, be clipped much worse than on the 24/bla version..
i wonder what the point of their advertising with 'higher quality' is it's basically just higher volume out of the box
[a href="index.php?act=findpost&pid=353312"][{POST_SNAPBACK}][/a]


Clipping has to do with maximum sample values, not with SNR.

The differences between CD and other formats are completely irrelevant for that. The clipping is not a technical problem. It's one of being AS LOUD AS POSSIBLE. Mastering engineers are REDUCING the dynamic range. That's yet another reason "better" formats are useless.
[a href="index.php?act=findpost&pid=353319"][{POST_SNAPBACK}][/a]


yeah.. forgive me for putting that a bit awkwardly, but as i understand it the things clip because of the 'saturation', which is directly linked to the bit depth, correct? and they clip because of the volume being too high, so in that regard sacd should be harder to saturate than redbook cd's.. correct?
Title: Why 24bit/48kHz/96kHz/
Post by: jussi on 2005-12-30 14:52:08
well, and what about the HDCD (20bit) technology? what are the other 4bits used for?
Title: Why 24bit/48kHz/96kHz/
Post by: William on 2005-12-30 14:54:54
Well, it sounds like these 24bits / whatever kHz are simply marketing gimmicks than really practically useful for the end user.
Title: Why 24bit/48kHz/96kHz/
Post by: Garf on 2005-12-30 14:56:39
Quote
Quote
Quote
how odd.. since the SNR is that much higher, it can't be that they just master it for sacd/dvda and then use that same mastering for cd, since it would, if i'm not mistaken, be clipped much worse than on the 24/bla version..
i wonder what the point of their advertising with 'higher quality' is it's basically just higher volume out of the box
[a href="index.php?act=findpost&pid=353312"][{POST_SNAPBACK}][/a]


Clipping has to do with maximum sample values, not with SNR.

The differences between CD and other formats are completely irrelevant for that. The clipping is not a technical problem. It's one of being AS LOUD AS POSSIBLE. Mastering engineers are REDUCING the dynamic range. That's yet another reason "better" formats are useless.
[a href="index.php?act=findpost&pid=353319"][{POST_SNAPBACK}][/a]


yeah.. forgive me for putting that a bit awkwardly, but as i understand it the things clip because of the 'saturation', which is directly linked to the bit depth, correct? and they clip because of the volume being too high, so in that regard sacd should be harder to saturate than redbook cd's.. correct?
[a href="index.php?act=findpost&pid=353333"][{POST_SNAPBACK}][/a]


No, no, no, no....please reread my post. You could have a 2560000 bit format and it would still be able to clip just as easily as a 8 bit wave file, as long as people are trying to make the music as loud as possible.

A CD doesn't store "loudness". The loudness is controlled by the person with the volume know. Given an equal settings, dynamics compression + maximized peak values will give a louder sound. Apparently a lot of people in mastering think this is desirable, so they do just that: maximize the peaks to the maximum the recording format can store, be it 15, 24 or 10000 bits. And very often, they go beyond that (clipping). Adding more bits (more SNR) is helpless against that stupidity. The formats have assloads of headroom but NOBODY USES IT!
Title: Why 24bit/48kHz/96kHz/
Post by: Garf on 2005-12-30 15:00:54
Quote
well, and what about the HDCD (20bit) technology? what are the other 4bits used for?
[a href="index.php?act=findpost&pid=353335"][{POST_SNAPBACK}][/a]


HDCD is CD + dither + noiseshaping + a control signal. And it's only 16 bit, which is why it is compatible with CD's.

The dither + noiseshaping will give an effective increase in resolution (marketing claims: 20bit effective). The control signal will prevent clipping in some circumstances (it's a workaround for braindead mastering engineers).

You can do the dithering + noiseshaping on any CD (not just a HDCD) and it will work effectively. But there are few CD's that are properly made like that.

Use the search function if you want to know more.
Title: Why 24bit/48kHz/96kHz/
Post by: Wintershade on 2005-12-30 15:04:51
Sorry for popping in like that, but here's my answer to the main question here...

I do a lot of audio editing (some people call me a "producer"). For my needs, 16/44.1 kHz is far from sufficient - after applying some filters and effects, artifacts are very audible and audio is heavily damaged. This is compensated with higher sampling rates and bit depths - for now I'm content with 32 bit/88.2 kHz audio.

However, this tosses me out from the "end users" basket. Yes, objectively, for the end users 16/44.1 audio is totally sufficient.

Why end users are being presented with "higher-quality" audio now is beyond me, since most of them can't even hear the difference between 192 kbps mp3 and a "common" audio CD.

Cheerz =]
Title: Why 24bit/48kHz/96kHz/
Post by: bug80 on 2005-12-30 15:25:50
Quote
Sorry for popping in like that, but here's my answer to the main question here...

I do a lot of audio editing (some people call me a "producer"). For my needs, 16/44.1 kHz is far from sufficient - after applying some filters and effects, artifacts are very audible and audio is heavily damaged. This is compensated with higher sampling rates and bit depths - for now I'm content with 32 bit/88.2 kHz audio.[Cheerz =]
[a href="index.php?act=findpost&pid=353341"][{POST_SNAPBACK}][/a]

I understand the part about 32 bits per sample (although 24 bit seems sufficient in most cases), but why 88.2 kHz? Do you have many subsequent A/D converters in your recording chain, like I explained above, because I don't think it is of much use if you do all your editing in the digital domain and resample to 44.1 kHz afterwards.
Title: Why 24bit/48kHz/96kHz/
Post by: Crissaegrim on 2005-12-30 16:02:09
Definitive answer... Artists record at as high as 192khz.  It has to be down-sampled to 44.1khz to fit on a media CD.  It just gives you (you = audiophiles) a fuzzy feeling knowing that it's "really close" to the original recording. 
Title: Why 24bit/48kHz/96kHz/
Post by: bug80 on 2005-12-30 16:09:41
Quote
Definitive answer... Artists record at as high as 192khz.
[a href="index.php?act=findpost&pid=353357"][{POST_SNAPBACK}][/a]

Most studios know that recording at > 44.1 kHz is pretty useless, unless it is a recording for DVD-A or SACD purposes. All studios I know record simply at 44.1 kHz / 24 bit. There are no artists that I know of that record at 192 kHz to give the end user a fuzzy feeling
Title: Why 24bit/48kHz/96kHz/
Post by: Axon on 2005-12-30 18:39:57
Some DSP algorithms may benefit from a high sampling rate, but strictly speaking, they can all be written to work fine at 44.1, or else the algorithm itself will just include its own resampler to do what it needs to do. (A trivially naive example: suppose you want to make a comb filter in less than 10 taps with a starting freq of 12khz. AFAIK, you can't do that with a 44.1 sampling rate, or else you could do it more easily with 96 than you could with 44.1.)

Low integral resamples (88.2 -> 44.1) are quite a bit easier on the math compared to 96->44.1. It shouldn't make an audible difference but the filters need to be much much more complex for the 96 downsample to be legitimate.

The 44.1khz frequency was chosen because the first PCM machines used (drumroll) analog videotape as its media. The math worked out that at 16 bits of resolution you could cram enough samples into a video frame to get 44.1.

The really ironic thing about 192khz is that the SNR numbers are objectively worse compared to 96khz recording (to say nothing of 44.1). Virtually all ADCs are sigma-delta nowadays anyway, and one of their fundamental characteristics is that you can always trade off frequency for resolution. There are a couple hardware vendors who actively speak out against 192khz recording for this reason.

In fact I'd almost argue that 192khz was just as much driven by hardware vendors than audiophiles. They're already trying to cram DxD down studios' necks. 8 bits at 384khz... yuck.
Title: Why 24bit/48kHz/96kHz/
Post by: Lyx on 2005-12-30 18:58:34
Regarding clipping and samplerate: according to a paper i read recently, a higher samplerate can indeed lower the amount of *analogue clipping*. But this can just as well be achieved on 44khz if its recorded with proper level meters (thus, not overcompressed AND then normalized to the maximum). So, the higher samplerate simply acts as a safeguard against malicious mastering practices. Notice that this will only lower the amount of clipping - dynamics will still suffer from overcompression.

Explanation is as following: even if the digital representation of a signal does not clip, it may still clip when its converted back to analogue. As you know, a digital representation of a waveform can simply be visualized as evenly spaced dots. So, only "snapshots" of the signal are stored, not lines or curves. When this representation goes through the DAC it is reconstructed to the real analogue wave - this however does not simply happen by connecting the "dots" with straight lines - instead it is a spline curve. The implication of this is that even if your signal peaks at 100% in the digital representation, it may still go "over the top" when its reconstructed - because the peak values of the digital snapshots are not equal to peak values of the resulting *analogue* signal. A higher samplingrate can lower this effect - but if proper level meters would be used during mastering, then this wouldn't be necessary at all...... of course, that would mean lowering the loudness a bit.... so.....
Title: Why 24bit/48kHz/96kHz/
Post by: Raptus on 2005-12-30 19:07:25
>16bit would be necessary if someone wanted to make a recording with dynamics ranging from treshold of hearing to treshold of pain...

While Nyquist states that every signal component below the Nyquist frequency can be preserved realworld filter constraints make the usable bandwidth smaller, CD specs garantee only up to 20Khz. But that is enough. Using higher samplerates though makes implementing the digital reconstruction filter of the DACs less critical, like Axon already mentioned.
Title: Why 24bit/48kHz/96kHz/
Post by: NoXFeR on 2005-12-30 23:05:49
On a sidenote; recorders may want higher frequencies if they want the frequencies generated above 22khz (analog) in their production. For instance, I have had interesting results playing back 96khz recordings of cymbals at 44khz. I believe quite some artists use this technique, especially for electronic music.
Title: Why 24bit/48kHz/96kHz/
Post by: bug80 on 2005-12-31 11:16:50
Quote
On a sidenote; recorders may want higher frequencies if they want the frequencies generated above 22khz (analog) in their production. For instance, I have had interesting results playing back 96khz recordings of cymbals at 44khz. I believe quite some artists use this technique, especially for electronic music.
[a href="index.php?act=findpost&pid=353442"][{POST_SNAPBACK}][/a]

What do you mean? If you playback anything on 44.1 kHz, you will have no energy content above 22.05 kHz, in fact, due to anti-aliasing filtering the roll-off will even start at a lower frequency.
Title: Why 24bit/48kHz/96kHz/
Post by: bug80 on 2005-12-31 11:23:37
Quote
Regarding clipping and samplerate: according to a paper i read recently, a higher samplerate can indeed lower the amount of *analogue clipping*. But this can just as well be achieved on 44khz if its recorded with proper level meters (thus, not overcompressed AND then normalized to the maximum). So, the higher samplerate simply acts as a safeguard against malicious mastering practices. Notice that this will only lower the amount of clipping - dynamics will still suffer from overcompression.

Explanation is as following: [...]
[a href="index.php?act=findpost&pid=353405"][{POST_SNAPBACK}][/a]

I've read this also. However, I think that this can only be avoided by using a sampling rate of infinity. Doubling the sampling frequency will reduce the number of clipping samples by a factor of two, or so. But it will still happen.

Second, when a producer records at 96 kHz and sees no clipping on his meters, the clipping effect you explained may occur as soon as he/she resamples back to 44.1 kHz for CD. So, it may be better to record at 44.1 kHz from the beginning in this case.
Title: Why 24bit/48kHz/96kHz/
Post by: Lyx on 2005-12-31 11:33:53
One more reason to instead simply use proper level meters, and use compressors in a sane way. Actually, just lowering the amount of compression alone would probably make the problem insignificant in practice.
Title: Why 24bit/48kHz/96kHz/
Post by: bug80 on 2005-12-31 12:33:27
Quote
One more reason to instead simply use proper level meters, and use compressors in a sane way. Actually, just lowering the amount of compression alone would probably make the problem insignificant in practice.
[a href="index.php?act=findpost&pid=353549"][{POST_SNAPBACK}][/a]

Indeed.

By the way, I thought a little bit longer about this. My question now is: is this clipping effect really a problem in practice? Let's assume the output of a D/A converter is in de range -/+ 1 V. Due to the clipping the output becomes -/+ 1.05 V, for instance. I think no analogue equipment will have a problem with that. Since it is all analogue, the sound will not clip, as it would in the digital domain.
Title: Why 24bit/48kHz/96kHz/
Post by: Garf on 2005-12-31 13:20:14
Quote
Quote
One more reason to instead simply use proper level meters, and use compressors in a sane way. Actually, just lowering the amount of compression alone would probably make the problem insignificant in practice.
[a href="index.php?act=findpost&pid=353549"][{POST_SNAPBACK}][/a]

Indeed.

By the way, I thought a little bit longer about this. My question now is: is this clipping effect really a problem in practice? Let's assume the output of a D/A converter is in de range -/+ 1 V. Due to the clipping the output becomes -/+ 1.05 V, for instance..
[a href="index.php?act=findpost&pid=353560"][{POST_SNAPBACK}][/a]


Huu, no. It will be +-1V and not more. The clipping will add distortion because the wave shape no longer corresponds to the original (has additional HF distortion).
Title: Why 24bit/48kHz/96kHz/
Post by: bug80 on 2005-12-31 13:23:16
Quote
Quote
Quote
One more reason to instead simply use proper level meters, and use compressors in a sane way. Actually, just lowering the amount of compression alone would probably make the problem insignificant in practice.
[a href="index.php?act=findpost&pid=353549"][{POST_SNAPBACK}][/a]

Indeed.

By the way, I thought a little bit longer about this. My question now is: is this clipping effect really a problem in practice? Let's assume the output of a D/A converter is in de range -/+ 1 V. Due to the clipping the output becomes -/+ 1.05 V, for instance..
[a href="index.php?act=findpost&pid=353560"][{POST_SNAPBACK}][/a]


Huu, no. It will be +-1V and not more.

Why, because the analogue output of a D/A converter is limited? Why should it be? You may be right, I'm no expert on D/A converters.

edit: We're not talking about conventional clipping here, but clipping due to the D/A conversion, so that's where my question comes from.
Title: Why 24bit/48kHz/96kHz/
Post by: Garf on 2005-12-31 13:26:41
Quote
As you know, a digital representation of a waveform can simply be visualized as evenly spaced dots. So, only "snapshots" of the signal are stored, not lines or curves. When this representation goes through the DAC it is reconstructed to the real analogue wave - this however does not simply happen by connecting the "dots" with straight lines - instead it is a spline curve. The implication of this is that even if your signal peaks at 100% in the digital representation, it may still go "over the top" when its reconstructed - because the peak values of the digital snapshots are not equal to peak values of the resulting *analogue* signal.
[a href="index.php?act=findpost&pid=353405"][{POST_SNAPBACK}][/a]


I believe this is wrong, or actually, not really explained as corresponding to reality. The DAC will follow the PCM representation as closely as possible, and thus add all kinds of HF distortion. Following up on the DAC should be an (as perfect as possible) lowpass filter, which will remove them again, and leave a "perfect" signal. This follows directly from DSP sampling theory.

The result of this process is more or less as you explain (though it isn't a spline curve, not in the mathemathical sense).

The problem of constructing the ideal as possible lowpass filter is one argument that is used for higher sampling rates (if you up the sampling rate, you can make a crappier filter and still get the same quality audio). It's a rather crappy argument though, because it's not as if we can't cheaply make good performing ones now!
Title: Why 24bit/48kHz/96kHz/
Post by: kwwong on 2005-12-31 13:30:06
The main reason why sampling rates > 44.1Khz is used is that at the hardware / player, the required transition bandwidth of the analogue low pass anti-aliasing filter is much - much wider and thus a simple low order analogue filter is sufficient.  Building a low-order linear phase low pass filter is very simple compare to a high order filter!

It has nothing to do with psychoacoustics but electronics. If the sampling rate is 192 kHz, do not expect to hear above the usual 18 - 20 kHz range.

As for the case of 24 bits, I supposed 16 bits audio isn't enough to represent the entire dynamic range of the human hearing. At its most sensitive region, around 2 - 4 kHz, this dynamic range is actually greater than 100 dB. 
Title: Why 24bit/48kHz/96kHz/
Post by: Garf on 2005-12-31 13:31:06
Quote
Quote
Quote
Quote
One more reason to instead simply use proper level meters, and use compressors in a sane way. Actually, just lowering the amount of compression alone would probably make the problem insignificant in practice.
[a href="index.php?act=findpost&pid=353549"][{POST_SNAPBACK}][/a]

Indeed.

By the way, I thought a little bit longer about this. My question now is: is this clipping effect really a problem in practice? Let's assume the output of a D/A converter is in de range -/+ 1 V. Due to the clipping the output becomes -/+ 1.05 V, for instance..
[a href="index.php?act=findpost&pid=353560"][{POST_SNAPBACK}][/a]


Huu, no. It will be +-1V and not more.

Why, because the analogue output of a D/A converter is limited? Why should it be? You may be right, I'm no expert on D/A converters.

edit: We're not talking about conventional clipping here, but clipping due to the D/A conversion, so that's where my question comes from.
[a href="index.php?act=findpost&pid=353573"][{POST_SNAPBACK}][/a]



Yes, of course it's limited. In practise it'll be limited to the supply voltage, i.e. 3.3V or something thereabouts.

Edit: Or a bit less depending on design etc. But in any case, the output voltage for sure is hard limited, it won't magically swing outside it's maximum range.
Title: Why 24bit/48kHz/96kHz/
Post by: Garf on 2005-12-31 13:41:31
Quote
The main reason why sampling rates > 44.1Khz is used is that at the hardware / player, the required transition bandwidth of the analogue low pass anti-aliasing filter is much - much wider and thus a simple low order analogue filter is sufficient.  Building a low-order linear phase low pass filter is very simple compare to a high order filter!


I know this argument, but does it have any bearing on practise? I mean, is this still a practical problem anywhere? It certainly doesn't seem to be! Even if we can't get the theorethical resolution because of limitations of real world filters, we certainly seem to be able to do well enough very cheaply that the results are perfect in a psychoacoustical sense.

Quote
As for the case of 24 bits, I supposed 16 bits audio isn't enough to represent the entire dynamic range of the human hearing. At its most sensitive region, around 2 - 4 kHz, this dynamic range is actually greater than 100 dB. 
[a href="index.php?act=findpost&pid=353575"][{POST_SNAPBACK}][/a]


Yes, and the amount of records that actually need such a dynamic range is? One in ten million?

As I already explained, I do not think that we need a dynamic range so far that the lower end of audibility corresponds with a higher end where the hearing of the listener is damaged by anything but very short term exposure. That's not only useless, it's actually dangerous.
Title: Why 24bit/48kHz/96kHz/
Post by: Garf on 2005-12-31 13:51:07
Another thing to consider is that with proper dithering and noiseshaping, the effective resolution at 44.1kHz can be boosted by as much as 15dB. This means that a properly produced CD (unfortunately very few exist, but HDCD are an example) will have an effective SNR of 111dB.

How many ADC or DAC's have you seen that can attain that resolution?
Title: Why 24bit/48kHz/96kHz/
Post by: NoXFeR on 2005-12-31 14:18:27
Quote
Quote
On a sidenote; recorders may want higher frequencies if they want the frequencies generated above 22khz (analog) in their production. For instance, I have had interesting results playing back 96khz recordings of cymbals at 44khz. I believe quite some artists use this technique, especially for electronic music.
[a href="index.php?act=findpost&pid=353442"][{POST_SNAPBACK}][/a]

What do you mean? If you playback anything on 44.1 kHz, you will have no energy content above 22.05 kHz, in fact, due to anti-aliasing filtering the roll-off will even start at a lower frequency.
[a href="index.php?act=findpost&pid=353546"][{POST_SNAPBACK}][/a]


True, but that's not the point.

If I record some instrument at 192khz, I may want the output played back at 44khz more for the piece I am making than for the usual sound of the instrument. The point is that if you play an instrument, it produces interesting sounds above the frequencies we hear that we cannot hear if we don't play back the music at lower rates.
Title: Why 24bit/48kHz/96kHz/
Post by: WmAx on 2005-12-31 14:29:34
Quote
The point is that if you play an instrument, it produces interesting sounds above the frequencies we hear that we cannot hear if we don't play back the music at lower rates.
[a href="index.php?act=findpost&pid=353588"][{POST_SNAPBACK}][/a]


That statement does not make sense.

-Chris
Title: Why 24bit/48kHz/96kHz/
Post by: Garf on 2005-12-31 14:31:53
I think what he's trying to say is this:

If you record up to 192kHz, and then shift the upper frequency range to a range that is within our human hearing range, you can make nice effects.
Title: Why 24bit/48kHz/96kHz/
Post by: Lyx on 2005-12-31 14:31:54
I think he means playing 192khz sample at 44khz WITHOUT converting it to 44khz beforehand...... so that in effect it will be played at much slower speed and pitched down.
Title: Why 24bit/48kHz/96kHz/
Post by: WmAx on 2005-12-31 14:39:35
Quote
As for the case of 24 bits, I supposed 16 bits audio isn't enough to represent the entire dynamic range of the human hearing. At its most sensitive region, around 2 - 4 kHz, this dynamic range is actually greater than 100 dB. 
[a href="index.php?act=findpost&pid=353575"][{POST_SNAPBACK}][/a]


Aside from the safety concerns that Garf has pointed out, it should be noted that when you consider the noise floor of even a very quiet(but not typical) environment(35dB range), plus the SPL capabilities of the transducers, it does not make technical sense to have greater than CD's dynamic range abilities. For starters, one would need large horn speakers or massive line arrays as a prerequisite. 2nd, live(unamplified) music simply does not use the dynamic range of CD, even if uncompressed, from an audience perspective. Now, obviously if you are sitting in the band(or in a small practice room with a band playing within it), you will be experiencing very high SPLs(and you had better be wearing hearing protection) that will strain, and possible exceed in some cases, the RBCD dynamic range capability. You will also be exceeding the safe limits of human hearing.

-Chris
Title: Why 24bit/48kHz/96kHz/
Post by: jmvalin on 2005-12-31 15:22:18
Quote
Practise disagrees with your suppositions. Is there any proof that we can hear beyond 96dB SNR?

Practical experiments have shown most listeners already get into problems hearing to the 13th bit (78db SNR). If you think you (or another person) can do better, please let them take the MAD challenge with whatever equipment and ears you want, come back aftwards, and see if you would still tell me there is a practical reason to have more.

For the record, claiming you hear over 100dB SNR equals claiming you can hear the person next to you breathing while listening to a live rock concert. Still sounds reasonable?
[a href="index.php?act=findpost&pid=353304"][{POST_SNAPBACK}][/a]


Actually, I can see only one reason why 24 bits *may* be worth it. While I really doubt you can hear noise with a 96 dB SNR, the extra bit can give you extra dynamic range. You would normally want to set the least significant bit to "match" the absolute threshold of hearing. That means that with 16-bit resolution, the loudest sound you can reproduce is 96 dB louder. That means you couldn't reproduce (in the same track) the sound of someone breathing 5 meters away *followed* by the sound of a jet taking off. Why anyone would want to do that is beyond me, but I'm wondering if 16 bits is actually enough for movie theatres.
Title: Why 24bit/48kHz/96kHz/
Post by: Garf on 2005-12-31 15:31:02
Quote
Actually, I can see only one reason why 24 bits *may* be worth it. While I really doubt you can hear noise with a 96 dB SNR, the extra bit can give you extra dynamic range. You would normally want to set the least significant bit to "match" the absolute threshold of hearing. That means that with 16-bit resolution, the loudest sound you can reproduce is 96 dB louder. That means you couldn't reproduce (in the same track) the sound of someone breathing 5 meters away *followed* by the sound of a jet taking off. Why anyone would want to do that is beyond me, but I'm wondering if 16 bits is actually enough for movie theatres.
[a href="index.php?act=findpost&pid=353600"][{POST_SNAPBACK}][/a]


1) Exactly the same was said a few posts before.

2) Movie theaters don't use CD's.
Title: Why 24bit/48kHz/96kHz/
Post by: WmAx on 2005-12-31 15:33:01
Quote
but I'm wondering if 16 bits is actually enough for movie theatres.
[a href="index.php?act=findpost&pid=353600"][{POST_SNAPBACK}][/a]


Just to point out the obvious: What's the noisefloor in a theatre full of people when they are being as quiet as possible? 55-60dB at best?

-Chris
Title: Why 24bit/48kHz/96kHz/
Post by: ivalladt on 2005-12-31 18:13:48
Quote
The question is why one would need a SNR of >96dB or a bandwidth of over 22kHz for end user playback.
[a href="index.php?act=findpost&pid=353050"][{POST_SNAPBACK}][/a]


Not needed for playback, of course, but for editing both images and sound, the higher the resolution, the more precise the result.

Quote
I have to step out and disagree that there is no audible difference between 16-bit and 24-bit recordings. The decision to use 16/44.1 was arrived at, not because it was audio nirvana, but because it was the best comprimise possible with the digital to analogue conversion technology of the 80's.
[a href="index.php?act=findpost&pid=353234"][{POST_SNAPBACK}][/a]


Also to make the lenght of most musical works fit into a CD.
Title: Why 24bit/48kHz/96kHz/
Post by: Shade[ST] on 2005-12-31 18:25:10
Quote
Just to point out the obvious: What's the noisefloor in a theatre full of people when they are being as quiet as possible? 55-60dB at best?

For theatre as in not movies, I've seen 40-45 dB;  of course, I didn't have a sonometer, but I could hear the actess whispering, and I was over 20 meters away.
Title: Why 24bit/48kHz/96kHz/
Post by: WmAx on 2005-12-31 19:52:07
Quote
,Dec 31 2005, 02:25 PM]
Quote
Just to point out the obvious: What's the noisefloor in a theatre full of people when they are being as quiet as possible? 55-60dB at best?

For theatre as in not movies, I've seen 40-45 dB;  of course, I didn't have a sonometer, but I could hear the actess whispering, and I was over 20 meters away.
[a href="index.php?act=findpost&pid=353630"][{POST_SNAPBACK}][/a]


I don't see how you can know it was 40-45dB without a SPL meter, but perhaps you are correct. I have never experienced a floor that I would rate that low in any audience. 40dB would be good for an average home environment, with no one making a sound. But let's say it's 40dB. What is the masking depth of the format floor(which is basicly broadband white noise) under this environment noise floor(which usually resembles braodband noise)? Assuming -10 or -15 dB is sufficient, you have 120 dB peak ability with that lowly 16 bit format. 120dB is extremely loud, and beyond the capability of most home playback systems(exepting large horn speakers and speaker arrays). In the only musical playback perceptual [1]test of SNR of which I'm aware, only -74dB was required for transparent playback on speakers, with a little bit more neded for headphones. But you(or someone here) brought up special effects like a plane flying directly overhead. I don't think that anyone would actually want to reproduce this, even for movie use.

-Chris

[1]Signal-to-Noise Ratio Requirement for Digital Transmission Systems
Spikofski, Gerhard
AES Preprint: 2196
Title: Why 24bit/48kHz/96kHz/
Post by: kwwong on 2006-01-01 03:30:54
Quote
Quote
The main reason why sampling rates > 44.1Khz is used is that at the hardware / player, the required transition bandwidth of the analogue low pass anti-aliasing filter is much - much wider and thus a simple low order analogue filter is sufficient.  Building a low-order linear phase low pass filter is very simple compare to a high order filter!


I know this argument, but does it have any bearing on practise? I mean, is this still a practical problem anywhere? It certainly doesn't seem to be! Even if we can't get the theorethical resolution because of limitations of real world filters, we certainly seem to be able to do well enough very cheaply that the results are perfect in a psychoacoustical sense.

Quote
As for the case of 24 bits, I supposed 16 bits audio isn't enough to represent the entire dynamic range of the human hearing. At its most sensitive region, around 2 - 4 kHz, this dynamic range is actually greater than 100 dB. 
[a href="index.php?act=findpost&pid=353575"][{POST_SNAPBACK}][/a]


Yes, and the amount of records that actually need such a dynamic range is? One in ten million?

As I already explained, I do not think that we need a dynamic range so far that the lower end of audibility corresponds with a higher end where the hearing of the listener is damaged by anything but very short term exposure. That's not only useless, it's actually dangerous.
[a href="index.php?act=findpost&pid=353578"][{POST_SNAPBACK}][/a]


Well I have to admit that I am not really an expert in analogue filter design but from what I have read "somewhere", it is very difficult to design a high order filter with linear phase characteristics besides requiring very expensive high precision resistors and capacitors and inductors !

I supposed this must have been the main drive to higher sampling rate systems as the audio hardware industry is a "cut-throat" business. 
Title: Why 24bit/48kHz/96kHz/
Post by: kwwong on 2006-01-01 03:39:39
Quote
Quote
Quote
The main reason why sampling rates > 44.1Khz is used is that at the hardware / player, the required transition bandwidth of the analogue low pass anti-aliasing filter is much - much wider and thus a simple low order analogue filter is sufficient.  Building a low-order linear phase low pass filter is very simple compare to a high order filter!


I know this argument, but does it have any bearing on practise? I mean, is this still a practical problem anywhere? It certainly doesn't seem to be! Even if we can't get the theorethical resolution because of limitations of real world filters, we certainly seem to be able to do well enough very cheaply that the results are perfect in a psychoacoustical sense.

Quote
As for the case of 24 bits, I supposed 16 bits audio isn't enough to represent the entire dynamic range of the human hearing. At its most sensitive region, around 2 - 4 kHz, this dynamic range is actually greater than 100 dB. 
[a href="index.php?act=findpost&pid=353575"][{POST_SNAPBACK}][/a]


Yes, and the amount of records that actually need such a dynamic range is? One in ten million?

As I already explained, I do not think that we need a dynamic range so far that the lower end of audibility corresponds with a higher end where the hearing of the listener is damaged by anything but very short term exposure. That's not only useless, it's actually dangerous.
[a href="index.php?act=findpost&pid=353578"][{POST_SNAPBACK}][/a]


Well I have to admit that I am not really an expert in analogue filter design but from what I have read "somewhere", it is very difficult to design a high order filter with linear phase characteristics besides requiring very expensive high precision resistors and capacitors and inductors !

I supposed this must have been the main drive to higher sampling rate systems as the audio hardware industry is a "cut-throat" business. 
[a href="index.php?act=findpost&pid=353693"][{POST_SNAPBACK}][/a]


Sorry, perhaps higher sampling rate allows the usage of better DA converters instead of the usual sigma-delta oversampling DA converters?
Title: Why 24bit/48kHz/96kHz/
Post by: Wintershade on 2006-01-02 11:45:15
Quote
Do you have many subsequent A/D converters in your recording chain


In most cases, yes.

When editing "purely" digital audio, I agree there is not much sense in upsampling and downsampling (this is not true for changing the bit depth), but when recording from analogue source, my ears told me many times that it is better to have higher sampling rates when editing.

Cheerz =]
Title: Why 24bit/48kHz/96kHz/
Post by: dekkersj on 2006-01-02 14:41:13
Quote
Another thing to consider is that with proper dithering and noiseshaping, the effective resolution at 44.1kHz can be boosted by as much as 15dB. This means that a properly produced CD (unfortunately very few exist, but HDCD are an example) will have an effective SNR of 111dB.

How many ADC or DAC's have you seen that can attain that resolution?
[a href="index.php?act=findpost&pid=353582"][{POST_SNAPBACK}][/a]



Hi Garf,

How does the calculation exactly go? Is there less noise involved, or is the effective resolution associated with a hearing curve?

Regards,
Jacco
Title: Why 24bit/48kHz/96kHz/
Post by: probedb on 2006-01-05 13:28:27
Quote
1) The 44.1kHz has its historical reasons, but how about 48kHz, 96kHz, or even 192kHz?
How are these numbers chosen? Any technical and practical advantage over 44.1kHz? And why creative cards resamples to these sampling rates?


Just thought I'd pipe up and answer this one. I think all sound cards that adhere to the AC'97 spec have to have the ability to resample to 48KHz, some vendors do this in hardware while others just have it as an option.

It's a common complaint on HTPC forums when people can't get bit-perfect output from their PCs into external DACs
Title: Why 24bit/48kHz/96kHz/
Post by: marcan on 2006-01-05 14:57:04
Quote
Another thing to consider is that with proper dithering and noiseshaping, the effective resolution at 44.1kHz can be boosted by as much as 15dB. This means that a properly produced CD (unfortunately very few exist, but HDCD are an example) will have an effective SNR of 111dB.

How many ADC or DAC's have you seen that can attain that resolution?
[a href="index.php?act=findpost&pid=353582"][{POST_SNAPBACK}][/a]

Hi Garf,
It's quite perturbing to talk about 111dB of SNR with a DAC having 96dB (16bits).
Actually dithering/NS decrease distortion while increasing the noise. This noise being created in order to be barely audible, we can say that we have a psychoacoustic SNR of 111 dB on a device with a dynamic of 96 dB.
Is that correct?
Title: Why 24bit/48kHz/96kHz/
Post by: enry2k on 2006-01-06 00:08:39
First: What about ensuring extremely controlled listening conditions, I mean using closed headphones, maybe using noise cancellation, in additon headphones requere much less power than speakers, so clipping of extremely loud peals should be prevented unlikely with the hadling capacity of even the most powerful amplifiers. Could these conditions lead to a significant reduction of the external noise floor in a way that extremely low level sounds could be audable and extremely loud peaks could result undistorted, so that 24 bit could make a difference?

Second: I know phase shift in an audio signal is not audable provided that it is constant in the entire audio spectrum. I was wondering whether filtering introduces frequency depending phase shift so it could become therefore audable?

Third: What do you think of this quote from the article titled "SACD vs. DVD-Audio:
High Definition Formats Evaluation "
http://www.digit-life.com/articles2/sacd-dvd-a/ (http://www.digit-life.com/articles2/sacd-dvd-a/)
  Do you think It is a pure nonsense?

Why do We Need Frequencies over 20 kHz?

We have come to an important stage. One and the same question has been asked on different forums in threads about formats and their quality - if a human being can't hear frequencies over 20 kHz (in rare instances 21 kHz) then the 22  kHz limitation in CD-DA (according to the Nyquist theorem it's a half of sampling frequency, for 44100 it will be 22050Hz) should pose no problems, shouldn't it? Why do we need 50 kHz, 70 kHz, and even 96 kHz, if we don't hear them anyway? And the second point - if our speakers have the stated pass band of 20Hz-20 kHz on the level of -3dB, then why do we need DVD-A or SACD? Our loudspeakers will not be able to reproduce the required spectrum, and moreover, intermodulation products of ultrasonic frequencies may get into the audible region and distort the original signal. Isn't it better don't have these ultrasonic images?

That's what said Nelson Pass, the well-known guru in the audiophile and audio engineerieng society:

"Although human hearing is generally very poor above 20,000 Hertz, ultrasonic frequency roll-offs produce phase and amplitude effects in the audible region; for example, a single pole (6dB/octave) roll-off at 30  kHz produces about 9 phase lag and 0.5 dB loss at 10  kHz. The effects may be subtle, but their audibility is undesirable in a piece of equipment whose performance is judged by its neutrality." (original). Thus, being aware of our hearing sensitivity to phase distortions, we can presume considerable decrease in the level of such distortions in the systems with a wider signal spectrum (including the quality LP playback).

Acoustical Spectrum of Real Instruments

The second point - if we limit the original signal with the analog anti-aliasing filter for the CD signal spectrum in comparison with the same filter for DVD-A 24bit 96 kHz signal when trying to record the spectrum of such instruments as a trumpet, there will be a considerable difference in phase distortions between the original and the record even before it is delivered to the speakers. The speakers (well-designed) have mildly sloping signal falloff characteristic at high frequencies, which mostly depends on the tweeter design, and thus introduces fewer phase distortions than the filter with a steep amplitude-frequency response falloff, which is in fact the CD-DA format for the wideband spectrum signals.

Thank you for your attention

Enrico
Title: Why 24bit/48kHz/96kHz/
Post by: Pio2001 on 2006-01-06 05:48:21
Quote
And why creative cards resamples to these sampling rates?[{POST_SNAPBACK}][/a] (http://index.php?act=findpost&pid=353238")


They need to be able to mix together different sounds. You would not like your music to stop when you receive a mail or when you launch a game.
Analog mixers would be much more expensive than a soundcard, plus the cost of one DAC per channel.
So the mix is done digitally. But what will be the output sample rate if one input track is 22050 Hz, another 32 kHz, and another one 44100 Hz ? Well, 48 kHz was chosen.

Quote
It's quite perturbing to talk about 111dB of SNR with a DAC having 96dB (16bits).
Actually dithering/NS decrease distortion while increasing the noise. This noise being created in order to be barely audible, we can say that we have a psychoacoustic SNR of 111 dB on a device with a dynamic of 96 dB.
Is that correct?
[a href="index.php?act=findpost&pid=354797"][{POST_SNAPBACK}][/a]


This is just a matter of frequency. It seems to me that in a 16 bits system, you can't get more than 96 dB SNR at 22049 Hz. But you may get more at lower frequencies.

Quote
I was wondering whether filtering introduces frequency depending phase shift so it could become therefore audable?


According to Nika Aldrich (  [a href="http://www.cadenzarecording.com/papers.html]The Audio Engineer's Approach to Understanding Digital Filters[/url], p35 ), filters used in ADCs and DACs are almost exclusively FIR filters, that do not introduce any frequency dependant phase shift, while getting -144 dB at 22050 Hz and 0 dB at 20000 Hz.
This dismisses all the arguments of Nelson Pass quoted above.
Title: Why 24bit/48kHz/96kHz/
Post by: krabapple on 2006-01-06 19:04:20
Quote
Quote
Why do we need 24bit/48kHz/96kHz/192kHz if 16bit/44.1kHz is good enough?
[{POST_SNAPBACK}][/a] (http://index.php?act=findpost&pid=353038")

We dont need it. It's just virtual useless number-games to give people the incentive to buy new equipment and then re-buy all our music. There are some *technical* arguments for using 48khz instead of 44khz.... but the actual benefit for normal endusers is zero.
[a href="index.php?act=findpost&pid=353041"][{POST_SNAPBACK}][/a]



The most thorough advocacy case I've ever seen for greater-than-redbook schemes was laid down by J. Robert Stuart of Meridian (who brought us Meridian Lossless Packing for DVD-Audio)
But even *it* boils down to a case of 'we don't *know* for sure that we really need it, but why not use it since we can?'  I just don't see the point of it for home playback, myself.

[a href="http://www.meridian-audio.com/ara/coding2.pdf]http://www.meridian-audio.com/ara/coding2.pdf[/url]
Title: Why 24bit/48kHz/96kHz/
Post by: krabapple on 2006-01-06 19:10:51
Quote
Quote
I agree that it makes no difference to end users...  though there are technical reasons why this would have its advantages...

most of it comes from the way hardware plays back music.... it has to do with the enforcements of peaks... each sample is essentially enforced as a peak in the reproduced sound wave... but this should NOT be the case... the sample may have been taken along any part of the wave (and most likely not at a peak)...

The affects of this are mostly negligable and hardware also works to minimize the affects... but it shouldn't have to... and higher sampling rates allow the hardware to start  with a closer reresentation to the original wave...

I'll look for some articles about this "phenom"... I actually looked them up a year or two ago cause an audiophile engineer I know was telling me about it and I didn't believe him...
[{POST_SNAPBACK}][/a] (http://index.php?act=findpost&pid=353155")


This really doesn't make any sense at all.
[a href="index.php?act=findpost&pid=353156"][{POST_SNAPBACK}][/a]


He's talking about 'phantom overs' -- signals that don't register as 'overs' during recording/production/mastering  due to poor digital metering.  Some hardware handles these intelligently, others just clip.

Here's two articles about them:

[a href="http://www.tcelectronic.com/media/Level_paper_AES109.pdf]http://www.tcelectronic.com/media/Level_paper_AES109.pdf[/url]

http://www.cadenzarecording.com/papers/Digitaldistortion.pdf (http://www.cadenzarecording.com/papers/Digitaldistortion.pdf)
Title: Why 24bit/48kHz/96kHz/
Post by: krabapple on 2006-01-06 19:16:29
Quote
Sorry for popping in like that, but here's my answer to the main question here...

I do a lot of audio editing (some people call me a "producer"). For my needs, 16/44.1 kHz is far from sufficient - after applying some filters and effects, artifacts are very audible and audio is heavily damaged.


Then I suggest you make sure all your work stays in the 32-bit domain until you're done, then convert it down to 16.  A higher sampling rate than redbook standard is probably unnecessary unless you are employing converters that are introducing audible aliasing artifacts with 44 kHz.
Title: Why 24bit/48kHz/96kHz/
Post by: krabapple on 2006-01-06 19:27:40
Quote
First: What about ensuring extremely controlled listening conditions, I mean using closed headphones, maybe using noise cancellation, in additon headphones requere much less power than speakers, so clipping of extremely loud peals should be prevented unlikely with the hadling capacity of even the most powerful amplifiers. Could these conditions lead to a significant reduction of the external noise floor in a way that extremely low level sounds could be audable and extremely loud peaks could result undistorted, so that 24 bit could make a difference?

Second: I know phase shift in an audio signal is not audable provided that it is constant in the entire audio spectrum. I was wondering whether filtering introduces frequency depending phase shift so it could become therefore audable?

Third: What do you think of this quote from the article titled "SACD vs. DVD-Audio:
High Definition Formats Evaluation "
http://www.digit-life.com/articles2/sacd-dvd-a/ (http://www.digit-life.com/articles2/sacd-dvd-a/)
  Do you think It is a pure nonsense?



Maybe you want to read an entire HA thread about it -- where the author also chimed in?

http://www.hydrogenaudio.org/forums/index....showtopic=26218 (http://www.hydrogenaudio.org/forums/index.php?showtopic=26218)
Title: Why 24bit/48kHz/96kHz/
Post by: enry2k on 2006-01-08 19:36:23
Quote from: krabapple,Jan 6 2006, 11:27 AM

Quote from: enry2k,Jan 5 2006, 07:08 PM
First: What about ensuring extremely Maybe you want to read an entire HA thread about it -- where the author also chimed in?

http://www.hydrogenaudio.org/forums/index....showtopic=26218 (http://www.hydrogenaudio.org/forums/index.php?showtopic=26218)
[a href="index.php?act=findpost&pid=355101"][{POST_SNAPBACK}][/a]


Thank you for quoting the thread, it is really interesting.

On the same cd audio quality subject, a few yaars ago Ivan Dimkovic, gave me this interesting reply in the audiocoding forum:
Quote [ ]Author: Ivan Dimkovic (---.verat.net)
Date:  01-05-02 06:32

For "studio" quality, sampling rate of 96 kHz and 24-bit fidelity is used. This is not used because somebody would be able to hear 40 kHz tone, or to distinguish between 110 and 105 dB signal strength, but because of filter design and many steps in production.

For "everyday" use, 64 kHz with 24 bit fidelity - with 25 kHz lowpass filter (to avoid AD-DA problems) would be "best of best" - I doubt that somebody would be able to distinguish this from 96 kHz without lowpass.]

So what are the A/D-D/A problems he is reffering to?
Couldn't be possible to use oversampling in the A/D  converter "say 4 times 44.1 Knz " to avoid sharp antialiasing filters and related problems, then in a following stage, throught the useage of FIR filters  to reduce the bandwidth to 20 Knz and sample rate converting the signal down to 44.1 knz without significant loss with no use of analogue steepy filters, to produce a perceptially valuable signal? Oversampling is also commonly employed in D/A converters.

Finally has AES or any other important organizations published any results yet regarding random tests involving the so-called golden ears on 44.1/16 vs 96/24 subjective comparison?

Regards

Enrico
Title: Why 24bit/48kHz/96kHz/
Post by: enry2k on 2006-01-08 20:28:24
Quote
Quote
Another thing to consider is that with proper dithering and noiseshaping, the effective resolution at 44.1kHz can be boosted by as much as 15dB. This means that a properly produced CD (unfortunately very few exist, but HDCD are an example) will have an effective SNR of 111dB.

How many ADC or DAC's have you seen that can attain that resolution?
[a href="index.php?act=findpost&pid=353582"][{POST_SNAPBACK}][/a]

Hi Garf,
It's quite perturbing to talk about 111dB of SNR with a DAC having 96dB (16bits).
Actually dithering/NS decrease distortion while increasing the noise. This noise being created in order to be barely audible, we can say that we have a psychoacoustic SNR of 111 dB on a device with a dynamic of 96 dB.
Is that correct?
[a href="index.php?act=findpost&pid=354797"][{POST_SNAPBACK}][/a]


I don't think dithering necessary adds a noise floor to the signal, in a book titled "Principles of Digital Audio" I read a few years ago, a A/D converter is decribed in which a pseudorandom signal is used to generate noise throught a D/A converter which produces a signal level comparable to the least significant bit. The noise signal is mixed up with the audio signal to be encoded. After A/D conversion, the same random signal is subtracted from the LSB. The resulting signal is said to retain the dither positive effect, without additional noise floor.
I guess this is the same idea  HD CDA was based upon. producing a result similar to 20 bit quatization.

Enrico
Title: Why 24bit/48kHz/96kHz/
Post by: Qjimbo on 2006-01-08 23:20:38
Ok, back to the original question of the thread. This is what I'd always thought was the reason for needing higher samplerates. Feel free to correct me if I'm wrong about this.

Firstly as humans, we can all hear up to roughly 22050Hz, and as nyquist worked out, to record all this this frequency and all the ones below it, you have to take a number of samples per second which is double that, so we get to 44100Hz aka 44.1Khz

However if you do this, although it might seem fine, the higher frequencies have less deatail the further up you go. A 22050Hz sound will only get an on and off point, but none of the smoothness of the curve, so it will be a triangle wave. Not to mention aliasing, which might make it capture a half on and half off point.

Now I've been told that this isn't correct, but from everything I've read online it seems to be.

There's also the more debatable issue that higher frequencies add to the warmth of the sound, even though they're inaudible on a concious level. Hence the term analog warmth, where valves would add random high frequencies to the audio.

Anyway thats just some of the stuff floating around in my mind I don't want another TOS 8 violation >_<
Title: Why 24bit/48kHz/96kHz/
Post by: CSMR on 2006-01-09 00:52:39
Somehow this question brings out all the know-alls whose world-view evidently depends on CD being a perfect medium. It really mystifies me how a person can become emotionally involved in this issue.
In any case some mathematics comes as a relief.
Quote
Firstly as humans, we can all hear up to roughly 22050Hz, and as nyquist worked out, to record all this this frequency and all the ones below it, you have to take a number of samples per second which is double that, so we get to 44100Hz aka 44.1Khz

However if you do this, although it might seem fine, the higher frequencies have less deatail the further up you go. A 22050Hz sound will only get an on and off point, but none of the smoothness of the curve, so it will be a triangle wave. Not to mention aliasing, which might make it capture a half on and half off point.

You are not right. Nyquist allows the original function to be reconstucted under certain conditions (that it's frequency components found by fourier transformation should be within a bound, say 20khz). To go from the samples to the original wave is slightly complicated I think and you will need to look at the proof of the theorem. It certainly isn't linear interpolation, which would leave you with your triangle wave. To say this again, nyquist says that the function which goes from waves with frequencies under 20khz to samples separated by 1/44khz has an inverse, but don't think that this inverse is simple.

Now Nyquist won't go up to 22050 hz it will go to any limit under 22050hz. Say 20khz. If you have a pure 20khz tone and sample at 22khz (for an infinite length of time) you will be able to deduce that your the original wave was a 20khz tone if you knew it had no components above 44khz.
Title: Why 24bit/48kHz/96kHz/
Post by: AndyH-ha on 2006-01-09 07:28:37
The Sampling Theorem says that signals containing frequencies up to 1/2 the sampling frequency (S/2) can be exactly represented by S samples per second. Restated, the highest frequency that can be recorded, and played back, is equal to 1/2 the sample rate. It isn't approximately, or "in the neighborhood of", or "a degraded representation of", it is exactly the same waveform, right up to S/2  (e.g. 22, 050 Hz at 44,100 Hz). No information is lost.

The input must be bandlimited to no more than S/2 (the Nyquist frequency) or there will be distortion. If higher frequencies exist in the input they will be partially sampled,  creating false information that portrays them as lower frequencies in the audio band. These are called alias images or foldover distortion. If S is the sample rate and F is some frequency between S/2 and S, then its image is created at S - F. Thus as F approaches S, the images are created at lower and lower audio frequencies.

Since it isn't feasible to build analogue filters that truly act as "brick walls," a stop band filter is employed that begins attenuating the input several thousand Hz below the Nyquist frequency. By S/2 the signal level is down sufficiently that aliasing isn't a significant problem. It is real world filter insufficiency, not any sampling theory constraints, that may limit actual performance to something less than the Nyquist frequency.

It is easier to do a better job by sampling at a much higher frequency and handling the details in the digital domain, either in hardware, or especially in software. In actual practice many ADCs, probably those in most of our soundcards, sample at several million times per second, use some form of sigma-delta conversion, then filter down to the desired sample rate and bit depth.

Sampling means that the continuous analogue waveform voltage can be measured only in terms of a limited number of discrete digital values. Although frequency information can be captured exactly (up to S/2), amplitude accuracy is limited by the size of the intervals into which the amplitude (from zero to maximum) must be assigned (e.g .01V, .02V, .03V, ...) The input is always recorded as exactly one of those values although the actual sampled input voltage is usually not exactly one of those values. The error can be as great as 1/2 interval.

Increasing the bit depth increases the number of possible values, increasing accuracy. Going from 16 bits to 24 bits increase the accuracy by a factor of 256, or decreases the average error amplitude by 1/256.

As signal level decreases the errors become larger relative to the signal. Also as signal level decreases, the number of bits, or levels, between it and zero decreases. This provides relatively fewer  choices to recorded its value. Thus the increased number of quantization levels available with greater bit depth becomes more important at lower input levels.

At high signal levels the quantization error is experienced as white noise but at very low signal level it is experienced as distortion. For every bit of depth added, the error, or quantization noise, is decreased by 6dB. Distortion of low level signals (e.g. -90dB to -96dB)  is much greater when using 16 bit rather than 24 bit.

There are  of course many other considerations to getting good digital audio reproduction . Whether or not some of the factors discussed here are always audible, they are real and measurable.
Title: Why 24bit/48kHz/96kHz/
Post by: bug80 on 2006-01-09 09:19:31
Quote
The Sampling Theorem says that signals containing frequencies up to 1/2 the sampling frequency (S/2) can be exactly represented by S samples per second. Restated, the highest frequency that can be recorded, and played back, is equal to 1/2 the sample rate. It isn't approximately, or "in the neighborhood of", or "a degraded representation of", it is exactly the same waveform, right up to S/2  (e.g. 22, 050 Hz at 44,100 Hz). No information is lost.[{POST_SNAPBACK}][/a] (http://index.php?act=findpost&pid=355646")

I thought that only frequencies under half the sampling rate can be represented exactly (so, Fs/2 is the first frequency that cannot be represented anymore). See also [a href="http://en.wikipedia.org/wiki/Nyquist_theorem]wikipedia[/url].

Quote
There's also the more debatable issue that higher frequencies add to the warmth of the sound, even though they're inaudible on a concious level. Hence the term analog warmth, where valves would add random high frequencies to the audio.
[a href="index.php?act=findpost&pid=355584"][{POST_SNAPBACK}][/a]

Well, as fas as I know these added harmonics are well within audible range.
Title: Why 24bit/48kHz/96kHz/
Post by: Lyx on 2006-01-09 15:17:05
Quote
Firstly as humans, we can all hear up to roughly 22050Hz...
[a href="index.php?act=findpost&pid=355584"][{POST_SNAPBACK}][/a]

The majority of people can NOT hear that high. Most are well below 20KHz, unless you turn up the volume to levels which will certainly damage your hearing.

This thread will end at the same conclusion as every other thread of this kind: for listening, 44KHz/16bits is sufficient and going higher has no benefit(for listening). Thus, the only advantage which media like DVD-A etc will bring is multichannel-support, storage-space for video, etc.

- Lyx
Title: Why 24bit/48kHz/96kHz/
Post by: marcan on 2006-01-09 16:59:57
Quote
Quote
Firstly as humans, we can all hear up to roughly 22050Hz...
[a href="index.php?act=findpost&pid=355584"][{POST_SNAPBACK}][/a]

The majority of people can NOT hear that high. Most are well below 20KHz, unless you turn up the volume to levels which will certainly damage your hearing.

This thread will end at the same conclusion as every other thread of this kind: for listening, 44KHz/16bits is sufficient and going higher has no benefit(for listening). Thus, the only advantage which media like DVD-A etc will bring is multichannel-support, storage-space for video, etc.

- Lyx
[a href="index.php?act=findpost&pid=355747"][{POST_SNAPBACK}][/a]

Absolutely and for multichannel 96 khz is overkill. Actually, compared to stereo, multichannel improves the dynamic (I’d say more than 10 db for 5.1).
There are other formats which are more supported with a better footprint and a good level of quality.
Title: Why 24bit/48kHz/96kHz/
Post by: AndyH-ha on 2006-01-09 17:02:35
My immediately available source is Principles of Digital Audio, fourth edition by Ken C. Pohlmann. What it says is consistent with what I remember reading elsewhere. I don't know about the "Wikipedia" as I get a message it is temporarily unavailable, but isn't that a collection to which anyone who thinks he knows something can contribute?

on page 25:
"a sampling frequency of S samples/second is needed to completely represent a signal with a bandwidth of S/2 Hz. ... For example, an audio signal with a frequency response of 0 to 20 kHz would theoretically require a sampling frequency of 40 kHz for proper sampling."
Further text extends the explanation, making clear that it is indeed S/2 Hz, not S/2 less something.

As for adding "random high frequencies to the audio" for "warmth," another way of saying that is "adding distortion that just happens to please some people."  Of course an original analogue source might have had higher order harmonics that cannot be captured. The Sampling Theorm is quite clear that only a properly bandlimited input can be sampled "correctly."
Title: Why 24bit/48kHz/96kHz/
Post by: SebastianG on 2006-01-09 17:30:07
Quote
on page 25:
"a sampling frequency of S samples/second is needed to completely represent a signal with a bandwidth of S/2 Hz. ... For example, an audio signal with a frequency response of 0 to 20 kHz would theoretically require a sampling frequency of 40 kHz for proper sampling."
Further text extends the explanation, making clear that it is indeed S/2 Hz, not S/2 less something.

As for adding "random high frequencies to the audio" for "warmth," another way of saying that is "adding distortion that just happens to please some people."  Of course an original analogue source might have had higher order harmonics that cannot be captured. The Sampling Theorm is quite clear that only a properly bandlimited input can be sampled "correctly."
[a href="index.php?act=findpost&pid=355778"][{POST_SNAPBACK}][/a]

We're being splitting hairs and in practice it really doesn't matter that much ... but here I go:
I agree with bug80 that it's everything under S/2, not S/2 itself.
Wikipedia is right and the book is not.

AndyH-ha, suppose there's a sine of amplitude 1.0 and frequency 10 kHz, sampled at 20 kHz with a certain phase offset 'ofs'. The samples will be:
1.0*sin(0+ofs) 1.0*sin(pi+ofs) 1.0*sin(2pi+ofs) ...
Now, if you set 'ofs' to pi/2 you'll get
+1 -1 +1 -1 +1  fine
set 'ofs' to pi/6 you'll get
+0.5 -0.5 +0.5 -0.5    whoa! what's that ?
Is it a sampled version of an 1.0-amplutude sine with a phase offset of pi/6 or an 0.5-amplitude sine with a phase offset of pi/2 ???
Even worse: Try a phase offset of zero  ;-)

Truth is: You can't properly sample and reconstruct a frequency of S/2 because there's this ambiguity ... hence you can only represent everything below S/2.


Sebi
Title: Why 24bit/48kHz/96kHz/
Post by: Woodinville on 2006-01-09 23:38:28
Quote
This thread will end at the same conclusion as every other thread of this kind: for listening, 44KHz/16bits is sufficient and going higher has no benefit(for listening). Thus, the only advantage which media like DVD-A etc will bring is multichannel-support, storage-space for video, etc.

- Lyx
[a href="index.php?act=findpost&pid=355747"][{POST_SNAPBACK}][/a]


Well, for some very, very limited sets of circumstances, one might imagine that 120dB might be required, i.e. a young person with good hearing in the quietest room in the world, listening to a very loud orchestra with lots of percussion, miked pretty close, and wanting to hear the crowd rustle and air conditioning sounds between movements.  This would also require quite an extraordinary playback system, indeed. Such playback systems ARE possible, but extremely rare.

But for most (if not all) living rooms, and most people with normal hearing, yes, I think it's pretty much sufficient.

N.B.  Before somebody gets nasty and screams TOS (Look, I'm tired of the audio-woo contingent, too, but not every surprising claim is audio woo.), I'm simply stating something that can be observed from Fletcher's zero-loudness curves.

It would be an extraordinary room and system, but both are possible, with effort and cost, and with a young person who doesn't listen to rock to listen.
Title: Why 24bit/48kHz/96kHz/
Post by: Woodinville on 2006-01-09 23:41:09
Quote
The Sampling Theorem says that signals containing frequencies up to 1/2 the sampling frequency (S/2) can be exactly represented by S samples per second. Restated, the highest frequency that can be recorded, and played back, is equal to 1/2 the sample rate. It isn't approximately, or "in the neighborhood of", or "a degraded representation of", it is exactly the same waveform, right up to S/2  (e.g. 22, 050 Hz at 44,100 Hz). No information is lost.[a href="index.php?act=findpost&pid=355646"][{POST_SNAPBACK}][/a]


True, however, you should also point out that as a filter becomes sharper and sharper, it's length grows without bound, to a filter with infinite sharpness that has infinite length.

For a standard FIR antialiasing filter, this requires infinite delay for both sampling and reconstruction. For an IIR filter, this requires infinite delay after the first filter.

So if you are within dF of fs/2, you need at least approximately 2/dF time for the filter to work. That is approximate, but a reasonable estimate of a minimum. When you get to fs/2, the filter delay is infinite in some fashion.
Title: Why 24bit/48kHz/96kHz/
Post by: Garf on 2006-01-10 11:33:05
Quote
Quote
Another thing to consider is that with proper dithering and noiseshaping, the effective resolution at 44.1kHz can be boosted by as much as 15dB. This means that a properly produced CD (unfortunately very few exist, but HDCD are an example) will have an effective SNR of 111dB.

How many ADC or DAC's have you seen that can attain that resolution?
[a href="index.php?act=findpost&pid=353582"][{POST_SNAPBACK}][/a]

Hi Garf,
It's quite perturbing to talk about 111dB of SNR with a DAC having 96dB (16bits).
Actually dithering/NS decrease distortion while increasing the noise. This noise being created in order to be barely audible, we can say that we have a psychoacoustic SNR of 111 dB on a device with a dynamic of 96 dB.
Is that correct?
[a href="index.php?act=findpost&pid=354797"][{POST_SNAPBACK}][/a]


Yes, that's well-explained. The 96dB is over the entire frequency range. You can make tradeoffs and gain at lower frequencies to lose at higher ones.

If you say that CD has an SNR of 96dB, you must also say that SACD has an SNR of 6dB.

The numbers are from Frank Klemm's page, BTW. He has a list of reasonably attainable audible SNR increase versus actual SNR decrease versus sampling rate, with dither + noise shaping.
Title: Why 24bit/48kHz/96kHz/
Post by: Garf on 2006-01-10 11:42:17
Quote
Well, for some very, very limited sets of circumstances, one might imagine that 120dB might be required, i.e. a young person with good hearing in the quietest room in the world, listening to a very loud orchestra with lots of percussion, miked pretty close, and wanting to hear the crowd rustle and air conditioning sounds between movements. 
[...]
It would be an extraordinary room and system, but both are possible, with effort and cost, and with a young person who doesn't listen to rock to listen.
[a href="index.php?act=findpost&pid=355886"][{POST_SNAPBACK}][/a]


Would you agree with my argument that prolonged exposure to such a large loudness variation (which means necessarily very loud levels at the loudest peaks) would eventually cause hearing damage, thereby making itself useless?
Title: Why 24bit/48kHz/96kHz/
Post by: Woodinville on 2006-01-10 22:41:33
Quote
Quote
Well, for some very, very limited sets of circumstances, one might imagine that 120dB might be required, i.e. a young person with good hearing in the quietest room in the world, listening to a very loud orchestra with lots of percussion, miked pretty close, and wanting to hear the crowd rustle and air conditioning sounds between movements. 
[...]
It would be an extraordinary room and system, but both are possible, with effort and cost, and with a young person who doesn't listen to rock to listen.
[a href="index.php?act=findpost&pid=355886"][{POST_SNAPBACK}][/a]


Would you agree with my argument that prolonged exposure to such a large loudness variation (which means necessarily very loud levels at the loudest peaks) would eventually cause hearing damage, thereby making itself useless?
[a href="index.php?act=findpost&pid=356006"][{POST_SNAPBACK}][/a]


It depends on the actual duration of the loud parts, but in general, I agree that listening over a short-term or long-term mean of 85dB SPL is bad news.

On the other hand, means to measure exposure are still rather primitive, and even those of us who prefer to be in quiet surroundings routinely see peaks well above that. Consider applause, but we don't all go deaf from it.
Title: Why 24bit/48kHz/96kHz/
Post by: crimsontide on 2006-03-13 13:36:57
Quote
I have to step out and disagree that there is no audible difference between 16-bit and 24-bit recordings. The decision to use 16/44.1 was arrived at, not because it was audio nirvana, but because it was the best comprimise possible with the digital to analogue conversion technology of the 80's.

To hear the benifits of 24-bit vs. 16-bit, you need a few things:

1) Capable and descriminating ears. The fact is that no everyone can hear the difference, and that's fine. However, even with a capable ear, the individual has to know what he's looking for. We're talking about very subtle points here, the difference is not going to be night and day.

2) Capable audio equipment. Just because you have a DVD-A or SACD player doesn't mean your system is capable of delivering an audible difference. You need very good equipment to make 24-bit listening worth your while.

3) Quality recording. It doesn't matter if the recording is 64-bits - if it was poorly recorded, and poorly produced, it's going to sound poor. Adding another 8 bits is not a magical fix-all.

Quote
Well, the advantage of DVDA and SACD is exactly that they are multichannel. And have DRM, which is obviously a big advantage to some people.


I disagree with this, especially with respect to SACD. Alot of hardcore audiophiles (the people who would be investing in expensive high bit recordings) feel very strongly that music should only bet two channels.

In SACD multichannel is optional, but 2-channel is required; in other words every SACD will have high-bit two channel audio, but not necessarily multichannel.
[a href="index.php?act=findpost&pid=353234"][{POST_SNAPBACK}][/a]


Agreed on all counts.

Sometimes I think people mistake failed ABX tests (which if successful, prove there IS a difference), with proof that there is NO difference - which is clearly not the case.

Anyone into Reason? - Might I suggest a project of ABXing with certain fully capable sound cards and a 44.1khz 16bit render, and a 24bit 96khz render direct from Reason? All soundwaves are then regenerated to actually use the accuracy available from the end format. Its hard to find CDs that havent been through a 48khz 16bit step at some point for example - this will eliminate that kind of variable compeltely - while unfortunately dislocating the tests from a real world model to some extent .... I think it would be a worthy project.

I have not abxed it but my subjective tests indicated a perceptible warmth from the 96khz render that was missing from the 44khz render. Some scientific results would be beneficial.
Title: Why 24bit/48kHz/96kHz/
Post by: Cyaneyes on 2006-03-13 14:55:17
Quote
I have not abxed it but my subjective tests indicated a perceptible warmth from the 96khz render that was missing from the 44khz render. Some scientific results would be beneficial.
[a href="index.php?act=findpost&pid=371261"][{POST_SNAPBACK}][/a]


Well, I have tried to ABX with good equipment and failed miserably, as have many others I would suspect. If you think you can hear a difference, why not try yourself. You might reach an interesting conclusion...
Title: Why 24bit/48kHz/96kHz/
Post by: crimsontide on 2006-03-13 15:49:27
Quote
Quote
I have not abxed it but my subjective tests indicated a perceptible warmth from the 96khz render that was missing from the 44khz render. Some scientific results would be beneficial.
[a href="index.php?act=findpost&pid=371261"][{POST_SNAPBACK}][/a]


Well, I have tried to ABX with good equipment and failed miserably, as have many others I would suspect. If you think you can hear a difference, why not try yourself. You might reach an interesting conclusion...
[a href="index.php?act=findpost&pid=371272"][{POST_SNAPBACK}][/a]


its not a conclusion though - I made that point before. you can only conclude positive results from ABX, not the other way around.

However I am going to try it.
Title: Why 24bit/48kHz/96kHz/
Post by: Societal Eclipse on 2006-03-13 16:40:36
Quote
Anyone into Reason? - Might I suggest a project of ABXing with certain fully capable sound cards and a 44.1khz 16bit render, and a 24bit 96khz render direct from Reason? All soundwaves are then regenerated to actually use the accuracy available from the end format. Its hard to find CDs that havent been through a 48khz 16bit step at some point for example - this will eliminate that kind of variable compeltely - while unfortunately dislocating the tests from a real world model to some extent .... I think it would be a worthy project.

I have not abxed it but my subjective tests indicated a perceptible warmth from the 96khz render that was missing from the 44khz render. Some scientific results would be beneficial.
[a href="index.php?act=findpost&pid=371261"][{POST_SNAPBACK}][/a]


I have been using Reason on and off but cannot test myself as my card only goes up to 48khz/24bit.
Title: Why 24bit/48kHz/96kHz/
Post by: stephanV on 2006-03-13 16:49:40
@crimsontide:

You can never prove a negative. The point you are trying to make is made often before... but it doesn't help anyone any further.

If you think there is an audible difference, please prove it under controlled circumstances.

Otherwise, if it seems no one yet (here) can prove to hear a difference, wouldn't it be more reasonable to assume that the difference is really inaudibble than to just keep assuming that it is, just because an abx-test can only prove positives?

If you would have seen a 1000 specimens of the same type of bird, and they were all black, would you assume the next one you see would also be black, or would you think it would be white?
Title: Why 24bit/48kHz/96kHz/
Post by: crimsontide on 2006-03-13 17:32:00
Quote
@crimsontide:

You can never prove a negative. The point you are trying to make is made often before... but it doesn't help anyone any further.

If you think there is an audible difference, please prove it under controlled circumstances.

Otherwise, if it seems no one yet (here) can prove to hear a difference, wouldn't it be more reasonable to assume that the difference is really inaudibble than to just keep assuming that it is, just because an abx-test can only prove positives?

If you would have seen a 1000 specimens of the same type of bird, and they were all black, would you assume the next one you see would also be black, or would you think it would be white?
[{POST_SNAPBACK}][/a] (http://index.php?act=findpost&pid=371304")


You have a good reasoning technique - however assumption does not belong in my scientific vocabulary.

subjective tests have more value than assumption in my book. Even with all those horrible random variables.

BUT NEVERTHELESS

I will prepare some samples and get ABXin.
Unfortunately reason is not installed and i cant find my discs so.....

I'm using these lovely samples from this lovely thread.
[a href="http://www.hydrogenaudio.org/forums/index.php?showtopic=17118]http://www.hydrogenaudio.org/forums/index....showtopic=17118[/url]

http://www.kikeg.arrakis.es/various/lovely_test.zip (http://www.kikeg.arrakis.es/various/lovely_test.zip) posted by kike G
Title: Why 24bit/48kHz/96kHz/
Post by: harlekeyn on 2006-03-19 05:58:23
16bit means 2^16 = 65536 possible values
for each 1/(44.1*10^3) portion of a second.

24bit would hence be 2^24=16.7 million values.

That is 2^(24-16) = 2^8 = 256 as many.

The misunderstanding people make is that one would
not really be able to hear the difference.
The curve is round enough. We are just humans.

But here is what they forget:

=== An example: ===
When you listen to a live recording,
the band plays at normal volume, with their
peaks till the full range of these (when recorded
in 16bit) 65536 values.

But the band does not always play loudly.
And even while they do, there are also sounds
that are not so loud. The drummer might
hit his ride gently.

Those rides have an interesting colorful sound.
A good drummer would be able to guess the brand
of the rides, with his eyes closed.
And also estimate what type of drum stick is used.

Since these rides are but played softly,
their range lies not in the full 65536 values,
but, say, only from +6% to -6%.

If one would be able to mute all loud instruments
(for this explanatory example), we would only
hear the rides in 3932 'blocks', or approximately 12bit.

If you now would turn up your volume
(of the headphones you use to listen to all this),
you would not hear an elegant ride sound at all.

So, in summery, a 24bit recording sounds nicer
than a 16bit recording because you can hear the
soft and gentle sounds better.

Someone in audience using a teaspoon and cup to
add some unasked percussion.
We can hear the stage has a wooden floor because
the horn player taps his foot.

Tristan

PS. I have not read all posts in this thread,
so forgive me if I am repeating something
already mentioned.

PPS. I am just an amateur too.
I could be mistaken in my details,
but I think the essence is correct.
Title: Why 24bit/48kHz/96kHz/
Post by: Pio2001 on 2006-03-19 21:07:37
Quote
If you now would turn up your volume[{POST_SNAPBACK}][/a] (http://index.php?act=findpost&pid=372423")


If you turn up your volume loud enough, you can always hear some noise.
It would be more interesting to use some extremely dynamic samples, and play them in a silent environment at the loudest possible volume for music listening. This would simulate the "worst case" situation.

Studying the advantage of high sample rates is extremely difficult.

First, one needs "super tweeters" in order to play ultrasonic content. Few speakers are capable of playing back sounds above 20 kHz.
Then, in [a href="http://www.davidgriesinger.com/intermod.ppt]the last version of his paper[/url], David Griesinger reports that intermodulation distortion at high frequencies, that can make a "counter-difference" between 44100 Hz and 96000+ Hz (96 khz sounding worse), occurs mainly in amplifiers, rather than in speakers.
Even with super tweeters, an amplifier may then produce distortion when fed with ultrasonic content.
On the other hand, there are evidences that ultrasonic intermodulation occurs in air : http://www.atcsd.com/hss.html (http://www.atcsd.com/hss.html)
However, it occurs at high sound pressure. Around 130 dB, for example. Ultrasonic content in musical instrument is below 60 dB. Moreover, in order to produce audible ultrasonic intermodulation, focused sonic beams are used, while a music instrument radiates sonic energy in all directions. It seems that ultrasonic intermodulation in air should be inaudible, but I don't know about studies on this.

Scientific papers featuring blind listening tests about high definition sound are rare.
Two are published on the web :
http://jn.physiology.org/cgi/content/full/83/6/3548 (http://jn.physiology.org/cgi/content/full/83/6/3548)
http://www.nhk.or.jp/strl/publica/labnote/lab486.html (http://www.nhk.or.jp/strl/publica/labnote/lab486.html)

The first one shows positive results, but give absolutely no details on how the statistic confidence have been evaluated. The result is even considered as minor, since it was not the goal of the experiment, that focuses on electro-encephalograms.
The second link above is a similar experiment, that failed.

I've heard about a third one, discussed here : http://www.hydrogenaudio.org/forums/index....st&p=307155 (http://www.hydrogenaudio.org/forums/index.php?showtopic=31759&view=findpost&p=307155)
However, the role of ultrasonic content in the success of this test is unsure : http://www.hydrogenaudio.org/forums/index....st&p=307619 (http://www.hydrogenaudio.org/forums/index.php?showtopic=31759&view=findpost&p=307619)

But unscientific blind listening tests have sometimes showed that CD players operating at 44100 Hz 16 bits are not fully transparent :
The 24/96 challenge : http://www.hydrogenaudio.org/forums/index....showtopic=17118 (http://www.hydrogenaudio.org/forums/index.php?showtopic=17118)
Analog copies from two different CD players : http://www.homecinema-fr.com/forum/viewtopic.php?p=169329500 (http://www.homecinema-fr.com/forum/viewtopic.php?p=169329500)
Analog copy (44100 Hz 16 bits playback -> 48000 kHz 16 bits recording) vs resampled digital file : http://www.hydrogenaudio.org/forums/index....f=21&t=6651 (http://www.hydrogenaudio.org/forums/index.php?act=ST&f=21&t=6651)

All these tests deal with files undergoing digital filtering near 22 kHz, and 16 bits rounding or dithering.
One can argue that 44100 Hz and 16 bits are enough in order to acheive transparency on a decent hardware, and that the above tests use samples suffering from bad recording or processing. But I listened to the samples of the first comparison in the second link, and could not hear any difference. The guy who got 8/8 have got better ears than me.

So we can wonder if raising the resolution of the digital format could improve the sound quality in these cases. I cannot tests this hypothesis because my listening abilities are not good enough to make the difference between two CD players.
Title: Why 24bit/48kHz/96kHz/
Post by: crimsontide on 2006-03-20 10:03:32
Well I agree with what Tristan said – For example on an old Led Zep, or to pick a rarer example Fleetwood Mac – Analogue Mix from 30 years ago, the dynamics were used in a more classical style.

Of course since many people listen in the car these days, or via the radio – most producers compress the music so this hasn’t much bearing on newer recordings.

By a classical style – I mean the overall volume level is in the lower regions of the dynamic range, saving the higher volumes for really powerful moments – Which is one of the benefits you get from Vinyl as well when compared to CD (or so enthusiasts would have you know). This would result in much less accuracy in the reproduction on 16/44 than 24/96 can afford.

So I think as long as the original mixdown is used and recorded direct to 24/96 – you might be able to ABX a quieter moment of a classic Fleetwood Mac record and have reasonable results. It’s a fair comment – scientifically.

HOWEVER

I tried that ABX from above and got only fractionally above complete chance. The results weren’t even worth noting down.
So I humbly concede that in practical applications – in 2006 – The difference between 24/96 and 16/44, is scientifically negligible, even if provable.

I didn’t believe it – I didn’t want to - but I couldn’t tell the difference.

Regards

Jon
Title: Why 24bit/48kHz/96kHz/
Post by: Woodinville on 2006-03-20 20:19:34
Quote
I tried that ABX from above and got only fractionally above complete chance. The results weren’t even worth noting down.
So I humbly concede that in practical applications – in 2006 – The difference between 24/96 and 16/44, is scientifically negligible, even if provable.

I didn’t believe it – I didn’t want to - but I couldn’t tell the difference.

Regards

Jon
[a href="index.php?act=findpost&pid=372871"][{POST_SNAPBACK}][/a]


Understood, but consider a couple of things. The atmosphere itself puts something like 6dB SPL white noise at the eardrum.

16 bits up from that is 102dB.

Now, how often do you listen to peaks above 102dB?

Note, we have not even discussed, yet, room noise, hearing loss, etc. So that explains why when it's done right, 16 bits shouldnt' be too problematic. I do expect one might be able to design a signal that, in a quiet area, caused a problem. I wonder, however, if the average good loudspeaker or headphone could actually reproduce it with anything approaching "fidelity".

Now, to 44.1 vs. 96.  Something you might try is to create some "dummy" data that might distinguish on very contrived signals.  I can't dismiss that outright, but I'd suggest that you try some broadband stimulii created at 96, and then the same downsampled to 44.1. The bit depth doesn't matter for this exercise.

Said stimulii ought to be something with both tonal (i.e. sinusoidal) components and peaky components (for isntance a gaussian pulse centered at 15kHz that's down to -60dB at 30kHz and DC... Or something like that. Perhaps a center frequency for which the aliasing for the 44.1 case would be obnoxious.
Title: Why 24bit/48kHz/96kHz/
Post by: harlekeyn on 2006-04-04 16:48:53
WHEN COMPARING digital audio to digital photography,
there is nothing wrong with matching :

            • bitrate < = > color depth, and 
            • sample-frequency < = > resolution.

With a higher bitrate, you can hear the tiny sounds-shades
with more accuracy as result of more variation.
Zoomed into the sound of, say, a cricket,
in stead of hearing just a soft and distant '' c r i c k e t '',
you can distinguish '' c r I c k ê t  ''.
More subtlety.

But, seeing the difference between light brown and
very slightly lighter brown is not easy.
You would have to use great concentration
and training to do so, if capable at all.

You look at a digital photo of a piece of wood.

1) The smallest details are seen because you have
    a high resolution image (compare: the sample frequency).
    It's sharp.

2) Now, the colors.
    If you would use software to increase the contrast,
    the patterns of wood grains become more obvious.
    This would be comparable to turning up the volume
    when listening to a quiet, subtle sound.

The next argument is clearly:

      Yes, all fine, but when listening to a music
      recording I cannot constantly change the volume.


Or --

      There is just a limit to my hearing.
      I -can- hear that a 4 bit recording, with
      a mere 2^4=16 possible values between loud and silent,
      is -not- enough for hearing soft, subtle background sounds.
      But, my brain and ears are limited and need no more
      than 2^16=65536 possible values between loud and silent.


                              . . . Note: I am not even sure if
                                    16 bit implies 2^16 values between
                                    +1 and -1, or between +1 and 0.
                                    Probably the former . . .

By typing the above paragraphs, I have not
succeeded at arguing one's mind does
appreciate more accuracy in amplitude.
Nor: the hypothesis that a human mind
is fast enough (read: fine enough) to prefer
a higher sample-frequency.

[ in the latter case, compare:
a digital camera with a higher mega-pixel capacity, or,
fingertips sensing silk woven with a finer thread. ]

Besides that I am still convinced I personally cán hear (feel?)
the difference between a normal audio cd and one of the
newcoming higher quality digital recordings,
the only comment I can make here is the following :

It does not suffice to blindfold people and ask if
they notice any difference between two recordings
of the same Fleetwood Mac track, one in cd-audio quality
and one in higher quality (with good speakers).

Rather, I would choose to count how many people
start to dance in a club. Or, what time they go home.
How much they enjoy it. ( <-- difficult to measure
of course, but the first two not, if done on large scale )

A final note.
      After a digital signal is analogue again,
it does not reach your ears as 'blocky' as
the digital wave form. The cone of the speaker
that produces the sound has some weight
and thus will round the sound.
      I know not enough about the kinetics and
materials (and..., and...) to say anything more
about this. Just that it is not to be overlooked.

                        Tristan Laurillard,
                        Vancouver, Canada


PS. Not everything I wrote here is verified knowledge.
Just my own understanding of these matters.
Title: Why 24bit/48kHz/96kHz/
Post by: benski on 2006-04-04 16:56:52
With somewhat expensive studio equipment (RME Hammerfall digital soundcard feeding DACs with about 110dB SNR, Mackie HR-824 speakers [120dB SPL, 102dB SNR], nice analog mixer) and professionally recorded 24bit source tracks, auditioned individually, I could readily differentiate 24bit and 16bit output (internal audio path was always 32bit).  48kHz versus 96kHz sounded identical.  However, once the tracks were mixed, there was no discernable difference between 16 and 24 bit.
Title: Why 24bit/48kHz/96kHz/
Post by: Garf on 2006-04-04 17:00:19
Quote
Besides that I am still convinced I personally cán hear (feel?)
the difference between a normal audio cd and one of the
newcoming higher quality digital recordings,
the only comment I can make here is the following :

It does not suffice to blindfold people and ask if
they notice any difference between two recordings
of the same Fleetwood Mac track, one in cd-audio quality
and one in higher quality (with good speakers).

Rather, I would choose to count how many people
start to dance in a club. Or, what time they go home.
How much they enjoy it. ( <-- difficult to measure
of course, but the first two not, if done on large scale )


This means that the difference is measurable in a blind test. Blind doesn't mean people must be blindfolded or anything like that! It means that a scientific test was done where nor the testees, nor the tester, knew what was being used, to rule out any "placebo" effects.

But there is simply zero solid proof the new formats have ANY effect at all.

Quote
A final note.
      After a digital signal is analogue again,
it does not reach your ears as 'blocky' as
the digital wave form. =
[a href="index.php?act=findpost&pid=379069"][{POST_SNAPBACK}][/a]


You seem to have the complete misconception that a digital signal is "blocky", but in fact, it cannot be blocky, because it's bandlimited!
Title: Why 24bit/48kHz/96kHz/
Post by: crimsontide on 2006-04-04 17:07:39
Quote
Understood, but consider a couple of things. The atmosphere itself puts something like 6dB SPL white noise at the eardrum.

16 bits up from that is 102dB.

Now, how often do you listen to peaks above 102dB?

Note, we have not even discussed, yet, room noise, hearing loss, etc. So that explains why when it's done right, 16 bits shouldnt' be too problematic. I do expect one might be able to design a signal that, in a quiet area, caused a problem. I wonder, however, if the average good loudspeaker or headphone could actually reproduce it with anything approaching "fidelity".

Now, to 44.1 vs. 96.   Something you might try is to create some "dummy" data that might distinguish on very contrived signals.  I can't dismiss that outright, but I'd suggest that you try some broadband stimulii created at 96, and then the same downsampled to 44.1. The bit depth doesn't matter for this exercise.

Said stimulii ought to be something with both tonal (i.e. sinusoidal) components and peaky components (for isntance a gaussian pulse centered at 15kHz that's down to -60dB at 30kHz and DC... Or something like that. Perhaps a center frequency for which the aliasing for the 44.1 case would be obnoxious.
[a href="index.php?act=findpost&pid=373052"][{POST_SNAPBACK}][/a]


1) Ambient Noise will always be present - and I cant change my ears either - therefore I consider both as constant - K. Therefore they can be ignored from the factor.

2) how often do i listen to peaks above 102db? erm..... who knows or cares? would i miss them if they were gone? probably not. The statement doenst make complete sense, since that db level is contained with 0 -102db on the cd - so that peak is INSIDE the dynamic range. I'm not missing out on anything - but they have compressed the dynamic rangte to fit. The point is: can i tell and do I care?

3) Preparing a sample you'd never listen to which can be abxed is fruitless.

consider this discussion as the choice between a 4*4 off roader and an F1 car.

Sure you could race them on different tracks and get different results, one better than the other in many ways.

Now consider this discussion is only interested in a vehicle which can be used for every application, with good performance in all..........the f1 car might rip it on the race track, but give it a pot hole and you're calling the AA.

My point is you must be practical with your investigation, and not seek to denounce anything, nor prove anything - just form a conclusion from your results.

I did that - and im now convinced that it really doesnt matter between the two formats.......and i already have a cd player in almost every room in the house. Ill go for CD thanks, its fine.......even if it does feel like an old rusty 4*4 in comparison..........it'll get me there - and much cheaper.
Title: Why 24bit/48kHz/96kHz/
Post by: crimsontide on 2006-04-04 17:15:06
Quote
With somewhat expensive studio equipment (RME Hammerfall digital soundcard feeding DACs with about 110dB SNR, Mackie HR-824 speakers [120dB SPL, 102dB SNR], nice analog mixer) and professionally recorded 24bit source tracks, auditioned individually, I could readily differentiate 24bit and 16bit output (internal audio path was always 32bit).   48kHz versus 96kHz sounded identical.  However, once the tracks were mixed, there was no discernable difference between 16 and 24 bit.
[a href="index.php?act=findpost&pid=379072"][{POST_SNAPBACK}][/a]


thats what im talkin about

I agree with everyone - even though i appear to have my own viewpoint!!!

I believe i can hear the difference too - if the tracks were split out - but I dont listen to music like that, so I have to discard abxing custom samples - and stick to the usual professional mixdowns I'm so concerned sound good....
Title: Why 24bit/48kHz/96kHz/
Post by: Garf on 2006-04-04 17:15:21
Was the 16-bit signal properly dithered+noiseshaped on playback? A properly dithered+noisehshaped 16-bit signal has an effective SNR of 111dB. This is more than your DAC's.

I don't see the sense in comparing not properly dithered 16-bit signals to the new generation formats, when one of those (SACD) has an SNR of 6dB when not dithered!
Title: Why 24bit/48kHz/96kHz/
Post by: Pio2001 on 2006-04-04 18:29:07
Quote
I could readily differentiate 24bit and 16bit output (internal audio path was always 32bit).   48kHz versus 96kHz sounded identical.  However, once the tracks were mixed, there was no discernable difference between 16 and 24 bit.
[{POST_SNAPBACK}][/a] (http://index.php?act=findpost&pid=379072")


Maybe, maybe, but outside controlled blind listening test, your experience bring no more information to us than the hundreds other user opinions. Remember the Terms of Service number 8 of this forum : [a href="http://www.hydrogenaudio.org/forums/index.php?showtopic=3974]http://www.hydrogenaudio.org/forums/index.php?showtopic=3974[/url]
Statistically significant blind listening tests only (see FAQ or Wiki).

Quote
It does not suffice to blindfold people and ask if
they notice any difference between two recordings
of the same Fleetwood Mac track, one in cd-audio quality
and one in higher quality (with good speakers).[a href="index.php?act=findpost&pid=379069"][{POST_SNAPBACK}][/a]


It suffices in order to test the claim that people notice obvious differences between two recordings of the same Fleetwood Mac track, one in cd-audio quality
and one in higher quality (with good speakers).

Quote
Rather, I would choose to count how many people start to dance in a club. Or, what time they go home.
How much they enjoy it. ( <-- difficult to measure
of course, but the first two not, if done on large scale )[a href="index.php?act=findpost&pid=379069"][{POST_SNAPBACK}][/a]


These can't lead to reliable statistics. The tested hypothesis must be clearly defined, and the measured parameters must be chosen before the test.

Beginning with the number of people starting to dance, you will get subjective evaluation problems. Is this guy dancing on his seat ? Is that one dancing on the way to the toilets ? I this one one the dancefloor or next the dancefloor ?
Then if you don't get the result you like, you will test the hour of people going home, and if it fails, the highest they jump, and if it doesn't show any correlation, the number of drinks they buy, their distance to the speakers, the number of couples looking at each other, etc
The number of testable things is infinite, thus the probability that you get a statistically significant result out of nothing is equal to 1.

The measured parameters must be the ones relevant to the tested hypothesis : if high definition formats are supposed to sound better, we must test if they sound better, and nothing else. And in order to do this, we must ask listeners if the sound is better or not.

The refinements that can improve the test performance are in the way the listeners are allowed to listen. It is perfectly right, and even recommended when testing small differences, to choose listeners that can easily hear these differences, and then let them train themselves with no time or protocol constraints.
The most relevant test would be the one made by the trained listener, after he has found the protocol (AB, ABX, ABA, AXY, etc), and listening conditions that allow him or her to always get statistically significant results during the training sessions.

Imposing a given musical content, a given sample duration, and demanding listeners to write down any answer after each listening session, played only once, are the most important obstacles to the success of such blind listening tests in my opinion.
I think that relaxing these restrictions, while maintaining the randomness and overall statistical requirements of the final test, should allow blind tests to show the existence of smaller audible differences than they usually do.
Title: Why 24bit/48kHz/96kHz/
Post by: jmartis on 2006-04-09 18:25:19
sorry if this was already posted, but 96k may have its benefit in reproducing (synthetic) sharp square wave (or saw-wave)- it always adds sub harmonic frequencies to it, but 96k should add less than 44k (in a very audible way) but when was the last time you heard a synthetic square wave in real music.
Title: Why 24bit/48kHz/96kHz/
Post by: Garf on 2006-04-09 21:21:30
sorry if this was already posted, but 96k may have its benefit in reproducing (synthetic) sharp square wave (or saw-wave)- it always adds sub harmonic frequencies to it, but 96k should add less than 44k (in a very audible way) but when was the last time you heard a synthetic square wave in real music.


Might be a good idea to read the thread you are replying to next time, I am sure that it would be quite enlightening.
Title: Why 24bit/48kHz/96kHz/
Post by: krabapple on 2006-04-13 23:43:15
The late Julian Dunn wrote a paper on dynamic range (bitdepth) requirements recording an distribution (playback)  in 1992

http://www.nanophon.com/audio/dynrange.pdf (http://www.nanophon.com/audio/dynrange.pdf)

essentially concluding that 16 bit with noise shaping was sufficient for playback, while 20 bit was miminum for recording.
Title: Why 24bit/48kHz/96kHz/
Post by: Woodinville on 2006-04-14 00:35:05
The late Julian Dunn wrote a paper on dynamic range (bitdepth) requirements recording an distribution (playback)  in 1992

http://www.nanophon.com/audio/dynrange.pdf (http://www.nanophon.com/audio/dynrange.pdf)

essentially concluding that 16 bit with noise shaping was sufficient for playback, while 20 bit was miminum for recording.


Well, let's see. Atmospheric noise at the eardrum is 6dB SPL give or take.  96dB over that is 102dB. Even without noise shaping, that's pretty close to the loudest most speakers can go.

How much does one assume they can get from noise shaping, and what shape are we proposing here?

While we're at it, what does any frequency over 30kHz have to do with human audition at all?
Title: Why 24bit/48kHz/96kHz/
Post by: Sgt_Strider on 2006-04-16 03:05:23
So let me get this straight, DVD-A and SACD will bring me no benefit in sound quality? There's no need to make the transition from CD to those formats?
Title: Why 24bit/48kHz/96kHz/
Post by: Cyaneyes on 2006-04-16 04:21:15
So let me get this straight, DVD-A and SACD will bring me no benefit in sound quality? There's no need to make the transition from CD to those formats?


They do if they feature superior mastering compared to their CD counterpart. Check out my thread from last summer comparing Porcupine Tree's Deadwing CD vs. DVD-A. http://www.hydrogenaudio.org/forums/index....topic=35572&hl= (http://www.hydrogenaudio.org/forums/index.php?showtopic=35572&hl=)
Title: Why 24bit/48kHz/96kHz/
Post by: Audionut on 2006-04-16 05:59:32
A piano's highest frequency is 4186Hz.  So, fuck it, let's sample the next-gen media down so that it only produces's frequency's up-to 4186Hz.

But wait, there's more!!

Quote
The tone with the lowest frequency is called the fundamental. The other tones are called overtones If the overtones have frequencies that are whole number multiples (x2, x3...up to x14) of the fundamental frequency they are called harmonics.  It is the difference in the harmonic content of notes that gives each musical instrument its characteristic sound or timbre ("tam-brah"). Therefore although the highest note of a piano has a fundamental frequency of just over 4kHz, equipment used to record music must be able to handle much higher frequencies to preserve the harmonics associated with each note.

Sounds produced by percussive effects are particularly rich in high harmonics. These occur mainly at the start of a sound, e.g. when a stringed instrument is plucked or a cymbal is struck. These starting transients are also characteristic of the instrument producing them. Sound equipment must be able to cope with these high frequencies otherwise the tonal quality of the sounds will be altered. Cymbals, for example, can produce frequencies around 20kHz to 25kHz.


http://www.users.globalnet.co.uk/~bunce/sound.htm (http://www.users.globalnet.co.uk/~bunce/sound.htm)


There's Life Above 20 kilohertz (http://www.its.caltech.edu/~boyk/spectra/spectra.htm)

So as you can see, there is a need for increased frequncy.

But hang on.  My ears can only hear frequency's from 20Hz-20khz.  Belive it or not, your subconscious mind can perceive sounds that your conscious mind cannot.

So while, Joe "doesn't know anybetter" Blow, who has never been to an orchestra, thinks that the piano he is listening to, derived from a frequncy limited 128kbps mp3, sounds transparent.
I think it sounds shit-house.


Piano, organ, bass drum all produce sound higher than 96db.  The bass drum alone can produce up-to 115db.  Mix in a few other intruments with it, and you can produce upwards of 120db.
It's all about dynamic range.  And just because when your at a live orchestra, the dynamic range is say 90db due to conversation.  When your listening to a recording at home, you should not be limited to the same dynamic range.

If your happy listening to music recorded in mp3 at 192kbps, Fine.
But don't bitch at companies that are helping the likes of myself, Enjoy music.
edit: or bitch at others, because "you" can't perceive the improvement.

Audionut.
Title: Why 24bit/48kHz/96kHz/
Post by: WmAx on 2006-04-16 06:32:51
There's Life Above 20 kilohertz (http://www.its.caltech.edu/~boyk/spectra/spectra.htm)

So as you can see, there is a need for increased frequncy.


No conclusive evidence found at that link to argue for increased bandwidth. The referred Ooashi paper was rejected for publication in the JAES(it never made it past submitted preprint status) and the Journal that did publish the paper(The Journal of Neurophysiology) only did so, classifying the paper as a paid for advertisement. The Ooashi paper never showed conclusive evidence of audibility(the mentioned listening test in the paper did not elaborate any of the test detail specifics, and NHK labs later did a follow up, and could not reproduce the audibility claimed results by Ooashi). Also, the Ooashi paper had very questionable results in the MRI scans. Note that no activity was detected for ultrasonic information by itself as a stimulus; only when both sonic and ultrasonic was produced. Very odd. Makes one wonder if something was up with their electronics or playback system(s). Or maybe, the entire paper is the result of poor researchers, or even fraudulent, since the only publication it could achieve is one that was an advertisement. I wonder if the MRI results(which are not proving any audibility) are reproducible.

-Chris
Title: Why 24bit/48kHz/96kHz/
Post by: audiofile on 2006-04-16 08:55:19
I think many of you here are missing the exciting aspect of the SACD/DVD-A, partly because you seem to be mainly focused on CD technology and it's accompanying MP3 format, and also, as suggested above, you're approaching it from a statistics-based view.

The fact is, it's basically an upgrade. And, of course Sony and the other companies are largely implementing it to make people buy more of their stuff, what else would be expected--but they can do all sorts of things to make people buy more stuff, which they've done. They seem to have done a piss-poor job of marketing SACD, so the idea that they've worked so hard on a new technology as a ploy to make people have to commit to a whole new technology doesn't seem to make sense. And you would think that's what they would do, but for some reason they haven't. Maybe they're waiting for some right moment.

I personally had been thinking about this for a while, before I had ever even heard of SACD and DVD-A, about the idea of higher density CDs; it just made sense to me. Then I finally learned of them and learned that for some reason they aren't being pushed at all, besides the fact that people are much more excited about MP3s than CDs that are higher than 44.1 kh and 16-bit, which is something that wouldn't mean anything to most people anyway.

The other major issue about CDs is that for many people Vinyl is still the highest fidelity, or at least the best-sounding format, and part of what people resisted about CDs in the first place is the idea listening to contiguous moments of music put together, where you're not actually hearing all of the music but a kind of simulated version. So basically, the way I would think about it, the higher the sampling rate (i.e. 196 kh), the closer you're coming to something that sounds more like analog and less "digital", while also having the ease of the digital format.

I'm not sure it's so much a matter of increasing the dynamic range as capturing more of the music, which I would assume the newer formats do.
Title: Why 24bit/48kHz/96kHz/
Post by: Firon on 2006-04-16 09:04:02
This is basically a repeat of an earlier discussion...
http://www.hydrogenaudio.org/forums/index....topic=41162&hl= (http://www.hydrogenaudio.org/forums/index.php?showtopic=41162&hl=)

And here's a very nice discussion about the higher sampling rates from James Johnston (co-inventor of MP3) not being necessary. http://www.skepticforum.com/viewtopic.php?...der=asc&start=0 (http://www.skepticforum.com/viewtopic.php?t=85&postdays=0&postorder=asc&start=0)
Title: Why 24bit/48kHz/96kHz/
Post by: WmAx on 2006-04-16 09:06:16
...


You should note the title of this forum[Scientifid/R&D Discussion]. Your post does not seem to fit the discussion.
Did you read this thread?

-Chris
Title: Why 24bit/48kHz/96kHz/
Post by: audiofile on 2006-04-16 09:15:25

...


You should note the title of this forum[Scientifid/R&D Discussion]. Your post does not seem to fit the discussion.
Did you read this thread?

-Chris


I read the first couple of pages or so. Maybe I missed something. Edit: Actually I read a lot of the first page and then read the last page. I think my post fits the discussion because I was addressing directly why some people would want higher bitrates/kHz, but maybe he was just asking about it whether it was technically a necessity whereas I just said why people would want it. I'm quite tired.
Title: Why 24bit/48kHz/96kHz/
Post by: legg on 2006-04-16 15:47:05
So basically, the way I would think about it, the higher the sampling rate (i.e. 196 kh), the closer you're coming to something that sounds more like analog and less "digital", while also having the ease of the digital format.


http://cm.bell-labs.com/cm/ms/what/shannon...shannon1948.pdf (http://cm.bell-labs.com/cm/ms/what/shannonday/shannon1948.pdf)

Theorem 13, page 34.
Title: Why 24bit/48kHz/96kHz/
Post by: Lyx on 2006-04-16 16:08:57
"Do we need more bits/samplerate?" Episode #543.... location: hydrogenaudio.org....... pagenumber #5...... still trying to find even a subtle difference.......

Those "obvious difference" must be hiding themselves really good......... good enough that almost no one can notice it..... mean, we're on THE forum for knowledgeable people regarding psychoacoustics including quite a few golden ears..... and even they after 5 pages still cannot tell the difference...... if its so fucking hard to notice at all, then why the hell should we need it, even if the difference actually exists? How useful is something in practice which makes almost no or no difference at all?

Or is the reason for the talk in this thread by any chance not about if there actually is a difference, but more about that some people want to BELIEVE that there is a difference? If yes, then why the heck are you discussing this on ha.org? You will get much more support to believe somewhere else.

- Lyx
Title: Why 24bit/48kHz/96kHz/
Post by: bug80 on 2006-04-16 22:15:03
Title: Why 24bit/48kHz/96kHz/
Post by: Kees de Visser on 2006-04-17 00:12:35
But seriously. I sometimes think of stopping to argument with people who claim that SACD and DVD-A really sounds much better and more natural and blabla, without performing proper blind testing.


Agreed, and that's why I'd like to investigate the possibility of a "professional user listening test" in a (top-quality) recording studio. It will be very difficult to set up a "proper" listening test so I'm hoping to get some useful information here, even though HA seems mainly targeted at endusers. Any insights are more than welcome. The test is still in an early (alpha) stage, so most options are still open.

A serious test would take quite some time, with quite some listeners and quite some equipment. A period of 3 days seems to be technically and financially feasible.
The studio can provide:
-high quality musical instruments as a source
-several recording rooms with very low noise levels
-several (large) mixing consoles, both analog and digital, as used in normal production
-a large choice of top quality microphones, pre-amps, AD/DA converters (both PCM and DSD) and monitoring equipment (stereo and surround). Any equipment that isn't already available can probably be arranged for the test.
-several experienced, professional recording/mixing/mastering engineers. Total group should probably be limited to about 20 people.

My assumption is that in order to compare (subtle) differences in audio devices, it is important to have a high quality source. In my view that has to be a microphone signal, fed into a high-quality pre-amp. I doubt if you can get higher quality than this (assuming acoustical music reproduction).

The test will be double blind with a reasonably long learning period (within the 3-days of the test).
Some of the potential contributors don't like ABX tests so another (double blind) system has to be agreed upon. I still have to collect opinions about the preferred method.

I'm basically trying to find out if a test like this can be done at all. One of the problems I foresee is that audio professionals might not even be interested in finding out what's just good enough, but want to use the best equipment they can afford, even if that means overkill.

In the eyes of HA, how should this listening test ideally be set up ?

Thanks for your input.
Title: Why 24bit/48kHz/96kHz/
Post by: krabapple on 2006-04-17 00:30:33

The late Julian Dunn wrote a paper on dynamic range (bitdepth) requirements recording an distribution (playback)  in 1992

http://www.nanophon.com/audio/dynrange.pdf (http://www.nanophon.com/audio/dynrange.pdf)

essentially concluding that 16 bit with noise shaping was sufficient for playback, while 20 bit was miminum for recording.


Well, let's see. Atmospheric noise at the eardrum is 6dB SPL give or take.


Fiedler sets the minimum at 4.

Quote
96dB over that is 102dB. Even without noise shaping, that's pretty close to the loudest most speakers can go.

How much does one assume they can get from noise shaping, and what shape are we proposing here?

While we're at it, what does any frequency over 30kHz have to do with human audition at all?



Did you read the paper?  Or do you imagine I'm actually advocating greater-than-redbook distribution parameters?

I'm not.  20 bits for *recording and production* has a rational basis, though.
Title: Why 24bit/48kHz/96kHz/
Post by: AndyH-ha on 2006-04-17 04:22:46
Of course you are familiar with the arguments of Brother ... Oh, excuse, I though this was the angels and pin heads discussion.
Title: Why 24bit/48kHz/96kHz/
Post by: legg on 2006-04-17 04:37:21
Some of the potential contributors don't like ABX tests so another (double blind) system has to be agreed upon.


ABX seems the only option if you expect a clear cut answer to "is there an audible difference ?".

There's ABC/HR, but you would have to make it ABX style (several trials for each test) to tell if they could hear an impairment or if it was mere luck.
Title: Why 24bit/48kHz/96kHz/
Post by: Firon on 2006-04-17 06:43:17
They don't like ABX because they know damn well that if they do it, they won't be able to hear any difference at all, and because they can't hear a difference, they claim that ABXing somehow screws up your hearing/perception and makes you miss those "subtle differences".
Title: Why 24bit/48kHz/96kHz/
Post by: kwwong on 2006-04-17 09:27:55
They don't like ABX because they know damn well that if they do it, they won't be able to hear any difference at all, and because they can't hear a difference, they claim that ABXing somehow screws up your hearing/perception and makes you miss those "subtle differences".


That is why ABXing is meant for the experts! During MPEG standardization processes, a team of audio experts can hear the differences. You need to "train" your hearing in order to tell the differences- plain practical skills.
Title: Why 24bit/48kHz/96kHz/
Post by: Firon on 2006-04-17 10:03:42
Even if you do train your hearing, you still can't magically hear frequencies higher than what physical limitations let you hear (not including insane tests with SPLs that are probably harmful to you)...

I'm not sure if you're being sarcastic or anything. If you're not, do you have any actual proof that higher sampling rates (for the SOURCE material, not oversampling anti-aliasing filters) actually make an audible difference? Especially from these so-called audio experts you speak of. I sure can't find any.
Title: Why 24bit/48kHz/96kHz/
Post by: krabapple on 2006-04-18 00:12:13
They don't like ABX because they know damn well that if they do it, they won't be able to hear any difference at all, and because they can't hear a difference, they claim that ABXing somehow screws up your hearing/perception and makes you miss those "subtle differences".


Congratulations.  You are now qualified to write editorials for Stereophile magazine.
Title: Why 24bit/48kHz/96kHz/
Post by: Firon on 2006-04-18 00:50:49
I don't really get it, because I don't know anything about that magazine. 
Title: Why 24bit/48kHz/96kHz/
Post by: Lyx on 2006-04-18 00:58:40
I don't really get it, because I don't know anything about that magazine.  :unsure:

Consider yourself lucky that you don't know. Hint: Imagine the opposite of ha.org.
Title: Why 24bit/48kHz/96kHz/
Post by: Pio2001 on 2006-04-18 01:37:09
In the eyes of HA, how should this listening test ideally be set up ?

Thanks for your input.


Hello Kees de Visser,

If you want to run a very serious test about high definition digital formats, here are some things that I thought about.

Documentation.

You first have to learn about the similar tests that have already been done. Here is all that I have got in my links :

http://www.hydrogenaudio.org/forums/index....ndpost&p=372649 (http://www.hydrogenaudio.org/forums/index.php?s=&showtopic=40134&view=findpost&p=372649)

http://www.hydrogenaudio.org/forums/index....40&#entry374740 (http://www.hydrogenaudio.org/forums/index.php?showtopic=3390&st=75&p=374740&#entry374740)

The first thing to notice is that this kind of test have already been done, and that it is very difficult to get any success.

Basic training.

Start with very low definition material. You can get some samples of low definition audio here : http://ff123.net/samples.html (http://ff123.net/samples.html)
Usually, it is hard to distinguish between 15 bits and 16 bits of resolution, and between a lowpass of 15 khz and a lowpass of 16 kHz !

Choice of sonic material

However, a sample was found where a 16 kHz lowpass was easily audible. I didn't keep the link but I'm sure that someone here have got it. Thus the samples used for the test are very important. The listener can fail to distinguish between the formats just because the presented material is not critical.

For the bit depth, you need very dynamic material. The "rach" sample in the above page contains no silence. I think that a critical sample should. I'm not used to classical music, but in the few CD that I've got, there is the recording of Grieg's Peer Gynt directed by Neeme Jarvi, (Deutsche Gramophon). The RMS level of the first instrument of the track "I Dovregubbens Hall" is -60 dB, while the end of the track is clipping ! This recording have a resolution of 14 bits, and the quantisation noise is very audible at the beginning of the track.

For the sample rate, since only Oohashi claims to have got significant results, why not include similar instruments ? The "gamelan" that was used is a set of metallic percussions (metallophons). James Johnston ( http://www.skepticforum.com/viewtopic.php?...der=asc&start=0 (http://www.skepticforum.com/viewtopic.php?t=85&postdays=0&postorder=asc&start=0) ) cites this class of instruments (bells, glockenspiel...) as very sensitive to phase shift, and capable of producing hypersonic intermodulation directly in air, regardless of recording.

Scope

What is the question that this test is to answer ? Are you going to compare sample rates ? Bit depths ? DSD ? Analog vs digital ? Online or recorded on CD ? On tape ? Commerical formats ?...

Result analysis

The classical statistic evaluation (ABX) doesn't seem to fit this kind of tests. The NHK test, the Detmold tests, Julian Dunn et al's tests about pressed CD all make the same mistake. They evaluate the individual statistical significance, then say "this result is significant, this one is not". This is mathematical nonsense ! Statistical significance can only be applied to one hypothesis, not to individual results.
There are other methods (Anova, Tukey...) more suited to listening tests with many people, but I don't think that they work well for this kind of test, because you don't want to know if the group can hear a difference, but if one people at least in the group can hear a difference.

I've got a solution for this kind of hypothesis, but it might require quite a lot of trials for everyone. I'll got it together in the next days. I set it up for my interconnect blind listening test, but the only written description of it is in a  french forum, scattered in several posts, with some mistakes and corrections.

Basically, it just consists in letting people perform ABX tests (or similar tests), each one with its individual p value, and then compute the P value of the following event : "one people at least get an individual result equal or inferior to p by chance".
It has the advantage that listeners can communicate, and help each other, without affecting the significance of the final result. And it let the possibility for anyone to get a significant result, while usual statistical evaluations dismiss individual successes as non representative.

Protocol

The ABX protocol is not required in itself. You can choose A/B, AXY, XY, or anything you want, as long as it is randomized. If the randomization does not lead to a binary choice with equal probabilities, the individual p values will have to be recalculated from the right formula, since the one that is used for ABX won't be correct anymore. It is however correct for any protocol that leads to choose between two options of equal probabilities. For example "X=A or X=B", or "A was first or B was first", or also "X was the same as Y,or X was not the same as Y". Just be careful that the randomisation assign an equal probability for both answers.
The most secure way to get random numbers, in my opinion, is to use dices, with dice cups (like these : http://www.bgshop.com/ (http://www.bgshop.com/) )

If listeners pass the test individually, each one can even choose the protocol that he prefers, as long as you get his individual p.

Hardware

Are you going to use active bi amplification, like Oohashi and the NHK ? If not, how to evaluate the role of intermodulation ? According to David Griesinger ( http://world.std.com/~griesngr/intermod.ppt (http://world.std.com/~griesngr/intermod.ppt) ) hypersonic intermodulation mostly occurs in amplifiers, not in tweeters, while James Jhonston (link above) says that tweeters are very prone to hypersonic non-linear distortion. If you get hypersonic intermodulation, it will mean that the high definition format is inferior to the low definition one, since it adds distortion !
If you get active biamplification, how to evaluate the atmospheric hypersonic intermodulation ? If it is significant, then the test result will completely depend on the distance between the microphones and the instruments. Far from the instruments, no high definition required, since hypersonic intermodulation occurs in the performance room. Close to the instruments, and high definition is required in order to let hypersonic intermodulation occur in the listening room.
And can the difference between Griesingers' conclusion and Johnston's one come from the fact that the former tested harmonic intermodulation while the later speaks about transient non linear distortion ?



This is the scientific approach. It have become very hard because of the tests already set up. If you want to learn more, you will have to run better tests than what was already done !

However, you can also choose the debunking approach : find people who claim they can hear the night-and-day difference immediately without possible mistake. Let them use their equipment and usual listening conditions, and just add randomization. Be sure to allow the listener to give null answers, otherwise, in case of failure, he will argue that the stress confused his hearing. If he can answer nothing when he hears nothing, answering something will mean that he heard something, and not that stress prevented him to do so.
Title: Why 24bit/48kHz/96kHz/
Post by: crimsontide on 2006-04-18 12:59:49
Well, let's see. Atmospheric noise at the eardrum is 6dB SPL give or take.  96dB over that is 102dB. Even without noise shaping, that's pretty close to the loudest most speakers can go.


Erm, not entirely.

SPL Car installs regularly reach in excess of 150db of bass, and i am sure that my Missions could blow my eardrums if i whacked them right up - which is at least 119db

so your only 2/3 of the way towards 150db of dynamic range, and 15% short on the human ear at 119db.

Surely with all that optical storage  available it would be good sense to overkill the digital capabilities of the format to dispell this argument completely?

Sorry for being anal, I kind of agree with you as well, but I'm playing devils advocate.

Or.....have I misunderstood something?
Title: Why 24bit/48kHz/96kHz/
Post by: Kees de Visser on 2006-04-18 13:00:17
Dear Pio2001,

thanks for your elaborate post. This is the kind of constructive advice I was hoping to find here.
I'll need some time to digest all your info and do more reading about the subject. Work is very busy so please be patient

"this kind of test have already been done"
I know, and the results indicate that high(er) bitrates have no (or very little) benefits. This can mean that audible differences are negligible (or non-existant), or the test was flawed. If it turns out that we can't design a better test, it's useless to start one.

"Start with very low definition material."
That's exactly what we had in mind. Don't discourage people with impossible tasks.

"Choice of sonic material"
We're in the luxury position of having a large library of high-quality recordings and at the same time being able to record "live" music or sounds in the studio. Basically any instrument can be used. A choice of sources should be made and agreed upon before the actual listening test takes place.

"For the bit depth, you need very dynamic material."
That's available, ranging from the microphone/room self-noise to (just) acceptable clipping levels.

"only Oohashi claims to have got significant results"
I've been thinking about adding EEG (or even MEG) measurements but in practice this will probably be too complicated (MEG even impossible to install in a studio). Have EEG results, besides Oohashi, ever been used to prove audibility of stimuli ?

"What is the question that this test is to answer ?"
Good question!  Actually I'm not sure wether the test should "prove" or "investigate" something. As a recording engineer I'm interested in capturing a microphone's output with as little loss as possible. Enduser formats are a different matter, though related, so this test should focus on the recording and mixing formats.

"The classical statistic evaluation (ABX) doesn't seem to fit this kind of tests."
I'm glad to find this open-minded attitude on this forum and I'm very interested to hear more about your proposed evaluation method.

"Are you going to use active bi amplification ?"
The studio is equipped with a large selection of active and passive monitors. What isn't available can probably be arranged. The large active Genelec 1035 monitors are tri-amp models, designed for high spl (136dB peak) but without "super" tweeters. Of course there are other, more subtle monitors as well. We would have to examine which monitors perform best with respect to hypersonic non-linear distortion.
I'm in doubt wether it's better to test under laboratory conditions or to use a real-life music studio setup with equipment that's actually available.
Thanks for the links to the Griesinger and Johnston theories. I'll read them and see what consequences they have for the proposed test.

"This is the scientific approach. It have become very hard because of the tests already set up. If you want to learn more, you will have to run better tests than what was already done !"
Again: if we think this test won't add anything meaningful to the existing ones, we simply won't even start.

"the debunking approach : find people who claim they can hear the night-and-day difference immediately without possible mistake."
This might be a good warming-up test to prepare for the real test. Is it safe to say that if the warming-up test fails (no significant results) it's useless to continue ?

Kees
Title: Why 24bit/48kHz/96kHz/
Post by: Klyith on 2006-04-18 14:22:18
"only Oohashi claims to have got significant results"
I've been thinking about adding EEG (or even MEG) measurements but in practice this will probably be too complicated (MEG even impossible to install in a studio). Have EEG results, besides Oohashi, ever been used to prove audibility of stimuli ?

EEG might be able to distinguish between hearing a sound and not hearing a sound, but there's no way it could tell you anything about hearing or not hearing particular frequency components of sounds*. Even MEG would not be any help unless by some chance the proposed "high frequency response" area of the brain is in a different physical location of the brain than the rest of our hearing circuits. These devices actually provide a surprisingly low amount of real information, despite the pretty graphs and pictures, and are as easy to misinterpret by a non-expert as the subtle details of the sounds you are studying.

I would be very suspicious of studies using these instruments as "proof" that we can hear or not hear particular sounds. Most of all I would want to know the credentials of the person analyzing the results, and whether this was a qualified neuroscientist or not.

*Edit for clatity: that is, broad frequency spectrum sounds. Does it really matter, from a audio reproduction standpoint, is there is some sort of physiological response in the brain to isolated high-frequency sounds? Until it can be proved to have an effect on our conscious hearing of the music I would say it does not.

"The classical statistic evaluation (ABX) doesn't seem to fit this kind of tests."
I'm glad to find this open-minded attitude on this forum and I'm very interested to hear more about your proposed evaluation method.

I think that people here are not necessarily married to the ABX test in particular. But we do want to see a list of things common to all good science including: detailed explanation of the set-up and procedure, logical and well-thought-out methods, controlled tests, and not least repeatability. Plenty of scientific studies have been done double blind, but then later found to be invalid because of some overlooked flaw elsewhere in the experiment. It's easy to sink yourself in this game.
  As Fenyman said: "The first principle is that you must not fool yourself -- and you are the easiest person to fool."

...


Heh, more off topic: I saw that and was totally struck by how much the ufo looks like a 1970's speaker cone. I wonder if that's what it actually is, suspended by a bit of fishing line? Anyways, it inspired me:
(http://home.earthlink.net/~klyith/forums/shops/audiophiles_want_to_believe_t.jpg) (http://home.earthlink.net/~klyith/forums/shops/audiophiles_want_to_believe.jpg)
(link to slightly bigger version)
Title: Why 24bit/48kHz/96kHz/
Post by: Kees de Visser on 2006-04-18 16:34:43
EEG might be able to distinguish between hearing a sound and not hearing a sound, but there's no way it could tell you anything about hearing or not hearing particular frequency components of sounds*. Even MEG would not be any help unless by some chance the proposed "high frequency response" area of the brain is in a different physical location of the brain than the rest of our hearing circuits.

That's what I guessed, but I didn't want to dismiss any possibility to make the listening test as scientific as possible. My understanding of the brain and neuroscience is very limited but from what I understood MEG has a very high time-resolution (in the one digit ms region). Now imagine a stimulus containing sound (music?) with intermittent hypersonic content, on and off. If the hypersonic content has an influence on human perception, wouldn't there be a possibility that this would show on an MEG readout ? (I'm not even sure if loudspeakers/magnets are allowed in the MEG room) Ah well, you're probably right that it will turn into a sort of SETI project that hardly serves the purpose of the test.

Quote
Plenty of scientific studies have been done double blind, but then later found to be invalid because of some overlooked flaw elsewhere in the experiment.

Exactly, that's why it's a good idea to study the flaws from other tests and try to avoid them by carefully preparing the test with help of experts and open communities like HA.
Title: Why 24bit/48kHz/96kHz/
Post by: WmAx on 2006-04-18 17:23:13
SPL Car installs regularly reach in excess of 150db of bass, and i am sure that my Missions could blow my eardrums if i whacked them right up - which is at least 119db


The user was talking about typical high quality full range home speakers. Only extremely large line arrays or very large horns(for example, Avente Garde) can reach SPLs of 120+ dB at the listening position. CD should be good for about 120dB SPL in real situations[assuming the recording noisefloor is sufficient]; read my text further down in this post.

Quote
so your only 2/3 of the way towards 150db of dynamic range, and 15% short on the human ear at 119db.


No need for this range, short of wanting to reproduce a live band in a small room at close range in a very quiet room [requiring very special high output speakers] to full SPL range, which I can not see a use for in a practical sense. Such levels would be painful/undesirable to almost anyone. And the reproduction of a live band in a small room would probably require a 130dB SPL ability at listening position for some peaks.

Another consideration is the masking noise floor of the room itself. An average room, considering HVAC noise, traffic noise, fans, etc., 40dB is doing fairly good. If you live in a secluded area like in a rural area, away from a busy road,  turn off the HVAC, the neighbor is not cutting the grass, and any/all other noise making devices in the house are disabled, you might get into the 20dB range in this very rare scenario. Then there is the noise floor of the recording to consider. I have not yet come across a commercial recording [made from microphones, as opposed to purely synthetic] that had a noise floor hovering around the limit of the CD format itself[though I could see it happening if a digital noise reduction filter was used aggressively -- which would probably result in other problems], the noise is always substantially higher. Such may exist, but it must be very rare and limited to a special demo recording. The range would be such as to require the aforementioned high output speakers. In reality, the 96dB range seems to allow in the neighborhood of 120 dB of practical SPL, if considering the noise floor of real rooms. Beyond what almost any high quality speaker can provide. In a car, the noise floor will be far worse than the examples above, unless the car is stopped, in a quiet area[very little exterior noise] and with engine turned off.

-Chris
Title: Why 24bit/48kHz/96kHz/
Post by: crimsontide on 2006-04-18 17:39:03
figured id missed something - thanks
Title: Why 24bit/48kHz/96kHz/
Post by: krabapple on 2006-04-19 06:44:30
I don't really get it, because I don't know anything about that magazine. 


No insult intended.  But if you were to write an editorial there based on the idea that ABX testing
*itself* is the problem...you'd probably get a bagful of subscriber letters praising your insight,
(along with anecdotes how even their *wives* can hear the differences),  and possibly a promotion
to a corner office. 
Quote
Pio wrote:
The classical statistic evaluation (ABX) doesn't seem to fit this kind of tests. The NHK test, the Detmold tests, Julian Dunn et al's tests about pressed CD all make the same mistake. They evaluate the individual statistical significance, then say "this result is significant, this one is not". This is mathematical nonsense ! Statistical significance can only be applied to one hypothesis, not to individual results.


Can you expand on this?  Are you saying results from a test must always be analysed in aggregate of all subjects? How does this account for an 'exceptional' listener hearing a real difference , among a group of 'average' who fail to perform above chance?
Title: Why 24bit/48kHz/96kHz/
Post by: Klyith on 2006-04-19 07:28:15
Can you expand on this?  Are you saying results from a test must always be analysed in aggregate of all subjects? How does this account for an 'exceptional' listener hearing a real difference , among a group of 'average' who fail to perform above chance?

You can't do that, because there is no way to tell a priori who the exceptional listeners are. They could just be the ones who by random chance got the "right" answers. You can't do that kind of selection on your data after the fact, and if you do you can "proove" just about anything.

I do a test of 100 people for psychic powers by having them guess the number/suit of a randomly drawn card from a standard deck, and three get the right card. I then say that these three are psychic and the rest are normals, and I have proven the existance of psi. I am wrong, because three of 100 is not much outside the normal statistical chance of 1/52 for this experiment. The three people were just lucky. This is a simplified example, but your idea is wrong for the same reasons.
Title: Why 24bit/48kHz/96kHz/
Post by: hel96 on 2006-04-19 15:13:33
Why do we need 24bit/48kHz/96kHz/192kHz if 16bit/44.1kHz is good enough? Are there any situations that 16bit/44.1kHz simply cannot satisfy? In other words, is there any real need for the higher bit depth and sampling rate?


There are such situations if you record (or produce) your own music. Any new mix decreases the S/N ratio as well as it increases the group delay of filters. Even if you
stay in the digital domain, you can experience these effects with a wave editor by using the "mix paste" function. The final product, however, is o.k. with the cdda standard.

An undisputed benefit of the industry's efforts is that the hardware codecs now finally live up to the standard of "good old" cd audio.
Title: Why 24bit/48kHz/96kHz/
Post by: NeoRenegade on 2006-04-19 15:24:19
Quote
Why do we need 24bit/48kHz/96kHz/192kHz if 16bit/44.1kHz is good enough?
[a href="index.php?act=findpost&pid=353038"][{POST_SNAPBACK}][/a]

We dont need it. It's just virtual useless number-games to give people the incentive to buy new equipment and then re-buy all our music. There are some *technical* arguments for using 48khz instead of 44khz.... but the actual benefit for normal endusers is zero.
Purely listening purposes, you would be about right. Obviously there are purists who would like to listen to as high-resolution a recording as possible, whether or not the difference can be heard.

The strongest point of using 24bit/48kHz/96kHz/192kHz is for mixing and recording. Heavy editng of a 16-bit/44.1kHz recording will introduce audible artifacts. Not so with a higher bit-depth/sampling-rate.
Title: Why 24bit/48kHz/96kHz/
Post by: legg on 2006-04-19 16:06:20
I do a test of 100 people for psychic powers by having them guess the number/suit of a randomly drawn card from a standard deck, and three get the right card. I then say that these three are psychic and the rest are normals, and I have proven the existance of psi. I am wrong, because three of 100 is not much outside the normal statistical chance of 1/52 for this experiment. The three people were just lucky. This is a simplified example, but your idea is wrong for the same reasons.


Yes, but on several trials ABX testing (say over 16), getting all the right answers is unlikely. So I'd safely assume that one who did it, could have heard a difference. I wouldn't jump on conclusions yet, I'd rather make him/her take another set of tests. If the subject succeeds is because there's an audible difference, which could lead to the conclusion that at least 96/24 sounds different (and in theory better).

One verified pair of "golden ears" is all it takes for this silly debate to end.
Title: Why 24bit/48kHz/96kHz/
Post by: Axon on 2006-04-19 17:18:03
One verified pair of "golden ears" is all it takes for this silly debate to end.


Exactly one pair, more like. One out of twenty is meaningless.
Title: Why 24bit/48kHz/96kHz/
Post by: legg on 2006-04-19 17:45:27

One verified pair of "golden ears" is all it takes for this silly debate to end.


One out of twenty is meaningless.


In that case I'd just call 19 people and make them do the same tests as guruboolez and if they ever fail I should say that guruboolez' results are meaningless.

The probability of someone guessing 16 trials is 1/65536. If you have 20 subjects you increase this probability merely to 20/65536 (0.000305). If one succeeded there's no reason to believe that just because he was on a group his result should be discarded.
Title: Why 24bit/48kHz/96kHz/
Post by: krabapple on 2006-04-19 18:48:23

Can you expand on this?  Are you saying results from a test must always be analysed in aggregate of all subjects? How does this account for an 'exceptional' listener hearing a real difference , among a group of 'average' who fail to perform above chance?

You can't do that, because there is no way to tell a priori who the exceptional listeners are. They could just be the ones who by random chance got the "right" answers. You can't do that kind of selection on your data after the fact, and if you do you can "proove" just about anything.


You can retest the putative 'exceptional' person.  If they keep scoring well above chance, don't you think that means something?

Quote
I do a test of 100 people for psychic powers by having them guess the number/suit of a randomly drawn card from a standard deck, and three get the right card. I then say that these three are psychic and the rest are normals, and I have proven the existance of psi. I am wrong, because three of 100 is not much outside the normal statistical chance of 1/52 for this experiment. The three people were just lucky. This is a simplified example, but your idea is wrong for the same reasons.


Again, retesting should return these people to 'normal', if they were just lucky during *that test*.

And what if you are testing one or more persons who *already claim to have psychic powers*?  This is analogous to the normal situation in the audio hobby (if not in codec testing).  The 'sighted' portion of the test usually involves fiding if out the subject think they hear a difference in the first place.  If not, there's no reason to continue.
Title: Why 24bit/48kHz/96kHz/
Post by: Pio2001 on 2006-04-25 15:30:49
"this kind of test have already been done"
I know, and the results indicate that high(er) bitrates have no (or very little) benefits. This can mean that audible differences are negligible (or non-existant), or the test was flawed. If it turns out that we can't design a better test, it's useless to start one.


I think that using dynamic recordings in dedicated listening rooms, you may be able to show the benefit of bit depths superior to 16 bits.
You may also try to generate artificial signals that would show the theoretical audibility of a given parameter, even if you fail to find a musical recording that suffers from this parameter. For example I recently tried to ABX a phase shift at 30 Hz. I chose a recording with 30 Hz notes with sharp attacks. I failed. But I was told that the ideal signal for this test was a low frequency "saw-teeth" signal.
The main problem is that strong high frequencies can damage tweeters.

"only Oohashi claims to have got significant results"
I've been thinking about adding EEG (or even MEG) measurements but in practice this will probably be too complicated (MEG even impossible to install in a studio). Have EEG results, besides Oohashi, ever been used to prove audibility of stimuli ?


I don't know, but I was not talking about the EEG results, that are also questionable because the EEG excitation started one minute after the stimulus was presented, and also ceased quite a lot of time after the stimulus have been removed
I was talking about the subjective appreciation of the sound by the listeners. Table 2 gives an impressive set of significant p values associated with direct listening test, not through EEG. However, not a word about the way they were computed.

If you want to try the same experiment, that is asking people if what they hear sound "harsh, dynamic etc" instead of asking them to identify X, then we would have to setup a mathematical model in order to get the statistical significance of the answers.

"The classical statistic evaluation (ABX) doesn't seem to fit this kind of tests."
I'm glad to find this open-minded attitude on this forum and I'm very interested to hear more about your proposed evaluation method.


For this kind of test, where we want to see if a difference can be heard by some people under some conditions at least, I suggest a protocol divided into three parts :

In part 1, the listeners, that are supposed to be familiar with the kind of difference tested, are allowed to play with the system. They must find the hardawre and the musical samples on which the difference is the easiest to spot. This phase goes on until they think that the difference is obvious enough for a blind test to easily succeed.
In part 2, some fake blind tests are done. This is the training. Listeners try to recognize the difference under the real test conditions. They can compare ABX with other methods. They can choose what seems to be the best delay between the trials. This part ends when the listeners, or at least some of them, consistently get statistically good results. Remember that this is only training. These results won't be taken into account in the final conclusion, no matter what happens.
In part three, the real test is done, according to the protocol chosen in part 2. If the number of trials was decided in advance, listeners are told their score after each trial. If they begin to make some mistakes, they can interrupt part 3 in order to undergo some more training, or stop for a while. In part 3, they are allowed to give null answers when they are not sure. In ABX, it would be a three choice test : "X is A", or "X is B", or "I'm not completely sure".
Only the X is A or X is B answers are recorded. The part 3 goes on until the right amount of  these kind of answers is collected.

The advantage of dividing the test in three parts is to dismiss the usual arguments opposed to blind tests :
The listeners are deaf : dismissed by part 1
The system is not good enough for the difference to be heard : dismissed by part 1
Listening in ABX doesn't allow to spot these kind of differences : dismissed by part 2
A decision process cannot account for the unconcious influences at work : dismissed by part 2

If one listener decides to do an ABX test in 8 trials, here is an example of phase 3 :

Trial 1 : X is A : right
Trial 2 : X is A : right
Trial 3 : X is A : right
Trial 4 : I'm not completely sure
Trial 5 : I'm not completely sure

Pause

Trial 6 : X is A : wrong

Training

Trial 7 : X is B : right
Trial 8 : X is A : right
Trial 9 : I'm not completely sure

Pause

Trial 10 : I'm not completely sure
Trial 11 : X is A : wrong

This is the second error, the test has failed. Otherwise, it would have gone on until one more "X is A" or "X is B" answer would have been got, which would have totalized 8 answers of this kind.

If more than one listener is taking part, the required number of right answers must be mathematically decided. We must compute the probability for one listener to fail its own ABX test by chance. Then put it to the power N, when N is the number of listeners. It gives the probability that everyone fails. The complementary event is that one listener at least have succeeded.
This is our final statistical result : the probability that among all the listeners, one of them at least gets by chance the same or more than the highest individual score recorded.

All the listeners can pass the test together, if they want. Uncontrolled influences between them can only decrease the probability of this event, thus increase the statistical significance of the result.

Advantages of this kind of statistical evaluation over a classical one :
-Listeners who cannot hear the difference don't prevent listeners who can hear it from demonstrating that the difference is audible
-Listeners can communicate and help each other during the test. They don't need to pass it one by one.

Drawback :
-More trials are needed in order to reach an acceptable level of confidence.

It is very probable, in case of a difference that cannot be heard at all, that the test doesn't get past part 2. The listeners must then explain why the differences heard in part 1 have vanished in part 2, and possibly get back in part 1 in order to find a better way to pass part 2. It's up to them. They are the one hearing a difference, they are the one who can tell how the test must be done.

This protocol was discussed here, in french, during the setup of the interconnect blind listening test : http://www.homecinema-fr.com/forum/viewtop...r=asc&start=195 (http://www.homecinema-fr.com/forum/viewtopic.php?t=29770792&postdays=0&postorder=asc&start=195)
Title: Why 24bit/48kHz/96kHz/
Post by: Ihmemies on 2006-06-21 18:08:52
I resample to 24/96, because if I just played files at 16/44,1, Audigy would just make its own crappy resampling. So I do the resampling with software and add some extra just because I have a reasonably fast cpu.
Title: Why 24bit/48kHz/96kHz/
Post by: pariah123 on 2006-06-21 19:05:40
Yes, I have searched the forum.
Yes, maybe I am dumb.

But it seems I cannot find the answer.

Why do we need 24bit/48kHz/96kHz/192kHz if 16bit/44.1kHz is good enough? Are there any situations that 16bit/44.1kHz simply cannot satisfy? In other words, is there any real need for the higher bit depth and sampling rate?

Thanks for answering.


CD's are encoded in 44.1khz, and most soundcards upsample them to 48khz (because they cant play at 44.1), and this actually degrades quality... not improving.
Title: Why 24bit/48kHz/96kHz/
Post by: Crystaljuggler on 2006-06-22 18:19:42
It's entirely possible I'm posting on the far end of a thread like this because I'm new, but having read through all six pages, there's something that strikes me :

The prevailing opinion seems to be that an arbitrary format (16/44.1) is in itself surplus to requirements.  So the next format (24/96) is even more surplus to requirements, except in highly specialised environments like studios.  Whether it's better or not is questionable, but is it actually worse?  If not, then what's the problem?  Now, the tangent while I try and structure my thoughts :

I'm a PC modder, and I'll do all sorts of things to get my PC to run faster and cooler and with more lights in it.  I have benchmarks and tests that will show just how much faster my PC is than other people's.  Numbers!  Never mind that it's physically impossible to tell, from the user's point of view, whether you're going at 29.4 fps or 31.2 fps in a game.  But if I was to sell you a graphics card I'd gloss over that, give you the numbers, and you would decide for yourself that the one that goes faster is the better one.  Apart from cost, what real difference is it going to make to your life?  Only one : it will make you feel better.  Your experience of playing a game will be better.  It's subjective, but you'll have more fun knowing that you're not losing out to the limitations of your own optical nerves, which can only detect rates up to 25fps or thereabouts.  So : should progress of graphics technology be stopped, because it's "good enough"?  Of course not. 

Everything goes faster than it did, especially electronics.  Once you reach a certain point, you're not going to notice much difference, but it will be there.  You wouldn't notice much difference between a car trip in this year's Rolls Royce and last year's.  Given the choice, would you go for the old one?  Especially - and here's a fun bit - if it cost the same?  Technological advancements are very rarely driven by what the consumer wants, but what he can be told and convinced that he wants.  What he can be told to buy.  You convince enough people to buy, it becomes the standard and then boom, there's no difference to argue about any more and it all costs the same anyway.

We can, so we do.  I certainly couldn't tell the difference between 24/96 and 16/44.1, but that doesn't mean I'm going to be feeding my 24/96-capable amp with half of what it can chew on.  I'm also not going to run out and get DVD-A replacements for all my CDs.  Is the difference there?  Yes, mathematically and by the oscillations of the crystals, there is a difference.  Is it perceptible? No.  Is there a difference in the experience of owning and operating one of these things?  Yes.  And that's what matters.  Limited edition CDs sound better for just this reason.  That, for me, is an integral part of the listening experience, and if I wasn't reasonably scientifically-minded and sceptical already, I'd hate for someone to take that part of the experience away from me.
Title: Why 24bit/48kHz/96kHz/
Post by: stephanV on 2006-06-22 19:21:39
I'm sorry, an empty bank-account caused by all the surplus on equipment and media I have to purchase for something I can't hear won't make me feel better. And I mean this seriously, I don't want to spend money on illusions just to make me feel better. I don't want to spend money just because company X invented a format which brings me nothing more than things there already are.

Give the money to charity, it will make you feel better too and it at least might end up somewhere where it is needed.
Title: Why 24bit/48kHz/96kHz/
Post by: unfortunateson on 2006-06-22 19:38:42
the limitations of your own optical nerves, which can only detect rates up to 25fps or thereabouts.


This sounds quite nonsensical to me
Title: Why 24bit/48kHz/96kHz/
Post by: saratoga on 2006-06-22 20:29:19

the limitations of your own optical nerves, which can only detect rates up to 25fps or thereabouts.


This sounds quite nonsensical to me


Its not correct, but its not really important to the point he was getting at, so I don't think its important.
Title: Why 24bit/48kHz/96kHz/
Post by: legg on 2006-06-22 21:37:15
I'm a PC modder, and I'll do all sorts of things to get my PC to run faster and cooler and with more lights in it.  I have benchmarks and tests that will show just how much faster my PC is than other people's.  Numbers!  Never mind that it's physically impossible to tell, from the user's point of view, whether you're going at 29.4 fps or 31.2 fps in a game.  But if I was to sell you a graphics card I'd gloss over that, give you the numbers, and you would decide for yourself that the one that goes faster is the better one.  Apart from cost, what real difference is it going to make to your life?  Only one : it will make you feel better.  Your experience of playing a game will be better.  It's subjective, but you'll have more fun knowing that you're not losing out to the limitations of your own optical nerves, which can only detect rates up to 25fps or thereabouts.  So : should progress of graphics technology be stopped, because it's "good enough"?  Of course not.


On a PC it does matter how fast can it go or how much info it can process. Getting 200fps on X game might not be noticeable, but in the near future there will be a more demanding Y game that will get you roughly 20fps.

It is not the same with audio, indeed you can improve the sound card, but there's no point in improving it to the point where the difference is impercetible or negligible if you know that the future will not demand any more.

I don't see a problem with 96kHz/24-bit either, but it is much like killing a flea with a cannon.
Title: Why 24bit/48kHz/96kHz/
Post by: Cosmo on 2006-06-22 22:00:53
The words ''sound better'' should not be used to describe sensations that have nothing to do with hearing.
Title: Why 24bit/48kHz/96kHz/
Post by: HbG on 2006-06-23 12:26:09
24 FPS was chosen because it was the lowest number to provide smooth-enough moving images. When there is a lot of movement in a scene, it becomes very easy to notice the jerkyness. Especially in a cinema.

In fast-paced computer games, 25fps equals unplayable for me. I seem to need about 40 as a minimum. Let's do some math. A fast turn at high mouse sensitivity can easily be ~500 degrees per second. At 25fps this equals 20 degrees per frame, with a FOV of 100 this means a displacement of 1/5th of your screen. Your eyes will have a lot of trouble tracking a moving object at such a low granularity. Framerates much higher than those needed for smooth images can also serve a purpose as it reduces processing latency, which can make a difference when everyone is playing at their limits, like in cyberathletics.

I can tell the refresh rate of a monitor when it's displaying a light surface, up to about 85Hz. When servicing computers i try to guess the screen's refresh rate before checking it. I'm nearly always right. Not a double blind test but as good as it gets.  LCD's are excempt from this of course, but at the right angle in direct sunlight, i've seen the old style LCD screens of digital watches flicker also. Curiously enough i can also notice some plasma screens refresh like TV's. Spotting the 100HZ tv's in a tv store is pretty easy also.


Sound isn't video, however.
Title: Why 24bit/48kHz/96kHz/
Post by: crimsontide on 2006-06-23 13:32:56
While its a whole different argument, I agree with the above post - that while 24fps may be the cinemas moving image benchmark, its not the limit of visual perception by any means whatsoever.

In the same way that 128kbit mp3 is not "CD Quality" audio, however it is suitable for the general populous with mobile mp3 players.

EG:
I bought 2 albums on itunes, 128kbit AACs, and they sound terrible. The album in question is Less than Jake (trumpets sound ok, but the cymbals sound like gravel in my ears)

I also can tell (also used to be video technician for Digital Video Computing in sussex, UK) the difference between 50hz, and 75hz for example, and between 75 and 85 (gets harder), between 85 and 100 is difficult though - although I think i would get reasonable results in double blind tests.

But noone is saying that CD audio is DVD-A quality.

They are saying that noone can scientifically prove that they can tell a difference.

But thats like saying a gherkin is a good gherkin:
how many gherkins have you tried? what kind of vinegar are they pickled in? How long have they been pickled for? did you read the label before tasting and voting on it?

Frankly I dont give a damn - I know a good gherkin when i taste one, and i know a bad one too.

I think 96khz 24bit is a good place to stop for audio, because I don't see why we should have the accuracy/reproduction of an analogue source limited in any way. The mathematical proof is - that DVD-A is far better when mixed down from an analogue recording session.

infinite wave sampled to  CD Quality

is an exponentially poorer reproduction

than sampling @ DVD-a quality.


Enough of the abx blind tests.....they are subjective.....YOU are listening through your EARS.  Just liek I did - i couldnt tell the difference, but I say, when we have a lot of cannons, and cannonballs, lets kill fleas with em!!!!
Title: Why 24bit/48kHz/96kHz/
Post by: stephanV on 2006-06-23 13:41:29
You are not killing fleas, you are shooting holes in the air. And in your wallet.
Title: Why 24bit/48kHz/96kHz/
Post by: saratoga on 2006-06-23 14:53:37
I think 96khz 24bit is a good place to stop for audio, because I don't see why we should have the accuracy/reproduction of an analogue source limited in any way. The mathematical proof is - that DVD-A is far better when mixed down from an analogue recording session.

infinite wave sampled to  CD Quality

is an exponentially poorer reproduction

than sampling @ DVD-a quality.


Analog recordings don't have infinate resolution, if thats what you're trying to say. 

Even ignoring that, saying something is a "mathematical proof" is meaningless if you neglect to include the actual math!
Title: Why 24bit/48kHz/96kHz/
Post by: crimsontide on 2006-06-23 15:50:41
Analog recordings don't have infinate resolution, if thats what you're trying to say. 

Even ignoring that, saying something is a "mathematical proof" is meaningless if you neglect to include the actual math!


Alright they are limited by the accuracies of the voltage applied to the magnetic head during recording, and the quality/thicknes/width/age of the tape, which in turn is limited by the musical signal you are recording to the tape etc. etc. blah..... Jeez, there are literally billions of factors affecting it, but at least they also affect digital recording excluding the voltage/current accuracies applied to the recording head and the efficiency and response curve of the recording head.

I didnt do any maths to prove that 1 to the power of 16 is exponentially smaller dynamic resolution than 1 to the power of 24, now are you happy that ive wasted 2 minutes telling you what you already should know to be posting comment in this thread? Go look up "exponential" and prove it yourself.

poking holes in my post while not addressing the point is a classic strawman argument - congratulations.
Or are you suggesting that its not exponentially a poorer representation? No - you didnt do that - just poked holes to make it SEEM like my point held no water. I don't like that - its childish and totally unscientific. Go away.

ITs friday - im hot, at work, and i was just making a point without POINTLESSLY DELVING into UNNECESSARY DETAILS.
Title: Why 24bit/48kHz/96kHz/
Post by: stephanV on 2006-06-23 16:25:21
I didnt do any maths to prove that 1 to the power of 16 is exponentially smaller dynamic resolution than 1 to the power of 24, now are you happy that ive wasted 2 minutes telling you what you already should know to be posting comment in this thread? Go look up "exponential" and prove it yourself.

In my math book 1^16 = 1^24. 

I assume you mean 2^16 < 2^24.

But the question is if the extra resolution will bring you anything you can hear. And a higher dynamic range might help a little in very extra-ordinary cases. But a higher sampling rate won't. But if you have nothing better to spend your money on, go ahead.
Title: Why 24bit/48kHz/96kHz/
Post by: crimsontide on 2006-06-23 16:44:35

I didnt do any maths to prove that 1 to the power of 16 is exponentially smaller dynamic resolution than 1 to the power of 24, now are you happy that ive wasted 2 minutes telling you what you already should know to be posting comment in this thread? Go look up "exponential" and prove it yourself.

In my math book 1^16 = 1^24. 

I assume you mean 2^16 < 2^24.

But the question is if the extra resolution will bring you anything you can hear. And a higher dynamic range might help a little in very extra-ordinary cases. But a higher sampling rate won't. But if you have nothing better to spend your money on, go ahead.


pwned.

I told ya it was a friday!!!! and it really is hot in my office..... hahaha

I agree with you essentially, but my point is - why not? when we can make cd players that read dvd-a as well? I don't see.....my hard drive is now 320+160 gigabytes, almost a half terabyte. My first HD was 40 MEgabytes!!!

As long as the production takes advantage of it at every stage, then i think its worth doing as a last step to arguably transparent digital sampling.
Title: Why 24bit/48kHz/96kHz/
Post by: WmAx on 2006-07-14 04:31:21
...


This is the scientific discussion forum. You are posting speculation(s), which are useless in this context, and parts of your post where you make audible claims without acceptable perceptual tests to back them up, may actually be a violation of the TOS of this website.

-Chris
Title: Why 24bit/48kHz/96kHz/
Post by: Taz PA-C on 2006-07-30 19:30:19
The reason we need more than 16 bit audio is because the human ear can hear through the noise floor level.  This is easily proven by yourself.  Go to a crowded party where the noise level is very high.  If you couldn't hear below this noise level, a private conversation between two people would be inaudible over the noise.  Imagine if suddenly one of them mentions your name in a less than complimentery way.  Suddenly, all the noise in the room is now quiet, and you can hear every word they say, even though they were noise previously.  Human hearing is like that.  Now imagine that you are listening to a band with several voices, and in the background is a tune that is being played very softly, but that is what you want to hear.  If the bit level is deep enough, you will be able to tune it in, otherwise, it is just interference to the main sound level.  I wish we had 32 bit/64KHz stereo minimum for all recordings, that would do justice to any music.
This is my first post.  The people here seem very educated.  I am educated too, but I am not an engineer or sound technician.  This is my personal opinion, but I wish that the subject of sound reproduction wasn't so cold and clinical, too many assumptions are made that make cold clinical sense, but not sense if you factor in the human element.  The subject of not being able to hear below the noise floor is a perfect example.
Title: Why 24bit/48kHz/96kHz/
Post by: Lyx on 2006-07-30 19:57:04
The reason we need more than 16 bit audio is because the human ear can hear through the noise floor level.  This is easily proven by yourself.  Go to a crowded party where the noise level is very high.  If you couldn't hear below this noise level, a private conversation between two people would be inaudible over the noise.  Imagine if suddenly one of them mentions your name in a less than complimentery way.  Suddenly, all the noise in the room is now quiet, and you can hear every word they say, even though they were noise previously.

Your argument-chain is invalid, because you are mixing up *MASKING* and Noise-Floor. Both are by far not the same. The fact that we can change "how we filter information" has nothing to do with the noise-floor. This is an argument from ignorance.

- Lyx
Title: Why 24bit/48kHz/96kHz/
Post by: TBeck on 2006-07-30 19:57:27
The reason we need more than 16 bit audio is because the human ear can hear through the noise floor level.  This is easily proven by yourself.  Go to a crowded party where the noise level is very high.  If you couldn't hear below this noise level, a private conversation between two people would be inaudible over the noise.  Imagine if suddenly one of them mentions your name in a less than complimentery way.  Suddenly, all the noise in the room is now quiet, and you can hear every word they say, even though they were noise previously.  Human hearing is like that.  Now imagine that you are listening to a band with several voices, and in the background is a tune that is being played very softly, but that is what you want to hear.  If the bit level is deep enough, you will be able to tune it in, otherwise, it is just interference to the main sound level.  I wish we had 32 bit/64KHz stereo minimum for all recordings, that would do justice to any music.

I don't understand, why this should be an argument for the need of higher bit resolutions. Perception can be selective, that is true. The selection here can be made by special properties of the signal (Frequency, patterns, the meaning it has for you...). But why should you need a high bit resoulution or dynamics to perform this selection? This would only be true, if the dynamic range would be too small to let you discriminate the properties you need for your selection process. I am quite sure, that for your examples even far less than 16 bit would be sufficient to perform the selection.
Title: Why 24bit/48kHz/96kHz/
Post by: greynol on 2006-07-30 20:15:27
Let's take a cold clinical approach to the example you have presented here:
Go to a crowded party where the noise level is very high.  If you couldn't hear below this noise level, a private conversation between two people would be inaudible over the noise.  Imagine if suddenly one of them mentions your name in a less than complimentery way.  Suddenly, all the noise in the room is now quiet, and you can hear every word they say, even though they were noise previously.

The SPL of a crowded party that is loud is probably no more than 95dB, 105dB tops with loud music going.

The SPL of a "whisper" that can be heard between two people at such a party will be no less than 35dB.

The difference between these levels is at most going to be 70dB.  The SNR due to the quantization noise of 16 bits is 96dB, more than enough to capture that whisper at a noisy party.
Title: Why 24bit/48kHz/96kHz/
Post by: jhbretz on 2006-08-14 18:33:19
While you are right that the SNR of the quantization noise of a 16 bit signal is 96dB, keep in mind that we're talking about the *quality* of the quietest sound.  So if this quietest sound was at -96dB, then it would have no quality at all because it would be a square wave.

The real question should be "what bit resolution is necessary *FOR THE QUIESTEST SOUND*."  At 16 bits, the quiet sounds are audible, but they don't have the same detail as the louder sounds.

  Now check this out.  You can get 24 bit audio from 16 bit CDs - how?  Use ogg vorbis at a very high quality level.  The file now contains *frequency* domain information.  That means that if there is quantization, it will disappear with a decoder running at 24 bit accuracy.

You could still do a lot of other things wrong.  I use Shure E5C headphones driven with a custom Sigma delta converter and amp.  Things are real clean and real quiet.  The difference between 16 bit and 24 bit is just OBVIOUS.  Especially quiet sounds with some texture, like the echos after a background percussion hit.


The reason we need more than 16 bit audio is because the human ear can hear through the noise floor level.  This is easily proven by yourself.  Go to a crowded party where the noise level is very high.  If you couldn't hear below this noise level, a private conversation between two people would be inaudible over the noise.  Imagine if suddenly one of them mentions your name in a less than complimentery way.  Suddenly, all the noise in the room is now quiet, and you can hear every word they say, even though they were noise previously.  Human hearing is like that.  Now imagine that you are listening to a band with several voices, and in the background is a tune that is being played very softly, but that is what you want to hear.  If the bit level is deep enough, you will be able to tune it in, otherwise, it is just interference to the main sound level.  I wish we had 32 bit/64KHz stereo minimum for all recordings, that would do justice to any music.

I don't understand, why this should be an argument for the need of higher bit resolutions. Perception can be selective, that is true. The selection here can be made by special properties of the signal (Frequency, patterns, the meaning it has for you...). But why should you need a high bit resoulution or dynamics to perform this selection? This would only be true, if the dynamic range would be too small to let you discriminate the properties you need for your selection process. I am quite sure, that for your examples even far less than 16 bit would be sufficient to perform the selection.
Title: Why 24bit/48kHz/96kHz/
Post by: Axon on 2006-08-14 19:04:13
While you are right that the SNR of the quantization noise of a 16 bit signal is 96dB, keep in mind that we're talking about the *quality* of the quietest sound.  So if this quietest sound was at -96dB, then it would have no quality at all because it would be a square wave.

The real question should be "what bit resolution is necessary *FOR THE QUIESTEST SOUND*."  At 16 bits, the quiet sounds are audible, but they don't have the same detail as the louder sounds.

That is a strawman argument and, in fact, is not the question.

Clearly, if you want to encode a signal using only the least significant bit, you will have large amounts of distortion - after all, it's a 1-bit signal. But those signals simply do not exist in real life. The most important reason is that nobody would want to listen to actual music on it - there's no known listening environment, or set of listeners, that can tolerate more than 60db of dynamic range. There is also no popular or classical music with 90db of dynamic range.

I don't think you are able to come up with a valid use case where the quantization noise is actually important, unless you want to cause permanent hearing loss in the listener. If you wanted to record the Space Shuttle taking off, then sure, you'd probably need 24 bits of precision. But that has no bearing on the use of CDs to store actual music, or failing that, sounds that people really want to listen to.

Quote
Now check this out.  You can get 24 bit audio from 16 bit CDs - how?  Use ogg vorbis at a very high quality level.  The file now contains *frequency* domain information.  That means that if there is quantization, it will disappear with a decoder running at 24 bit accuracy.

You are confusing "accuracy" with "precision". Look it up. You will never get better than 16 bits of accuracy out of CD, even if you decode to 64-bit floating point, no matter which format you use.
Title: Why 24bit/48kHz/96kHz/
Post by: Woodinville on 2006-08-14 20:11:18
So if this quietest sound was at -96dB, then it would have no quality at all because it would be a square wave.


Please go to a good reference library and read Bart Locanthi Sr's paper on dithering.

Thank you.
Title: Why 24bit/48kHz/96kHz/
Post by: saratoga on 2006-08-14 21:06:11

So if this quietest sound was at -96dB, then it would have no quality at all because it would be a square wave.


Please go to a good reference library and read Bart Locanthi Sr's paper on dithering.

Thank you.


heh that was my thought.  You can do a lot with just one bit.  Certainly more then a "square wave".
Title: Why 24bit/48kHz/96kHz/
Post by: greynol on 2006-08-14 21:20:22
heh that was my thought.  You can do a lot with just one bit.  Certainly more then a "square wave".
I tried to point that out to someone here once but it didn't go over too well.

http://www.hydrogenaudio.org/forums/index....&pid=414672 (http://www.hydrogenaudio.org/forums/index.php?act=findpost&pid=414672)
Title: Why 24bit/48kHz/96kHz/
Post by: HotshotGG on 2006-08-14 23:21:12
Quote
Clearly, if you want to encode a signal using only the least significant bit, you will have large amounts of distortion - after all, it's a 1-bit signal. But those signals simply do not exist in real life. The most important reason is that nobody would want to listen to actual music on it - there's no known listening environment, or set of listeners, that can tolerate more than 60db of dynamic range. There is also no popular or classical music with 90db of dynamic range.


This is true under practical conditions. 
Title: Why 24bit/48kHz/96kHz/
Post by: Pio2001 on 2006-08-15 19:11:05
The real question should be "what bit resolution is necessary *FOR THE QUIESTEST SOUND*."


Answer : 1 bit, because the quietest audible sound by the ear has no quality either.

Now check this out.  You can get 24 bit audio from 16 bit CDs - how?  Use ogg vorbis at a very high quality level.  The file now contains *frequency* domain information.  That means that if there is quantization, it will disappear with a decoder running at 24 bit accuracy.


No. Quantization introduces noise. When you decode at 24 bits, you still have got the 16 bits noise introduced by the ADC process. It doesn't disappear. It is encoded and decoded.

The difference between 16 bit and 24 bit is just OBVIOUS.  Especially quiet sounds with some texture, like the echos after a background percussion hit.


Please, provide ABX results.
Title: Why 24bit/48kHz/96kHz/
Post by: Woodinville on 2006-08-15 19:32:03
Quote
Clearly, if you want to encode a signal using only the least significant bit, you will have large amounts of distortion - after all, it's a 1-bit signal. But those signals simply do not exist in real life. The most important reason is that nobody would want to listen to actual music on it - there's no known listening environment, or set of listeners, that can tolerate more than 60db of dynamic range. There is also no popular or classical music with 90db of dynamic range.


This is true under practical conditions. 



Really, if I encode a signal with a massively oversampled 1-bit signal, I'll get large amounts of distortion?

***cough***

Really?????

I use Shure E5C headphones driven with a custom Sigma delta converter and amp.  Things are real clean and real quiet.  The difference between 16 bit and 24 bit is just OBVIOUS.  Especially quiet sounds with some texture, like the echos after a background percussion hit.

(Emphasis added)

That's a pretty good 1-bit system, now, isn't it?
Title: Why 24bit/48kHz/96kHz/
Post by: Patsoe on 2006-08-15 19:49:29
[...]

Really, if I encode a signal with a massively oversampled 1-bit signal, I'll get large amounts of distortion?

***cough***

Really?????

[...] (Emphasis added)

That's a pretty good 1-bit system, now, isn't it?


I think Axon was talking about the LSB in a multi-bit system, which is not the same as a 'massively oversampled 1-bit signal' - it's not oversampled on the disc... Also the sigma/delta converter you're boldfacing is a dac: it's a decoder, so it can do nothing about your encoding...

Maybe I just don't get the joke?
Title: Why 24bit/48kHz/96kHz/
Post by: jhbretz on 2006-08-15 19:51:59
Okay, so you're saying that to hear the 1 bit sound, you would end up blasting your ears with everything else.

I agree.

Now let's ask "what bit resolution is necessary" for this quietest sound?  When I say "quiestest sound" I mean the quietest sound on the track of which you can still discern its quality.

If this "quietest sound" were 8 bits, would you be able to tell?  That means that it has %0.4 steps on it.  I would guess that this is probably around the threshold of hearing.

This quietest sound would only have to be 48dB down on the 16 bit recording to render it 8 bit precision.

I don't know the answer to this question, but the limit of perception is definately bigger than 1 bit, so the 96dB figure is strictly an upper limit.

While you are right that the SNR of the quantization noise of a 16 bit signal is 96dB, keep in mind that we're talking about the *quality* of the quietest sound.  So if this quietest sound was at -96dB, then it would have no quality at all because it would be a square wave.

The real question should be "what bit resolution is necessary *FOR THE QUIESTEST SOUND*."  At 16 bits, the quiet sounds are audible, but they don't have the same detail as the louder sounds.

That is a strawman argument and, in fact, is not the question.

Clearly, if you want to encode a signal using only the least significant bit, you will have large amounts of distortion - after all, it's a 1-bit signal. But those signals simply do not exist in real life. The most important reason is that nobody would want to listen to actual music on it - there's no known listening environment, or set of listeners, that can tolerate more than 60db of dynamic range. There is also no popular or classical music with 90db of dynamic range.

I don't think you are able to come up with a valid use case where the quantization noise is actually important, unless you want to cause permanent hearing loss in the listener. If you wanted to record the Space Shuttle taking off, then sure, you'd probably need 24 bits of precision. But that has no bearing on the use of CDs to store actual music, or failing that, sounds that people really want to listen to.

Quote
Now check this out.  You can get 24 bit audio from 16 bit CDs - how?  Use ogg vorbis at a very high quality level.  The file now contains *frequency* domain information.  That means that if there is quantization, it will disappear with a decoder running at 24 bit accuracy.

You are confusing "accuracy" with "precision". Look it up. You will never get better than 16 bits of accuracy out of CD, even if you decode to 64-bit floating point, no matter which format you use.
Title: Why 24bit/48kHz/96kHz/
Post by: jhbretz on 2006-08-15 20:04:58
I will definately look up the reference.  Thank you.

But the question of dithering is a side issue.  You can certainly use the LSB of any data stream and modulate it at a higher frequency to emulate a higher precision, provided that you oversample at a sufficient rate.

My main point was that on a 16 bit recording, the softer sounds are not 16 bit precision.  This begs the question of what precision is transparent for these softer sounds.


So if this quietest sound was at -96dB, then it would have no quality at all because it would be a square wave.


Please go to a good reference library and read Bart Locanthi Sr's paper on dithering.

Thank you.
Title: Why 24bit/48kHz/96kHz/
Post by: Patsoe on 2006-08-15 20:06:54
jhbretz, your post above is very cryptic... could you take a shot at rewriting that?

edit: oh - I meant the first of the two above here
Title: Why 24bit/48kHz/96kHz/
Post by: jhbretz on 2006-08-15 20:34:57
This quote of mine below intuitively sounds like nonsense, doesn't it?  (Maybe this is why I got "warned")  But hear me out - I was just posting this to see whether anyone was listening.  And someone is!  Very exciting!

Let's say for the purposes of discussion, the "softest sound" on a recording is at -60dB.  On a 16 bit recording, that means that it is only at 6 bit precision.  It has "steps" of 1.5% signal amplitude at the sampling frequency.  Now the act of encoding with vorbis fits this 6 bit signal with the RMS minimum error frequency components.  In doing so, an interpolated signal has been created that doesn't have the 1.5% steps in it.  (CT Fourier transform coeffiecients are discrete but the frequencies themselves are continuous)

The fact that sounds in general are very well correlated means that the interpolated signal is a good fit for what the signal should have been.

Now check this out.  You can get 24 bit audio from 16 bit CDs - how?  Use ogg vorbis at a very high quality level.  The file now contains *frequency* domain information.  That means that if there is quantization, it will disappear with a decoder running at 24 bit accuracy.
Title: Why 24bit/48kHz/96kHz/
Post by: Patsoe on 2006-08-15 21:02:35
This quote of mine below intuitively sounds like nonsense, doesn't it? (Maybe this is why I got "warned") But hear me out - I was just posting this to see whether anyone was listening. And someone is! Very exciting!

Let's say for the purposes of discussion, the "softest sound" on a recording is at -60dB.  On a 16 bit recording, that means that it is only at 6 bit precision.  It has "steps" of 1.5% signal amplitude at the sampling frequency.  Now the act of encoding with vorbis fits this 6 bit signal with the RMS minimum error frequency components.  In doing so, an interpolated signal has been created that doesn't have the 1.5% steps in it.  (CT Fourier transform coeffiecients are discrete but the frequencies themselves are continuous)

The fact that sounds in general are very well correlated means that the interpolated signal is a good fit for what the signal should have been.


Ah... I think I know what you mean by "steps of 1.5%" now. You mean the quantization error has a relative size of 1.5% on a signal of 2^6 amplitude, right?

Well, the bad news is: no Fourier transform is going to give you back the lost information. The error just takes another form in another domain.
Thus, without the transform, the quantized signal is just as good a fit for what it should have been. And since encoding to vorbis does a bit more than just transforming to the frequency domain, the fact is that the original wav is a better fit than the ogg.

If you got a warning, I think it wasn't for this erroneous thought, but rather for claiming you could obviously hear it...
Title: Why 24bit/48kHz/96kHz/
Post by: jhbretz on 2006-08-15 21:02:50
Please see previous post where I said:

Let's say for the purposes of discussion, the "softest sound" on a recording is at -60dB. On a 16 bit recording, that means that it is only at 6 bit precision. It has "steps" of 1.5% signal amplitude at the sampling frequency. Now the act of encoding with vorbis fits this 6 bit signal with the RMS minimum error frequency components. In doing so, an interpolated signal has been created that doesn't have the 1.5% steps in it. (CT Fourier transform coeffiecients are discrete but the frequencies themselves are continuous)

The fact that sounds in general are very well correlated means that the interpolated signal is a good fit for what the signal should have been.

  Now check this out.  You can get 24 bit audio from 16 bit CDs - how?  Use ogg vorbis at a very high quality level.  The file now contains *frequency* domain information.  That means that if there is quantization, it will disappear with a decoder running at 24 bit accuracy.


No. Quantization introduces noise. When you decode at 24 bits, you still have got the 16 bits noise introduced by the ADC process. It doesn't disappear. It is encoded and decoded.

The difference between 16 bit and 24 bit is just OBVIOUS.  Especially quiet sounds with some texture, like the echos after a background percussion hit.


Please, provide ABX results.


I just did an informal ABX and was able to discern a 16 bit wav from its vorbis encoded & 24 bit decoded equivalent consistently.  I did another test decoding the vorbis at 16 bit and 24 bits and got the same results.  How should I go about providing these results?
Title: Why 24bit/48kHz/96kHz/
Post by: saratoga on 2006-08-15 21:12:48
Please see previous post where I said:

Let's say for the purposes of discussion, the "softest sound" on a recording is at -60dB. On a 16 bit recording, that means that it is only at 6 bit precision. It has "steps" of 1.5% signal amplitude at the sampling frequency. Now the act of encoding with vorbis fits this 6 bit signal with the RMS minimum error frequency components. In doing so, an interpolated signal has been created that doesn't have the 1.5% steps in it. (CT Fourier transform coeffiecients are discrete but the frequencies themselves are continuous)

The fact that sounds in general are very well correlated means that the interpolated signal is a good fit for what the signal should have been.



  Now check this out.  You can get 24 bit audio from 16 bit CDs - how?  Use ogg vorbis at a very high quality level.  The file now contains *frequency* domain information.  That means that if there is quantization, it will disappear with a decoder running at 24 bit accuracy.


No. Quantization introduces noise. When you decode at 24 bits, you still have got the 16 bits noise introduced by the ADC process. It doesn't disappear. It is encoded and decoded.

The difference between 16 bit and 24 bit is just OBVIOUS.  Especially quiet sounds with some texture, like the echos after a background percussion hit.


Please, provide ABX results.


I just did an informal ABX and was able to discern a 16 bit wav from its vorbis encoded & 24 bit decoded equivalent consistently.  I did another test decoding the vorbis at 16 bit and 24 bits and got the same results.  How should I go about providing these results?


Describe the process you did to create the test, list the sample(s) you used, and post the score from your ABX test.
Title: Why 24bit/48kHz/96kHz/
Post by: Patsoe on 2006-08-15 21:20:33
I just did an informal ABX and was able to discern a 16 bit wav from its vorbis encoded & 24 bit decoded equivalent consistently.  I did another test decoding the vorbis at 16 bit and 24 bits and got the same results.  How should I go about providing these results?


You'll need to go from informal to formal

* what equipment were you using? (e.g. if you played back the 24bit file on a 16bit device it might sound different due to lack of dithering)

* were the levels the same after the encode/decode cycle? (you can check this in your wav-editor)

* how did you do the blinding (what software)?

* how many rounds did you do (without checking how you were doing!)?
Title: Why 24bit/48kHz/96kHz/
Post by: Woodinville on 2006-08-15 21:23:50
Now let's ask "what bit resolution is necessary" for this quietest sound?  When I say "quiestest sound" I mean the quietest sound on the track of which you can still discern its quality.


You are, I presume, aware of the absolute threshold of hearing?

I should, I suppose, also point out the fact that the atmosphere, being made of molecules of gasses (N2, O2, CO2, H2O primarily) and atoms of Argon, is discrete in nature, and that "air pressure" is in fact exactly the result of the momentum of the molecules and atoms bouncing off something.

This means that there is a noise level to the atmosphere.  For something the size of an eardrum, it's reasonably close to 6dB SPL, white noise, in the 20Hz to 20kHz range. This is a noise in the system that can not be removed, ever, unless you remove the air from both sides of the eardrum, which I believe may create some difficulties with the listener involved.

Standard understanding of noise-masking-noise shows that if you have noise in a critical band, and you wish to use to to mask some smaller noise, that an SNR of 3 to 4dB is sufficient to mask that noise.

That is what we have when we have quantization noise being masked by the atmosphere (which, by the way, is just SLIGHTLY below the threshold of hearing in the most sensitive ERB), with the quantization noise being at, oh, about 3dB SPL. That puts the peak level from a system operating in the quietest room on earth at 98dB SPL for peak levels.  For much presentation, this is completely sufficient.

Now, if we consider a quiet, normal room, what happens? We have at least 20dB more headroom. So, in other words, 16 bits is likely to be more than sufficient if we have normal speakers, which may get to 110dB on peaks without frying, in a QUIET normal room.

If we argue for 18 bits, we have enough dynamic range to get from the noise level of the atmosphere to the loudest most speakers can get, and certainly to the loudest that the human ear should ever actually be exposed to.

If we want to reproduce a rimshot, up close, first we have to invent new speakers. We should also, in that case, encourage research in to hair-cell regeneration on the Organ of Corti.

If, just for kicks, we argue for 32 bits, we can go from the atmospheric noise level (6dB) to +4dB re: 1 ATMOSPHERE RMS. Of course nobody can realize that level on the negative excursion side, and any one device is only likely to render it on the positive side for one use...
Title: Why 24bit/48kHz/96kHz/
Post by: jhbretz on 2006-08-15 21:54:59
I know where you are coming from.  You are saying "you can't possibly create new information" in decoding 16 bit data at 24 bits.  True - you can't, but the key is that sound is well correlated.  If we were talking about random noise, then it wouldn't work.  But sounds that come from real instruments have amplitudes that don't jump around a whole lot between samples.  So I would argue that doing an interpolation (and in fact much better than a linear interpolation) is valid.

In terms of "creating information," you aren't - you're just guessing that the source of the sound was correlated.  And since uncorrelated things sound awful, this is a good guess.

The frequency components of the actual sound are limited (real instruments have limited overtones) and therefore the vorbis encoder can store these limited number of frequencies and get very close to the original sound.


This quote of mine below intuitively sounds like nonsense, doesn't it? (Maybe this is why I got "warned") But hear me out - I was just posting this to see whether anyone was listening. And someone is! Very exciting!

Let's say for the purposes of discussion, the "softest sound" on a recording is at -60dB.  On a 16 bit recording, that means that it is only at 6 bit precision.  It has "steps" of 1.5% signal amplitude at the sampling frequency.  Now the act of encoding with vorbis fits this 6 bit signal with the RMS minimum error frequency components.  In doing so, an interpolated signal has been created that doesn't have the 1.5% steps in it.  (CT Fourier transform coeffiecients are discrete but the frequencies themselves are continuous)

The fact that sounds in general are very well correlated means that the interpolated signal is a good fit for what the signal should have been.


Ah... I think I know what you mean by "steps of 1.5%" now. You mean the quantization error has a relative size of 1.5% on a signal of 2^6 amplitude, right?

Well, the bad news is: no Fourier transform is going to give you back the lost information. The error just takes another form in another domain.
Thus, without the transform, the quantized signal is just as good a fit for what it should have been. And since encoding to vorbis does a bit more than just transforming to the frequency domain, the fact is that the original wav is a better fit than the ogg.

If you got a warning, I think it wasn't for this erroneous thought, but rather for claiming you could obviously hear it...
Title: Why 24bit/48kHz/96kHz/
Post by: Patsoe on 2006-08-15 22:28:17
I know where you are coming from.  You are saying "you can't possibly create new information" in decoding 16 bit data at 24 bits.  True - you can't, but the key is that sound is well correlated.  If we were talking about random noise, then it wouldn't work.  But sounds that come from real instruments have amplitudes that don't jump around a whole lot between samples.  So I would argue that doing an interpolation (and in fact much better than a linear interpolation) is valid.


Ofcourse the values can't jump around a lot between samples... that's what you ensure by low-passing the signal before a/d conversion.
What you want to accomplish - interpolating out the "steps" in the digital signal - is exactly what low-passing after d/a conversion does for you. And indeed that's better than a linear interpolation.

Seriously, your statement that the Fourier coefficients are quantized says it all. There is no improvement - the errors are now in those coefficients and will still sound the same.

I should, I suppose, also point out the fact that the atmosphere, being made of molecules of gasses (N2, O2, CO2, H2O primarily) and atoms of Argon, is discrete in nature, and that "air pressure" is in fact exactly the result of the momentum of the molecules and atoms bouncing off something.


Maybe you shouldn't have pointed that out
It serves no purpose at all for making your point (everything else in that post did not rely on the atomic structure of air), plus it seems to suggest that the discreteness of gasses would justify the discreteness in digital audio... that discreteness is on a completely different scale, and for all audio reproduction purposes on earth (not in outer space) we can consider air a continuous medium, I think
Title: Why 24bit/48kHz/96kHz/
Post by: Woodinville on 2006-08-15 22:43:46
Maybe you shouldn't have pointed that out
It serves no purpose at all for making your point (everything else in that post did not rely on the atomic structure of air),

Actually, it sets an absolute lower limit on what detectable noise from an audio system can be, and I did that, ergo it is rather germane to the discussion. Since that sets a peak level of 98dB or so for a 16 bit system that meets that constraint, as I pointed out in that post, that rather does tie in to most audio reprodction.
Quote
plus it seems to suggest that the discreteness of gasses would justify the discreteness in digital audio...

Perhaps it seems to do so to you, it certainly relates to discrete levels in audio for me, since it does show, quite indisputably, the lowest useful level, EVER, for actual reproduced quantization noise.  It does not assert that any particular form of discrete system is the right one, or that systems even must be discrete.

Since (we do remember our QM, yes?) all systems in the real world ARE discrete, your point seems rather pointless, as it were.
Quote
that discreteness is on a completely different scale, and for all audio reproduction purposes on earth (not in outer space) we can consider air a continuous medium, I think


Really, I don't think so. Just look at sound propagation through air as a function of moisture content. Right then and there, the issue rises its ugly head again.

As to "on a completely different scale", I have no idea what you mean. In fact, the threshold of hearing is very near the level of atmospheric noise at the eardrum, and very near the potential low-level noise of a 16 bit system set up for a reasonable playback level.  That is hardly "a completely different scale" it is very, very much on the same part of the scale, and very very nearly equal. So whatever DO you mean?
Title: Why 24bit/48kHz/96kHz/
Post by: Patsoe on 2006-08-15 23:37:02
So whatever DO you mean?


Hey, no need to bring out the all-caps

OK, I have reread your post - and think I get it now. The point is that thermal motion would account for 6dB SPL already, right? I read too fast the first time, and missed that connection  Also, now that I'm getting the message, I'm quite stunned by the fact that the absolute threshold of hearing would be this low.

Quote
Since (we do remember our QM, yes?) all systems in the real world ARE discrete, your point seems rather pointless, as it were.


I think I remember some QM  But what effects that we're discussing would be lost in a classical treatment?

My point is that quantizing the audio signal would be just as legitimate if the real world was not discrete in any way. So I was thinking that bringing that point up blurs the discussion (and we're only making it worse now ).

Different scale: I meant that the sheer number of atoms that touch your ear drum every "integration time" (of the sensory nerves, that is) would be such that there's no point discretizing that. Thinking some more about that and your point about thermal noise: can't we predict that 6dB thermal noise level within the framework of classical thermodynamics, without thinking of independent particles? edit: hmmm, I guess not... a classical gas pressure could just generate a constant force...
Title: Why 24bit/48kHz/96kHz/
Post by: bhoar on 2006-08-15 23:57:15
If this is off-topic, I apologize...

I assume the thread (currently, at least) is only addressing the "why 24bit/48kHz/96kHz" issue for playback only purposes.

There are several reasons that recording, mixing and processing should be done at higher resolution and/or sampling rates than the targeted playback format, especially for music sourced from large #s of individual tracks...

-brendan
Title: Why 24bit/48kHz/96kHz/
Post by: Woodinville on 2006-08-16 18:58:33

So whatever DO you mean?


Hey, no need to bring out the all-caps

OK, I have reread your post - and think I get it now. The point is that thermal motion would account for 6dB SPL already, right? I read too fast the first time, and missed that connection  Also, now that I'm getting the message, I'm quite stunned by the fact that the absolute threshold of hearing would be this low.


It's not clear that it is, but it's within 5 dB at the frequency of your ear canal resonance.  Needless to say it's hard to run this experiment.
Quote
I think I remember some QM  But what effects that we're discussing would be lost in a classical treatment?

Well, amplifiers, the atmosphere, etc, would be a whole lot quieter
Quote
My point is that quantizing the audio signal would be just as legitimate if the real world was not discrete in any way. So I was thinking that bringing that point up blurs the discussion (and we're only making it worse now ).

Well yes, I don't think that's in dispute, however, the quantized nature of the universe does set lower limits to perception and electronics, and those lower limits are NOT very much smaller than our perceptions.
Quote
Different scale: I meant that the sheer number of atoms that touch your ear drum every "integration time" (of the sensory nerves, that is) would be such that there's no point discretizing that. Thinking some more about that and your point about thermal noise: can't we predict that 6dB thermal noise level within the framework of classical thermodynamics, without thinking of independent particles? edit: hmmm, I guess not... a classical gas pressure could just generate a constant force...


Exactly. But the force isn't constant.

Just some numbers. 1 Atmosphere RMS is 194dB SPL. We can hear to about -6dB SPL at our ear canal resonance. That's a factor of 10^20 in energy, or 10^10 in amplitude. Yeah, we do hear really small variations in air pressure at audio frequencies. Really small.

Fortunately we don't hear low frequencies at all. Consider the implications of a change in barometric pressure caused by a cloud going over. That's nothing but low-frequency, nonlinear acoustics. Consider the SPL, too, strictly speaking. Yeah, milliHz, fortunately.
Title: Why 24bit/48kHz/96kHz/
Post by: jlt on 2006-08-19 17:45:27
Yes, I have searched the forum.
Yes, maybe I am dumb.

But it seems I cannot find the answer.

Why do we need 24bit/48kHz/96kHz/192kHz if 16bit/44.1kHz is good enough? Are there any situations that 16bit/44.1kHz simply cannot satisfy? In other words, is there any real need for the higher bit depth and sampling rate?

Thanks for answering.



If this is off-topic, I apologize...

I assume the thread (currently, at least) is only addressing the "why 24bit/48kHz/96kHz" issue for playback only purposes.

There are several reasons that recording, mixing and processing should be done at higher resolution and/or sampling rates than the targeted playback format, especially for music sourced from large #s of individual tracks...

-brendan


brendan,
you answered right(There are several reasons...) but William don't ask for playback only purposes.
for editions we need more than 16bit(dithering is horrible)
Title: Why 24bit/48kHz/96kHz/
Post by: Pio2001 on 2006-08-20 02:56:51
the key is that sound is well correlated.  If we were talking about random noise, then it wouldn't work.  But sounds that come from real instruments have amplitudes that don't jump around a whole lot between samples.  So I would argue that doing an interpolation (and in fact much better than a linear interpolation) is valid.


This has nothing to do with quantization. It deals with sample rate. When the amplitude doesn't jump around in both directions between samples, it means that there are no high frequencies, and that a low sample rate is enough.
When amplitude doesn't jump around much in a given direction, it means that the sound is quiet.

Since (we do remember our QM, yes?) all systems in the real world ARE discrete, your point seems rather pointless, as it were.


In QM, the cinetic energy of a free air molecule is not discrete, but continuous.
Title: Why 24bit/48kHz/96kHz/
Post by: Taz PA-C on 2006-08-20 04:06:01
Fascinating discussion.  Excellent question.  Check out the book by Malcolm Gladwell called "Blink".  There is report in this book of a double blind taste test conducted by Pepsi which proved that on the single sip, people much preferred Pepsi over Coke.  Coke then conducted their own double blind taste tests using the single sip taste test, and confirmed Pepsi's results.  As a result of these tests, Coke reformulated their recipe, and the rest is history.  In case you missed it, the new Coke failed miserably in the market.  Why?!?  Taste tests proved in double blind tests that the new formula was a winner.  The reason the new Coke failed was the very nature of the sip test, it was too brief a test.  It didn't reflect the days and weeks of using a product that on the initial test, won.  If buyers took a case of pop home, the results were completely different from a double blind test.  Coke eventually went back to "Classic" Coke, then the "New" Coke quietly dissappeared from the market.
Now lets apply the lesson from this to audio.  16 bits may be enough for a brief listen, but if you listen day after day, week after week, month, 16 bits aren't enough, I can hear too many limitations on my mid-fi system with 16 bit audio.  I can hear it when a 16 bit music CD is truncated as it fades to silence, and I hate it.  I want more.  To heck with double blind audio "sip" tests.  I want more.
Title: Why 24bit/48kHz/96kHz/
Post by: saratoga on 2006-08-20 04:56:45
16 bits may be enough for a brief listen, but if you listen day after day, week after week, month, 16 bits aren't enough, I can hear too many limitations on my mid-fi system with 16 bit audio.


Prove it.

I can hear it when a 16 bit music CD is truncated as it fades to silence, and I hate it.


There is no trunication.  Theres a noise floor, and the as the signal gets quieter it vanishes under quantinization error.  It seems to me you've fooled yourself into hearing things.
Title: Why 24bit/48kHz/96kHz/
Post by: Radetzky on 2006-08-20 05:11:16
Fascinating discussion.  Excellent question.  Check out the book by Malcolm Gladwell called "Blink".  There is report in this book of a double blind taste test conducted by Pepsi which proved that on the single sip, people much preferred Pepsi over Coke.  Coke then conducted their own double blind taste tests using the single sip taste test, and confirmed Pepsi's results.  As a result of these tests, Coke reformulated their recipe, and the rest is history.  In case you missed it, the new Coke failed miserably in the market.  Why?!?  Taste tests proved in double blind tests that the new formula was a winner.  The reason the new Coke failed was the very nature of the sip test, it was too brief a test.  It didn't reflect the days and weeks of using a product that on the initial test, won.  If buyers took a case of pop home, the results were completely different from a double blind test.  Coke eventually went back to "Classic" Coke, then the "New" Coke quietly dissappeared from the market.
Now lets apply the lesson from this to audio.  16 bits may be enough for a brief listen, but if you listen day after day, week after week, month, 16 bits aren't enough, I can hear too many limitations on my mid-fi system with 16 bit audio.  I can hear it when a 16 bit music CD is truncated as it fades to silence, and I hate it.  I want more.  To heck with double blind audio "sip" tests.  I want more.


What kind of analogy is that?

I don't even bother to explain anymore.  I guess reading posts about how a WAV sounds better than a FLAC just killed my enthusiasm... (head to head-fi.org if you need to laugh a bit...)

No wonder there is a market out there for all kind of "stereo-snake-oil" products... there is a world of suckers out there !!
Title: Why 24bit/48kHz/96kHz/
Post by: Patsoe on 2006-08-20 08:44:30

[...]
Now lets apply the lesson from this to audio.  16 bits may be enough for a brief listen, but if you listen day after day, week after week, month, 16 bits aren't enough,
[...]


What kind of analogy is that?

I don't even bother to explain anymore.  I guess reading posts about how a WAV sounds better than a FLAC just killed my enthusiasm... (head to head-fi.org if you need to laugh a bit...)

No wonder there is a market out there for all kind of "stereo-snake-oil" products... there is a world of suckers out there !!


What kind of a reaction is that?
Don't take this wrong, I know the feeling, but if you don't bother to explain, just don't post

Taz PA-C: nothing in abx protocols forbids you to listen for days and weeks...
Title: Why 24bit/48kHz/96kHz/
Post by: Pio2001 on 2006-08-20 13:19:50
Blind listening for days have even been done recently : http://www.hydrogenaudio.org/forums/index....showtopic=45432 (http://www.hydrogenaudio.org/forums/index.php?showtopic=45432)
The results ? People gets much worse results than listening for seconds. Psychological illusions seem to grow up and reinforce themselves day after day.

The failure of the new Coke taste can be explained by the fact that people have liked the old taste for years. It was a traditional product. Change the taste, for better or worse, and it is no more a traditional product. It is a new attempt from beginners, made by chemists and mathematicians instead of gastronoms, welcomed with suspucion.
That's now how a drink should be made, thus it has to taste bad. The psychological effect is at work, and makes people dislike the new taste, or not even try it.
Title: Why 24bit/48kHz/96kHz/
Post by: Kees de Visser on 2006-08-20 15:12:53
I can hear it when a 16 bit music CD is truncated as it fades to silence, and I hate it.

If you hear artifacts (grainy sound e.g.) when music fades to silence it probably means that either your monitoring system (DAC) or the audio on the cd (or both) hasn't been properly dithered. With correct dithering the sound will smoothly fade into the (dither-)noisefloor, which should be at such a low level (1LSB is about -90dBFS) that it's almost or completely inaudible under normal listening conditions.
Properly dithered music can still sound decent at very low levels (<-80dBFS).
Title: Why 24bit/48kHz/96kHz/
Post by: jlt on 2006-08-20 16:17:58
good explanations Kees de Visser 

the "issue" that i don't like in 16bit is that after each effect applyed(volume,equalize,etc) in the source when editing encrease the noise floor is summed one more time. 
then,after some effects you can hear clearly the "white noise"(hiss)....too bad.
Title: Why 24bit/48kHz/96kHz/
Post by: saratoga on 2006-08-20 17:57:54
good explanations Kees de Visser 

the "issue" that i don't like in 16bit is that after each effect applyed(volume,equalize,etc) in the source when editing encrease the noise floor is summed one more time. 
then,after some effects you can hear clearly the "white noise"(hiss)....too bad.


What software edits in 16 bit precision?  Even winamp plugins from the late 90s use floating point.
Title: Why 24bit/48kHz/96kHz/
Post by: Radetzky on 2006-08-20 22:05:42
Blind listening for days have even been done recently : http://www.hydrogenaudio.org/forums/index....showtopic=45432 (http://www.hydrogenaudio.org/forums/index.php?showtopic=45432)
The results ? People gets much worse results than listening for seconds. Psychological illusions seem to grow up and reinforce themselves day after day.


You are soooo wrong !  Philip Greenspun proves it! (http://philip.greenspun.com/materialism/stereo) (scroll to the "A/B Testing" section)

p.s.: nah.. I didn't think you were wrong.
Title: Why 24bit/48kHz/96kHz/
Post by: jlt on 2006-08-20 22:30:36
Quote
What software edits in 16 bit precision?

who knows in what software we can really trust?

Quote
Even winamp plugins from the late 90s use floating point.
yes but not means that winamp plugins are for advanced editions and good results,right?
Title: Why 24bit/48kHz/96kHz/
Post by: Pio2001 on 2006-08-20 23:24:58

Blind listening for days have even been done recently : http://www.hydrogenaudio.org/forums/index....showtopic=45432 (http://www.hydrogenaudio.org/forums/index.php?showtopic=45432)
The results ? People gets much worse results than listening for seconds. Psychological illusions seem to grow up and reinforce themselves day after day.


You are soooo wrong !  Philip Greenspun proves it! (http://philip.greenspun.com/materialism/stereo) (scroll to the "A/B Testing" section)


If the difference that they are talking about is psychologic, then it proves my point ! It gets stronger after a long time.
Title: Why 24bit/48kHz/96kHz/
Post by: Patsoe on 2006-08-21 07:36:55
You are soooo wrong !  Philip Greenspun proves it! (http://philip.greenspun.com/materialism/stereo) (scroll to the "A/B Testing" section)

p.s.: nah.. I didn't think you were wrong.


LOL... I loved the part where he says "If it is built in Japan, audio equipment is designed by engineers who couldn't get jobs designing video equipment. If it is built in the US, audio equipment is designed by engineers who couldn't get jobs designing high frequency electronics or computers."

Quote
What software edits in 16 bit precision?

who knows in what software we can really trust?

Quote
Even winamp plugins from the late 90s use floating point.
yes but not means that winamp plugins are for advanced editions and good results,right?


1. that sounds paranoid
2. you're right... I think that this is even one of the reasons foobar2000 was conceived: winamp plugins could work with floats internally but had to pass the stream from one plugin to the next in 16bit integers (but please correct me if I remembered wrong).
Title: Why 24bit/48kHz/96kHz/
Post by: cabbagerat on 2006-08-21 09:33:50
Quote
What software edits in 16 bit precision?

who knows in what software we can really trust?
Here's how you can learn to trust your software:But that would be taking it a bit far, wouldn't it? If you don't think the author of a piece of software is competent, then don't use it. Or use open source and audit the code yourself. You are just being silly now.

A couple months ago I tried to ABX 24 bit versus 16 and failed. Then I tried 16 versus 15 and passed on some samples (dynamic classical stuff, mostly) and 15 versus 14 and passed on most samples (dynamic classical, vocal and jazz). For me - 12bits seems good enough for most rock. Of course, anybody who is mixing or applying effects in 12bits isn't going to get good results.
Title: Why 24bit/48kHz/96kHz/
Post by: jlt on 2006-08-21 14:37:16
@ Patsoe
1- lol    but it's true,for audio i'm paranoic.
2- i can't correct you because i think you're right.
 

@ cabbagerat
Quote
But that would be taking it a bit far, wouldn't it? If you don't think the author of a piece of software is competent, then don't use it. Or use open source and audit the code yourself. You are just being silly now.
humm...your answer is a little...hard!    we (well,i am) are trying to answer the first question of the thread and not to be "correct" or wise.

Quote
Buy expensive distortion analyzer from Agilent
Generate a variety of sweeps, square waves, etc
Measure the distortion
A couple months ago I tried to ABX 24 bit versus 16 and failed.

failed? ok but don't need ABX test for this conditions:
Quote
the "issue" that i don't like in 16bit is that after each effect applyed(volume,equalize,etc) in the source when editing encrease the noise floor is summed one more time.

the noise floor is summed after each effect and is audible...and this is why we need more than 16bit for editions(here again i'm trying to answer the first post.)
do one little test and you'll listen the big hiss.
this is one fact and not one doubt(and don't deserve or need ABX test)
Title: Why 24bit/48kHz/96kHz/
Post by: puntloos on 2006-08-21 15:29:13
While I might be totally redundant here in this discussion (I have skipped over some technicals), there is one 'fact' that has become important for me:

1/ Upsampling from 16 bit to 24 bit is lossless (mind you: I mean 16/44 to 24/44, for example)
2/ Digitally reducing volume of a 16bit recording to say '33.333333% volume' will result in a greater accuracy error than reducing the upsampled 24bit version.
3/ The same applies for all other 'conversion/mogrification' steps.

Yes, it would be futile to upsample -> reduce volume -> dither back to 16, since a SMART volume control will do the dithering anyway, but this will still be (more) lossy (than necessary). You will get the improvement if you play back the new 24bit version with 24bit DA convertors.

Or am I missing some key points?
Title: Why 24bit/48kHz/96kHz/
Post by: jlt on 2006-08-21 17:41:33
Quote
Or am I missing some key points?
maybe.look the bar levels,the noise floor and the "dither" in the screenshots.

01 source 44.1k-16bit have ~ -84.3dB
[a href="http://img242.imageshack.us/my.php?image=0116bitfh5.png" target="_blank"]
Title: Why 24bit/48kHz/96kHz/
Post by: markanini on 2006-08-22 14:46:24
Recording and editing in 24 bit makes sense. After dithering 16 bits is enough tho.
Title: Why 24bit/48kHz/96kHz/
Post by: dex Otaku on 2006-08-26 09:26:28
There's a really, really simple answer to this question.  It has nothin to do with equipment being capable of playing back with 144dB of dynamic range or >48kHz bandwidth, either.

Higher resolution is higher resolution.  Period.

It's almost irrelevant that we can't hear above ~20kHz.  It's also next to irrelevant that we can't perceive even the 96dB of dynamic range that 16 bits provide, or that the average listening environment has a high enough ambient noisefloor that attempting to play back with a full 96dB of dynamic range would probably blow all the windows out and make our ears bleed, for that matter.

The real answer is this: higher bit depths and sampling rates mean higher accuracy within the normal audio band, regardless of the dynamic range, SNR, or actual transduced bandwidth of the signal. 

Higher accuracy = higher accuracy, period. 


From where I'm standing, the point behind high resolution audio formats isn't to annoy your dog's ears or to attempt to reproduce a higher dynamic range than we can perceive.  Actually attempting to reproduce the full 96kHz bandwidth of a 192kHz-sampled recording is beside the point; the point is that the audible bandwidth is >4x the resolution of a 44.1kHz recording.  Likewise, a higher bit depth means higher amplitude resolution, plain and simple. 

This is all leaving out the fact that most microphones [even expensive professional equipment] can't transduce the full bandwidth to be recorded by even 48kHz equipment. 

If we have 24-bit ADCs and DACs that can sample at 192kHz, cheap mass-storage solutions for recording and editing with, broadband delivery systems, and mass-produceable storage media capable of holding hours of high-resolution audio - then why not use it? 

It's not a point of novelty.  We have the capability to do this, so why not use it?  It doesn't even matter than your playblack equipment rolls off above 18kHz.  [Though it can be argued that >30kHz signals can damage some equipment.]


All that said - yes, it's true that the vast majority of people [myself included, and I'm a recordist/engineer] can't tell the difference between a 24-bit 96kHz recording and a properly mastered 16-bit 44.1kHz one.  I, myself, record in 16-bit [the limitation of the medium I use], edit in 32-bit fp, store edit masters in 24-bit [the highest resolution any playback equipment can currently use directly anyway], and distribute in 16-bit or lossy-compressed forms. 

It does make a difference to record and edit with the highest resolution possible, then distribute with what is common and works well - CDDA exceeds the needs of most people by quite a ways. 

That doesn't mean we should limit ourselves to that format, though.
Title: Why 24bit/48kHz/96kHz/
Post by: jlt on 2006-08-26 11:01:51
  wonderful explanations dex Otaku,congrats.

i'm a single home user and i can hear the noise floor with round 85/90dB in low parts of the musics(when the sound is loud) and it bore me.

as a recordist/engineer and have good equipment, do you agree that the hiss in the noise floor encrease if any effect is used editing 16bit?

thanks.
Title: Why 24bit/48kHz/96kHz/
Post by: cabbagerat on 2006-08-26 11:26:38
The real answer is this: higher bit depths and sampling rates mean higher accuracy within the normal audio band, regardless of the dynamic range, SNR, or actual transduced bandwidth of the signal.
No. They dont.
Let's start with sampling rate. If we make the assumption that sampling is done with an impulse and that filters are perfect, sampling at a frequency of 2N Hz will *perfectly* represent a signal with a bandwidth of N Hz. This is the Nyquist (sometimes called Nyquist Shannon and other things) sampling theorem. If you managed to show otherwise, you would rock the world of mathematics.

If we relax the requirement on the filtering and the shape of the sampling pulse, the picture changes somewhat, with the bandwidth of the signal that can be reconstructed reducing from N (for a 2N Hz sampling rate) to some number smaller than N. On audio DACs and ADCs, the bandwidth is commonly better than 90% of the Nyquist frequency limit.

On to sample bit depth. Assume for a moment that you take an arbitrarily accurate sample of the incoming wave form. This sample will consist of some amount of signal (S) and some amount of noise (N). The ratio between signal and noise is limited by the temperature of the components involved (a discussion of this would be fairly technical, but can be found in may good physics texts). When we reduce the bit depth of the samples (quantization) we increase the strength of the Noise signal - effectively adding white noise of a certain amplitude to the signal. Noise shaping and certain other trickery can be used to change the PDF and frequency distribution of this noise, but they will not effect the total energy.

So, a 44100 Hz sampling rate and 16bit samples can represent any signal with a bandwidth of 22050Hz to a signal to noise ratio of ~96dB. Sorry, but assertions like this just without any proof or evidence just pisses me off.

None of this is new - it's been discussed on this thread for the last nine pages.
Title: Why 24bit/48kHz/96kHz/
Post by: stephanV on 2006-08-26 11:53:51
Quote
then why not use it?

I think a repetition (http://www.hydrogenaudio.org/forums/index.php?showtopic=40134&st=148#) of all arguments used in this thread is not very useful.
Title: Why 24bit/48kHz/96kHz/
Post by: Pio2001 on 2006-08-26 18:13:40
The frequency resolution of a 44100 Hz or any other sample rate is infinite within its frequency range. Sampling at 96 or 192 kHz doesn't make it more accurate.
Title: Why 24bit/48kHz/96kHz/
Post by: bigshot on 2006-08-27 03:40:55
I, myself, record in 16-bit [the limitation of the medium I use], edit in 32-bit fp, store edit masters in 24-bit [the highest resolution any playback equipment can currently use directly anyway], and distribute in 16-bit or lossy-compressed forms.


Recording at 16 bit and bumping it up to high bitrate to edit is counter-productive, unless you are doing heavy duty processing of the sound. I don't know why you wouldn't record at a high bitrate if you are going to mix at a high bitrate. The only times that 24 bit sound has come in handy for me when I mix is bringing up low level sound. 24 bit has higher resolution at lower volume levels allowing more flexibility in mixing.

See ya
Steve
Title: Why 24bit/48kHz/96kHz/
Post by: Radetzky on 2006-08-27 15:58:59
I, myself, record in 16-bit [the limitation of the medium I use], edit in 32-bit fp, store edit masters in 24-bit [the highest resolution any playback equipment can currently use directly anyway], and distribute in 16-bit or lossy-compressed forms.


Recording at 16 bit and bumping it up to high bitrate to edit is counter-productive, unless you are doing heavy duty processing of the sound. I don't know why you wouldn't record at a high bitrate if you are going to mix at a high bitrate. The only times that 24 bit sound has come in handy for me when I mix is bringing up low level sound. 24 bit has higher resolution at lower volume levels allowing more flexibility in mixing.

See ya
Steve


Yeah, I too thought it was strange to record in 16-bits and THEN work in 32-bits.  Just record in whatever highest resolution you can to work with the audio.  Then, when all is done, just mix down the 16/44.1.

And, dex Otaku, I believe you have SOME of your understandings correct.  It IS better to record audio and work with audio in the highest resolution possible.  But, I think you misinterpret WHY is it better to do so.  Hint: It has not much to do with the final version you produce.

I don't say so because I am insecure... 16/44.1 IS enough for stereo reproduction.  Saying otherwise is stating you are badly informed.  But now, I am just being a jerk...
Title: Why 24bit/48kHz/96kHz/
Post by: jlt on 2006-08-27 16:53:44
in all my tests from years editing audio as home user/eletronic technician and from all cool posts and comments in this thread,here are 2 phrases resuming what is correct and answer the first post:

to record and/or edition:
Quote
Just record in whatever highest resolution you can to work with the audio. Then, when all is done, just mix down the 16/44.1.


to listen:
Quote
16/44.1 IS enough for stereo reproduction.


Title: Why 24bit/48kHz/96kHz/
Post by: saratoga on 2006-08-27 17:58:48

I, myself, record in 16-bit [the limitation of the medium I use], edit in 32-bit fp, store edit masters in 24-bit [the highest resolution any playback equipment can currently use directly anyway], and distribute in 16-bit or lossy-compressed forms.


Recording at 16 bit and bumping it up to high bitrate to edit is counter-productive, unless you are doing heavy duty processing of the sound. I don't know why you wouldn't record at a high bitrate if you are going to mix at a high bitrate. The only times that 24 bit sound has come in handy for me when I mix is bringing up low level sound. 24 bit has higher resolution at lower volume levels allowing more flexibility in mixing.

See ya
Steve


Yeah, I too thought it was strange to record in 16-bits and THEN work in 32-bits.  Just record in whatever highest resolution you can to work with the audio.  Then, when all is done, just mix down the 16/44.1.

And, dex Otaku, I believe you have SOME of your understandings correct.  It IS better to record audio and work with audio in the highest resolution possible.  But, I think you misinterpret WHY is it better to do so.  Hint: It has not much to do with the final version you produce.

I don't say so because I am insecure... 16/44.1 IS enough for stereo reproduction.  Saying otherwise is stating you are badly informed.  But now, I am just being a jerk...


Recording in 16 bit isn't going to automatically give you bad results, but it is going to leave you a lot less headroom to get the volume levels right.  However, if you have the levels right, the limitation won't be the bitdepth, but rather the equipment you use.  Pretty big if though

Kind of a funny example:  I was saw an experiement that recorded audio data, and needed about 8 bit precision at most to get good results.  The team used a 24 bit ADC and no amplifiers at all on any of the equipment.  Waste of a good ADC, but it worked
Title: Why 24bit/48kHz/96kHz/
Post by: Woodinville on 2006-08-28 23:58:33
The failure of the new Coke taste can be explained by the fact that people have liked the old taste for years. It was a traditional product. Change the taste, for better or worse, and it is no more a traditional product. It is a new attempt from beginners, made by chemists and mathematicians instead of gastronoms, welcomed with suspucion.
That's now how a drink should be made, thus it has to taste bad. The psychological effect is at work, and makes people dislike the new taste, or not even try it.


There's another issue, in that taste and smell receptors can be saturated, and can be blocked for quite a while by various tastes and smells.

The history of the olfactory system can run to at least 10's of minutes.

The history of the hearing apparatus barely makes it to 200 milliseconds, using the same meaning for "history", as in "where chemical differences remain", unless you've hurt your ears by listening too loud.
Title: Why 24bit/48kHz/96kHz/
Post by: dvda-sacd on 2006-09-05 11:20:04
DSD and PCM 24/192 increase the resolution of music by more closely following the original waveform of the music.
A picture is worth a thousand words:

Regards.
Title: Why 24bit/48kHz/96kHz/
Post by: Firon on 2006-09-05 11:43:16
It'd be nice if you read the damn thread before posting garbage.
Title: Why 24bit/48kHz/96kHz/
Post by: 2Bdecided on 2006-09-05 11:47:11
It'd be nice if you read the damn thread before posting garbage.


LOL! So true!

Or as we used to say "Harsh... but fair!"

Cheers,
David.
Title: Why 24bit/48kHz/96kHz/
Post by: Hollunder on 2006-09-05 12:16:48
After looking at his other posts I would say he's simply violating T.O.S. 14 (http://www.hydrogenaudio.org/forums/index.php?s=&showtopic=3974&view=findpost&p=292534)
Title: Why 24bit/48kHz/96kHz/
Post by: smz on 2006-09-05 12:31:22
After looking at his other posts I would say he's simply violating T.O.S. 14 (http://www.hydrogenaudio.org/forums/index.php?s=&showtopic=3974&view=findpost&p=292534)


And you know what? I tried to give a look at his other posts and ZoneAlarm (my antivrus/antimalware/firewall/etc.) blocks access to his site webcindario.com as a "Spy site"
(There is a link to an image hosted in that domain in the post at http://www.hydrogenaudio.org/forums/index....howtopic=35839) (http://www.hydrogenaudio.org/forums/index.php?showtopic=35839))

Sergio
Title: Why 24bit/48kHz/96kHz/
Post by: dvda-sacd on 2006-09-05 13:35:14
It'd be nice if you read the damn thread before posting garbage.

Sorry, I don't think I have posted garbage. I tryed to show why some people (like me) do want high resolution digital audio formats.

With respect to this link (http://dvda-sacd.webcindario.com/), I must point out that it's a non commercial free website. Webcindario (http://webcindario.miarroba.com/) offers a free web hosting service, in exange for including publicity in your web pages; therefore you may find pop-ups and banners in that web site.

Regards.
Title: Why 24bit/48kHz/96kHz/
Post by: smz on 2006-09-05 13:40:37
With respect to this link (http://dvda-sacd.webcindario.com/), I must point out that it's a non commercial free website. Webcindario (http://webcindario.miarroba.com/) offers a free web hosting service, in exange for including publicity in your web pages; therefore you may find pop-ups and banners in that web site.

Regards.

So how do you explain that ZoneAlarm lists your domain as a "Spy Site"?

EDIT: Maybe now I understand. webcindario is *not* your domain but a hosting service on which you have hosted your pages. If this is the case, my best advice is to move your pages somewhere else because your host is blacklisted at least by ZoneAlarm (a very well known and diffused firewall) and those using it will not get to your pages. As for those that do not use it and get to your site... maybe they will be get caught by some malware and they will give you all their blessings!

Sergio
Title: Why 24bit/48kHz/96kHz/
Post by: Hollunder on 2006-09-05 13:43:34
well, you should read the T.O.S. anyways:

8. All members that put forth a statement concerning subjective sound quality, must -- to the best of their ability -- provide objective support for their claims. Acceptable means of support are double blind listening tests (ABX or ABC/HR) demonstrating that the member can discern a difference perceptually, together with a test sample to allow others to reproduce their findings. Graphs, non-blind listening tests, waveform difference comparisons, and so on, are not acceptable means of providing support.
Title: Why 24bit/48kHz/96kHz/
Post by: dvda-sacd on 2006-09-05 13:49:26
So how do you explain that ZoneAlarm lists your domain as a "Spy Site"?

Make that question to Miarroba Networks, S.L. (http://miarroba.com/servicios/contacta.php), that is the owner of Webcindario.

well, you should read the T.O.S. anyways:

8. All members that put forth a statement concerning subjective sound quality, must -- to the best of their ability -- provide objective support for their claims. Acceptable means of support are double blind listening tests (ABX or ABC/HR) demonstrating that the member can discern a difference perceptually, together with a test sample to allow others to reproduce their findings. Graphs, non-blind listening tests, waveform difference comparisons, and so on, are not acceptable means of providing support.

OK, but I haven't done a statement concerning subjective sound quality, but objective sound quality. If that type of statements lacks of interest in this forum I apologize.
Title: Why 24bit/48kHz/96kHz/
Post by: smz on 2006-09-05 13:52:35

So how do you explain that ZoneAlarm lists your domain as a "Spy Site"?

Make that question to Miarroba Networks, S.L. (http://miarroba.com/servicios/contacta.php), that is the owner of Webcindario.

Sorry, I was editing my post while you were answering. Understood. My considerations (in the "Edit") still holds, anyway.

Cheers

Sergio
Title: Why 24bit/48kHz/96kHz/
Post by: dvda-sacd on 2006-09-05 13:57:50


So how do you explain that ZoneAlarm lists your domain as a "Spy Site"?

Make that question to Miarroba Networks, S.L. (http://miarroba.com/servicios/contacta.php), that is the owner of Webcindario.

Sorry, I was editing my post while you were answering. Understood. My considerations (in the "Edit") still holds, anyway.

Cheers

Sergio

OK, thank you.

Cheers!
Title: Why 24bit/48kHz/96kHz/
Post by: 2Bdecided on 2006-09-05 14:16:02
DSD and PCM 24/192 increase the resolution of music by more closely following the original waveform of the music.
A picture is worth a thousand words:

With DSD, you can have a filter on the DAC, or not. Some consumer devices offer a choice between filtering at 50kHz or 100kHz.

When capturing scope screen shots, you have the choice to average (few or many times), or not at all.

Before we discuss them in detail, you must understand that no single filter and scope setting could produce all of the pictures you've posted. For each picture, the filter (or not) and averaging (or not) has been chosen to make DSD "look good". Whether this is a reasonable thing to do (or not!) requires at least some understanding - the graphs you have posted suggest little or no understanding, and they're hardly new!

Quote
  • Impulse response of an analog, a PCM and a DSD system -> http://www.merging.com/2002/images/dsdresponseneon.gif (http://www.merging.com/2002/images/dsdresponseneon.gif)

And what is the impulse response of the human ear? It's non linear, asymmetric, and more widely spread than the widest of those.

Given that we are targeting the human ear, this means there is a level of accuracy beyond which it is pointless to strive. Things may be objectively better, but they may also be pointless!

For example, we can easily make systems with a frequency response flat up to 130MHz (which means an impulse response with a rise time of 0.004us - a little better than the "best" graphs on your link). I could make a similar graph to show how such a system is vastly superior to DSD. Does it sound better? Of course not - all the extra information is irrelevant to the human ear! (But is very useful for clear sharp HD video signals).

As for "the impulse response of analogue" - analogue what? What they've shown isn't tape or vinyl, that's for sure. Maybe it's an analogue "piece of wire"(!).


Finally, to get a clean plot like that, the DSD response must have been averaged or carefully filtered. DSD itself has so much intrinsic noise, that you can't even see the impulse response without these measures. So even as a claim of objective superiority, these plots are flawed.

(It is reasonable to correctly filter the signals, so I'm not arguing that the plot is misleading in that sense - just that, as we move through these plots, we'll get to some that aren't filtered. You can't have it both ways!)


Quote
  • Esoteric X-01, waveform of undithered 1kHz sinewave at –90.31dBFS:

    44.1 kHz PCM -> http://stereophile.com/images/archivesart/esEX1FIG4.jpg (http://stereophile.com/images/archivesart/esEX1FIG4.jpg)

And why would you have a linear PCM system without dither? It's like having a car engine without lubrication! i.e. broken! With dither, you'll have a noisy sine wave. Noisier than below, but it is "only" 16-bits, running at the limit of 16-bits (i.e. ~ -90dB).


Quote
DSD -> http://stereophile.com/images/archivesart/esEX1FIG5.jpg (http://stereophile.com/images/archivesart/esEX1FIG5.jpg)[/li][/list]

24-bit PCM would look even better

Quote

  • Analog 10 kHz square wave -> http://www.smr-home-theatre.org/surround20...s/image_051.jpg (http://www.smr-home-theatre.org/surround2002/images/show_images/image_051.jpg)

Correctly titled: "One specific analogue approximation to an ideal square wave" - a genuine ideal square wave has an infinite rise time. As mentioned above, engineers routinely do 1000 times better in an HD video system!

Quote
10 kHz square wave sampled using 44.1 kHz PCM -> http://www.smr-home-theatre.org/surround20...s/image_052.jpg (http://www.smr-home-theatre.org/surround2002/images/show_images/image_052.jpg)

Yes, that is the only audible component of a 10kHz square wave, and will sound identical to the "analogue" one. Therefore, storing any more is pointless. (Disagree? ABX please!)

Quote
10 kHz square wave sampled using 192 kHz PCM -> http://www.smr-home-theatre.org/surround20...s/image_054.jpg (http://www.smr-home-theatre.org/surround2002/images/show_images/image_054.jpg)

That is a clear 10kHz square wave with a low pass filter around 100kHz, but...


Quote
10 kHz square wave sampled using DSD -> http://www.smr-home-theatre.org/surround20...s/image_055.jpg (http://www.smr-home-theatre.org/surround2002/images/show_images/image_055.jpg)

...that is a noisy 10kHz square wave with a low pass filter around 100kHz. i.e. inferior!


Of course, if you read this thread (and similar threads - see the FAQ), you would know all this before posting.

Cheers,
David.
Title: Why 24bit/48kHz/96kHz/
Post by: dvda-sacd on 2006-09-05 14:47:29
2Bdecided, you can make all the objections you want (I've read all them before), but I haven't found any graphic that shows a clearly better performance (except noise floor) of LPCM over DSD.
Title: Why 24bit/48kHz/96kHz/
Post by: smz on 2006-09-05 14:54:30
2Bdecided, you can make all the objections you want (I've read all them before), but I haven't found any graphic that shows a clearly better performance (except noise floor) of LPCM over DSD.

Your problem is not finding a graphic that shows better performance of LPCM over DSD, but to find ears that confirms (ABX) better performance of DSD over LPCM!

Sergio
Title: Why 24bit/48kHz/96kHz/
Post by: dvda-sacd on 2006-09-05 15:00:40

2Bdecided, you can make all the objections you want (I've read all them before), but I haven't found any graphic that shows a clearly better performance (except noise floor) of LPCM over DSD.

Your problem is not finding a graphic that shows better performance of LPCM over DSD, but to find ears that confirms (ABX) better performance of DSD over LPCM!

Sergio

No, that's not my problem. I do prefer DSD because it's the most analogue-like digital system and my ears do confirm it to me. I don't need more tests.

Cheers!
Title: Why 24bit/48kHz/96kHz/
Post by: stephanV on 2006-09-05 15:04:51
But you are not willing to back your statements up, so what value do they have?

I think there is more than enough anecdotal evidence in the audio world, and too little real.
Title: Why 24bit/48kHz/96kHz/
Post by: Axon on 2006-09-05 15:10:01
What he said. You started off on claiming that DSD is numerically superior and requires no explanation, then when that is shot down - or just when somebody competent chimes in - you claim it always sounds better to you so you require no actual evidence that it's better? Make up your mind.

It sounds to me like you've been suckered into the marketing speak of DSD and you have no knowledge of digital audio besides the fluff pieces you've read. "analogue-like"?
Title: Why 24bit/48kHz/96kHz/
Post by: smz on 2006-09-05 15:12:48
No, that's not my problem. I do prefer DSD because it's the most analogue-like digital system and my ears do confirm it to me. I don't need more tests.

It is not a problem for you if this is just your personal preference, but yes, it is your problem if you come here to tell the rest of the world that things are like you think they are.

In a previous post you said that you:
Quote
... haven't done a statement concerning subjective sound quality, but objective sound quality

You can't say that. You, at most, can say that you have pictures showing a better signal quality. How this relate to sound quality as perceived by the human ear is a matter of discussion and if you say that there is a correlation, then it's up to you to prove that.

Cheers.

Sergio
Title: Why 24bit/48kHz/96kHz/
Post by: jlt on 2006-09-05 15:46:00
back to the center of the topic...

@ 2Bdecided

Quote
Why 24bit/48kHz/96kHz/, If 16bit/44.1kHz is good enough?


...pay atention 2B  (kiddin),you answered the question(again) :

Quote
And why would you have a linear PCM system without dither? It's like having a car engine without lubrication! i.e. broken! With dither, you'll have a noisy sine wave. Noisier than below, but it is "only" 16-bits, running at the limit of 16-bits (i.e. ~ -90dB).


you're right.
Title: Why 24bit/48kHz/96kHz/
Post by: legg on 2006-09-05 16:34:51
I've been reading this thread since the beginning and it has become a loop since a few pages back.
Title: Why 24bit/48kHz/96kHz/
Post by: 2Bdecided on 2006-09-05 16:48:42
2Bdecided, you can make all the objections you want (I've read all them before), but I haven't found any graphic that shows a clearly better performance (except noise floor) of LPCM over DSD.


Try Stanley P. Lipshitz & John Vanderkooy, "Why Professional 1-Bit Sigma-Delta Conversion is a Bad Idea".

I especially like the graph which shows 20dB (!!!!) of noise modulation (in the audio band!) in DSD systems.


That graph isn't a reason to avoid DSD in the real world, no more than your graphs are reasons to use it.


I do prefer DSD because it's the most analogue-like digital system and my ears do confirm it to me. I don't need more tests.


You've done double blind testing of the same source material direct, via LPCM 24/96, and via DSD?

Didn't think so.

What you mean is that you like the mastering on the DSD discs which exist. It's often very good. It still sound good when copied onto CD! (Not the CD layer of DSD, which is often taken from a different master).

Cheers,
David.


I've been reading this thread since the beginning and it has become a loop since a few pages back.


Yes, because the same false arguments get put forward again and again - they're bound to draw the same rebuttals.


If DSD, 24/96 PCM or whatever does sound better than CD, let's see the proof and figure out why.

Anyone who simply spouts marketing nonsense to back up their uninformed claims does their cause no good at all.


It make me wonder why such people bother to register at HA - they can't all be troll, surely?!

Cheers,
David.

(P.S. - it's great for me though - I have toothache - explaining the old arguments yet again takes my mind off it - you don't think I'd bothered otherwise, do you?!)
Title: Why 24bit/48kHz/96kHz/
Post by: TBeck on 2006-09-05 18:44:45
Sorry, nothing really substantial, but i couldn't resist:

I am wondering, if there would be a need to increase the frequency range of visual digitalization. Wouldn't it be nice to have DVD's with infrared content?

Reviewers could tell you about extended warmth of the new remastered infrared version of some well known DVD. And indeed, after some extended ABX-ing the reviewer definitely will have got another teint...
Title: Why 24bit/48kHz/96kHz/
Post by: krabapple on 2006-09-05 19:35:41

No, that's not my problem. I do prefer DSD because it's the most analogue-like digital system and my ears do confirm it to me. I don't need more tests.

It is not a problem for you if this is just your personal preference, but yes, it is your problem if you come here to tell the rest of the world that things are like you think they are.


It *always* the same problem from the audiophile camp: trying to 'prove' why they hear what they hear, and getting it wrong.  IT's never enough to stop at "I like SACD/DVD-A/LP better than CD', they always make that fateful leap into 'because....'. Though unlike this guy, usually they state their preference first and *then* make the quixotic attempt to justify the preference on objective ground...often with the SAME DAMN BOGUS SQUARE WAVE graphs that have been debunked over and over again.

It's like scientists dealing with creationists -- you shoot one down in an argument, and up pops another throwing the *same* agruments at you. It's like they *never* read anything but their own arguments, never do any research beyond their own limited circle of references, so they always think they're packing heat, and  counterarguments are always new and surprising to them.
Title: Why 24bit/48kHz/96kHz/
Post by: dvda-sacd on 2006-09-05 22:06:13
That graph isn't a reason to avoid DSD in the real world, no more than your graphs are reasons to use it.

We are all agreed. 

I know how it works, I know how it sounds, and DSD is definitely my personal preference. 

I don't mind if you prefer LCPM, MP3 or whatever you want.  That's your business.

Kind regards.
Title: Why 24bit/48kHz/96kHz/
Post by: dekkersj on 2006-09-05 22:39:43
Hi,

It has been a while but I see that some discussions will never end...

For those who want to find out for themselves what kind of dynamic range and frequency range is applicable for you, they can download a EAC image here: Logaritmic sweep (EAC) (http://amorgignitamorem.nl/Audio/LogsweepSinCosImages/LogSweep89109.zip) or a Nero version: Log sweep (Nero) (http://amorgignitamorem.nl/Audio/LogsweepSinCosImages/NeroLogSweep89109.zip)

After burning it to an audio cd, one can find 3 tracks. All three are sweeps from a few Hertz up to 20 kHz or so. The difference is the amplitude and they are at -89 dBFS, -99 dBFS and -109 dBFS.

The files are created in MatLab and at a resolution of 32 bits. In Audition, I dithered them correctly to 16 bits.

The test is as follows:
Set your listening level to normal and play track 1. If it remains silent, 16 bits is good enough for you. Otherwise, you need more.
Then put up the volume to somewhat louder to hear the sweep, I expect it to be in the order of 10 dB for the best hearable frequencies. If the track stops and you were able to hear frequencies up to that point in time, you need more than 44k1. Otherwise, you don't.

In my listening environment, 44k1/16b is sufficient.

Regards,
Jacco
Title: Why 24bit/48kHz/96kHz/
Post by: MikeQuell on 2006-09-06 00:35:09
It pitches better.
Title: Why 24bit/48kHz/96kHz/
Post by: KikeG on 2006-09-06 10:09:59
In my opinion DSD is a contrived format that is very inneficient bitrate-wise, and such a pain to process that currently nearly all SACDs are mastered from a PCM source.

The 2 "tricks" of SACD are the high sampling frequency and the high noise shaping. In practice, DSD frequency response hardly reaches 100 KHz if lots of ultrasonic noise want to be avoided:

http://www.stereophile.com/digitalsourcere...ony/index4.html (http://www.stereophile.com/digitalsourcereviews/401sony/index4.html)
http://stereophile.com/hirezplayers/814/index6.html (http://stereophile.com/hirezplayers/814/index6.html)
http://www.stereophile.com/digitalsourcere...180/index7.html (http://www.stereophile.com/digitalsourcereviews/180/index7.html)
http://www.stereophile.com/digitalsourcere...inn/index4.html (http://www.stereophile.com/digitalsourcereviews/1203linn/index4.html)
http://www.stereophile.com/digitalsourcere...515/index5.html (http://www.stereophile.com/digitalsourcereviews/515/index5.html)

The response looks similar to that of a 192KHz PCM device. In fact, 16-bit 192 KHz PCM with agressive noise shaping would be much better that DSD in terms of noise, similar in terms of effective frequency response, could be perfectly dithered (as opposed to DSD), and would take much less space (Edit: wrong, it would would take similar space, see what I meant in my next post).
Title: Why 24bit/48kHz/96kHz/
Post by: dvda-sacd on 2006-09-06 10:55:18
The response looks similar to that of a 192KHz PCM device. In fact, 16-bit 192 KHz PCM with agressive noise shaping would be much better that DSD in terms of noise, similar in terms of effective frequency response, could be perfectly dithered (as opposed to DSD), and would take much less space.

Much less space? Explain this, please.

Regards.
Title: Why 24bit/48kHz/96kHz/
Post by: KikeG on 2006-09-06 11:43:03
  • 16 bit x 192 kHz = 3,072 bit per second
  • 1 bit x 2,822.1 kHz = 2,822.1 bit per second
Regards.


OK, it seems my memories betrayed me, sorry. To be precise, quoted from this paper http://sjeng.org/ftp/SACD.pdf (http://sjeng.org/ftp/SACD.pdf)

Quote
Finally, consider 8-bit, four-times-oversampled PCM with
noise shaping. This is also a data rate one-half that of DSD and
double that of CD, with a sampling rate of 4 × 44,100 =
176,400 Hz. It can achieve a noise floor 120 dB below full
scale up to 20 kHz, using 96 dB of noise shaping, and a total
noise power of –19 dBFS. Its frequency response would be
flat to 80 kHz. This example is perhaps the most instructive of
the lot. For a data rate one-half that of DSD, it achieves a
comparable signal bandwidth, with a similar noise power
density up to 20 kHz, but much lower power above this
frequency, and 28 dB lower total noise power. It is fully
TPDF-dithered, and so is completely artefact free. At one-half
the data rate it outperforms DSD on every count! DSD is a
profligate wastrel of capacity.
Title: Why 24bit/48kHz/96kHz/
Post by: dvda-sacd on 2006-09-06 12:16:56
  • 16 bit x 192 kHz = 3,072 bit per second
  • 1 bit x 2,822.1 kHz = 2,822.1 bit per second
Regards.


OK, it seems my memories betrayed me, sorry.

It's all right.

Regards.
Title: Why 24bit/48kHz/96kHz/
Post by: krabapple on 2006-09-06 17:12:26
classic stuff from the Linn Unidisc universal player measurements in Stereophile

Quote
The occasional low-level clicks [produced on CD playback only] may well be specific to our review sample, but the Unidisk's somewhat disappointing measured performance on CD playback compared with SACD and DVD caused my eyebrows to rise a little


If the manufacturers can't be trusted to give all formats a fair shake in their 'uni-players' , it's not reasonable to assume audible differences are due to *the formats*

And this was a pricey ($11,000) 'high-end' player!
Title: Why 24bit/48kHz/96kHz/
Post by: Radetzky on 2006-09-07 13:42:21
classic stuff from the Linn Unidisc universal player measurements in Stereophile

Quote
The occasional low-level clicks [produced on CD playback only] may well be specific to our review sample, but the Unidisk's somewhat disappointing measured performance on CD playback compared with SACD and DVD caused my eyebrows to rise a little


If the manufacturers can't be trusted to give all formats a fair shake in their 'uni-players' , it's not reasonable to assume audible differences are due to *the formats*

And this was a pricey ($11,000) 'high-end' player!


I remember Stereophile proving the Redbook layer of the 30th Anniversary Edition of Pink Floyd Dark Side Of The Moon was deliberately flawed compared to the SACD layer.  The Redbook layer was very often clipped (the wave would hit 0dB and just be flat for a while... hugh !!).  The SACD layer wasn't touching 0dB once.

The weirdest part was the fact that STEREOPHILE made that public.

Anyway.. the moral of the story is that Sony has a monetary interest in making sure Redbook CDs don't sound as good as SACD.

I am just glad Vinyl sold more than SACD and DVD-A combined.

p.s.: sorry, I can't back-up my claims with references.  Too long and my coffee is getting cold.
Title: Why 24bit/48kHz/96kHz/
Post by: dekkersj on 2006-09-07 15:16:04
[...]
I remember Stereophile proving the Redbook layer of the 30th Anniversary Edition of Pink Floyd Dark Side Of The Moon was deliberately flawed compared to the SACD layer.  The Redbook layer was very often clipped (the wave would hit 0dB and just be flat for a while... hugh !!).  The SACD layer wasn't touching 0dB once.
[...]

Old story and it is very known that mastering is the issue here. Not that it is clipped, although it is a sign on the wall. I believe that the 30th anniversary DSOTM PCM layer was not a downsample from the DSD master. It was the same as a previously released version, probably from a cost cutting point of view.

Regards,
Jacco
Title: Why 24bit/48kHz/96kHz/
Post by: Radetzky on 2006-09-07 16:56:15

[...]
I remember Stereophile proving the Redbook layer of the 30th Anniversary Edition of Pink Floyd Dark Side Of The Moon was deliberately flawed compared to the SACD layer.  The Redbook layer was very often clipped (the wave would hit 0dB and just be flat for a while... hugh !!).  The SACD layer wasn't touching 0dB once.
[...]

Old story and it is very known that mastering is the issue here. Not that it is clipped, although it is a sign on the wall. I believe that the 30th anniversary DSOTM PCM layer was not a downsample from the DSD master. It was the same as a previously released version, probably from a cost cutting point of view.

Regards,
Jacco


Funny thing is, I have both the 1992 Shine On Box Set version and the MFSL Ultradisc II version and neither are clipped.

... they went out of their way to make it clipped.  They had an economic reason to do so.  SACD cannot sound better (technically speaking, it does not make sense).  So what are you gonna do if you are an exec at Sony and you want to push the SACD format?  Hmmm... why not re-release CDs but badly mastered!!

The only thing that holds them back, IMHO, is that there are already correct versions out there.  It would be too obvious if they re-released everything clipped or otherwise incorrect.  People could easily see what is going on.

Nahhh... IMHO, again, the thing SACD has is multichannel.  Now, remix stuff properly by really taking advantage of that technology (not like what they did with Kind of Blue for example) AND remove the pesky copy-protection and I MIGHT look at it.

But hey... if you really want to throw your money at SACD, be my guest.  I don't care.
Title: Why 24bit/48kHz/96kHz/
Post by: krabapple on 2006-09-20 20:36:12

[...]
I remember Stereophile proving the Redbook layer of the 30th Anniversary Edition of Pink Floyd Dark Side Of The Moon was deliberately flawed compared to the SACD layer.  The Redbook layer was very often clipped (the wave would hit 0dB and just be flat for a while... hugh !!).  The SACD layer wasn't touching 0dB once.
[...]

Old story and it is very known that mastering is the issue here. Not that it is clipped, although it is a sign on the wall. I believe that the 30th anniversary DSOTM PCM layer was not a downsample from the DSD master. It was the same as a previously released version, probably from a cost cutting point of view.

Regards,
Jacco



I know that the mastering of the DSD and CD layers are significantly different on this disc -- I've verified that with my own rips and AD captures --  but I have never heard that the CD layers was the same as a previous version -- do you remember which one?

What's funny to me is that the CD layer is louder than the DSD layer , due to the compression...and thus in light of psychoacoustic 'loudness bias', it could well be judged  *better sounding* than the DSD layer to lots of people in lots of situations.    (Though of course, difference in EQ might mitigate the 'loudness' effect...)
Title: Why 24bit/48kHz/96kHz/
Post by: krabapple on 2006-09-20 20:47:38


[...]
I remember Stereophile proving the Redbook layer of the 30th Anniversary Edition of Pink Floyd Dark Side Of The Moon was deliberately flawed compared to the SACD layer.  The Redbook layer was very often clipped (the wave would hit 0dB and just be flat for a while... hugh !!).  The SACD layer wasn't touching 0dB once.
[...]

Old story and it is very known that mastering is the issue here. Not that it is clipped, although it is a sign on the wall. I believe that the 30th anniversary DSOTM PCM layer was not a downsample from the DSD master. It was the same as a previously released version, probably from a cost cutting point of view.

Regards,
Jacco


Funny thing is, I have both the 1992 Shine On Box Set version and the MFSL Ultradisc II version and neither are clipped.

... they went out of their way to make it clipped.  They had an economic reason to do so.  SACD cannot sound better (technically speaking, it does not make sense).  So what are you gonna do if you are an exec at Sony and you want to push the SACD format?  Hmmm... why not re-release CDs but badly mastered!!



That is onle possibility -- teh otehr is that they were simply following common 'modern' CD remastering practice, which seems to assume that people will do most of their CD listening in cars or other noisy environments, where louder and less dynamic  *is* often subjectively better.

What *is* underhanded, though, is implying that such discs are demonstrations of how SACD (the format) sounds better than CD.  I've seen some ignorant reviewers do just that...and I've never seen Sony put out any disclaimers regarding its own hype of SACDs.

Quote
The only thing that holds them back, IMHO, is that there are already correct versions out there.  It would be too obvious if they re-released everything clipped or otherwise incorrect.  People could easily see what is going on.

Nahhh... IMHO, again, the thing SACD has is multichannel.  Now, remix stuff properly by really taking advantage of that technology (not like what they did with Kind of Blue for example) AND remove the pesky copy-protection and I MIGHT look at it.



I like the Kind of Blue multichannel mix -- it's definittely a laid-back use of multichannel (essentially the 'surround' is for ambience), but I don't really think this music needs something more aggressive.  'In A Silent Way' has more actual 'surround' content, and that fits it.


Quote
But hey... if you really want to throw your money at SACD, be my guest.  I don't care.


SACDs *are* often mastered more tastefully than their CD remaster counterparts (whether hybrid or independent) -- perhaps because the Scarlet Book spec forbids 'clipping' in the DSD domain (though IAUI you can take it through a PCM step, clip the hell out of it, lower the level, then transcode to DSD --  if you wanted to)
Title: Why 24bit/48kHz/96kHz/
Post by: Radetzky on 2006-09-21 00:37:23
SACDs *are* often mastered more tastefully than their CD remaster counterparts (whether hybrid or independent) -- perhaps because the Scarlet Book spec forbids 'clipping' in the DSD domain (though IAUI you can take it through a PCM step, clip the hell out of it, lower the level, then transcode to DSD --  if you wanted to)


I know the Analog Productions Bill Evans Hybrid SACD have been equally well mastered on their SACD layer and on their Red Book layer.  They CAN do a great job on Red Book. Oh well...

Redbooks are supposedly (I don't have the expertise to make a blanket statement) so badly mastered these days that people are often looking foir the non-re-mastered version of their favourite records (look at SteveHoffman.tv Music forums for an idea of what I am talking about).

IMHO, the current model of the music industry should be completely rehauled.  Image if remastering engineers could select whatever tapes they wished to work on and people could choose the version they wanted.  That would generate competition between mastering engineers and they would have to make a great job.
Title: Why 24bit/48kHz/96kHz/
Post by: Johnnywellas on 2006-10-20 03:20:56
Well, i haven't really read all of the topic's posts thoroughly. I did scan them reasonably, though, so i'm going to post a question regarding samplerates :

I read somewhere (don't remember exactly where right now, and it's 3 a.m., so...  ) that psychoacoustically, our hearing response to spatial perception was based on differences regarding the sound that reaches each ear, and there were a mainly 3 key factors - amplitude (volume) , spectral differences (high frequency rolloff) and phase differences. The cool remark was that we (human beings ?!  ) can't distiguish the directionality of sounds intervalled below about 5ms.

Given this, at 192khz samplerates, we have 5,2083(3) microseconds between samples, which is a very close call to the beferementioned value. Now i ask you - wouldn't this, per se, give us an extremely more detailed stereo/multichannel image? Does anybody work on a studio/facility that has equipment capable of performing tests on this?

Anyway, don't bother to kill me if what i just said is complete nonsense. I don't have any TonMeister or other acoustics degrees, and i'm way beyond my usual lack of sleep threshold, so... 
Title: Why 24bit/48kHz/96kHz/
Post by: saratoga on 2006-10-20 03:57:52
Given this, at 192khz samplerates, we have 5,2083(3) microseconds between samples, which is a very close call to the beferementioned value. Now i ask you - wouldn't this, per se, give us an extremely more detailed stereo/multichannel image?


No.  The sample period has absolutely nothing to do with that.  It only determines the maxium possible frequency in the signal.  For frequencies below half the sample rate, the signal produced by your DAC is continuous (and thus the distance between adjacent samples is 0).  Or at least it is for an ideal DAC (what you actually hear will depend on your equipment)!

Does anybody work on a studio/facility that has equipment capable of performing tests on this?


The effect of sample rate was mathamatically proved in the 1930s.  No need to test, but you're welcome to if you feel like proving that mathamatics works

Anyway, don't bother to kill me if what i just said is complete nonsense. I don't have any TonMeister or other acoustics degrees, and i'm way beyond my usual lack of sleep threshold, so... 


EE degree would be best for this sort of thing
Title: Why 24bit/48kHz/96kHz/
Post by: Woodinville on 2006-10-20 05:56:40
Given this, at 192khz samplerates, we have 5,2083(3) microseconds between samples, which is a very close call to the beferementioned value. Now i ask you - wouldn't this, per se, give us an extremely more detailed stereo/multichannel image?


You can easily resolve your 5 or 10 microsecond interaural ITD's with a 44.1 kHz sampled signal.

Try it, using a Gaussian pulse, yourself.
Title: Why 24bit/48kHz/96kHz/
Post by: smack on 2006-10-20 09:05:36
The cool remark was that we (human beings ?!  ) can't distiguish the directionality of sounds intervalled below about 5ms.

Given this, at 192khz samplerates, we have 5,2083(3) microseconds between samples, which is a very close call to the beferementioned value.

Did you notice that "ms" means millisecond and that one millisecond equals 1000 microseconds? 
Title: Why 24bit/48kHz/96kHz/
Post by: Axon on 2006-10-20 16:12:05
ms can refer to microseconds in a few circumstances.(But IIRC, they mostly involve SPICE plots.)
Title: Why 24bit/48kHz/96kHz/
Post by: Woodinville on 2006-10-20 19:00:22

The cool remark was that we (human beings ?!  ) can't distiguish the directionality of sounds intervalled below about 5ms.

Given this, at 192khz samplerates, we have 5,2083(3) microseconds between samples, which is a very close call to the beferementioned value.

Did you notice that "ms" means millisecond and that one millisecond equals 1000 microseconds? 



ITD's have been detected (by humans) in the 5 to 10 MICROsecond range.

In fact, they have been so detected in digital audio signals at 44.1/16, which ought to be a great big hint for those who think they can't be.
Title: Why 24bit/48kHz/96kHz/
Post by: Johnnywellas on 2006-10-20 19:21:12
A big thumbs up for the replies. Thanks for helping a rookie on this question. Is there any practical advantage on working on such high samplerates (96k+)?
Title: Why 24bit/48kHz/96kHz/
Post by: Axon on 2006-10-20 19:37:11
There are very few cut and dry reasons why you'd really want to. If you're doing an analog capture of a high res digital format (SACD/DVD-A) it would make sense. You'd also need 96khz (and maybe a little more?) to record quadraphonic records. And of course if you're recording something that you know has huge amounts of ultrasound (car keys, castanets, etc), that needs to be recorded for some reason.
Title: Why 24bit/48kHz/96kHz/
Post by: marcan on 2006-10-21 00:00:53
A big thumbs up for the replies. Thanks for helping a rookie on this question. Is there any practical advantage on working on such high samplerates (96k+)?

For recording/production/mixing, it might make sense. Converters and plugins (digital effects) have a better result with 96 or 88.2 khz (you better get 88.2 if you want a final release digitally resample at 44.1).
However, you have to work on a music that require very open high end which is not the case with typical pop/rock (Beatles for example rarely went higher than 12 khz), because it takes more CPU and HD space
Title: Why 24bit/48kHz/96kHz/
Post by: jmartis on 2006-12-13 18:37:41
Sorry for buming old thread, but this has been lying in my head for months: someone said, that you can perfectly "capture" all frequencies up to samplerate/2. So, how do you then capture a 22049Hz sine wave with a sampling frequency of 44100Hz without a 1hz "breathing volume" ?? Sorry if this question is stupid, but I have no answer!
Title: Why 24bit/48kHz/96kHz/
Post by: Axon on 2006-12-13 19:09:54
A good enough reconstruction filter will reproduce a 22049 hz sine wave, um, good enough.
Title: Why 24bit/48kHz/96kHz/
Post by: Mercurio on 2006-12-13 19:36:23
Sorry for buming old thread, but this has been lying in my head for months: someone said, that you can perfectly "capture" all frequencies up to samplerate/2. So, how do you then capture a 22049Hz sine wave with a sampling frequency of 44100Hz without a 1hz "breathing volume" ?? Sorry if this question is stupid, but I have no answer!


Maybe there is the answer

http://www.hydrogenaudio.org/forums/index....amp;#entry44506 (http://www.hydrogenaudio.org/forums/index.php?showtopic=4347&pid=44506&mode=threaded&show=&st=25&#entry44506)

http://www.hydrogenaudio.org/forums/index....d&pid=19328 (http://www.hydrogenaudio.org/forums/index.php?showtopic=2035&mode=threaded&pid=19328)
Title: Why 24bit/48kHz/96kHz/
Post by: christopher on 2006-12-16 06:14:42
Sorry for buming old thread, but this has been lying in my head for months: someone said, that you can perfectly "capture" all frequencies up to samplerate/2. So, how do you then capture a 22049Hz sine wave with a sampling frequency of 44100Hz without a 1hz "breathing volume" ?? Sorry if this question is stupid, but I have no answer!



As previously mentioned, the DAC will have to do a bit of waveform reconstruction in some shape or form, because you just can't sample above the Nyquist, and I guess that's (well, it IS) where the quality of the circuitry and chips used comes into play.


Here's a thought... You have your 24/96, your 192, your 44.1 and 88.2... DSD only has a bitdepth of ONE bit, but its samplerate is 2.8224MHz. How cool is that  (though granted it is sigma delta modulation, not plain pulse code, when we're at these kind of levels of samplerates it really is just 'you say potatoe, I say potatoe') And I know this is common knowledge, it's just a part of my brain still marvels at the fact that you can have a single bit of bitdepth and still have amazing quality recordings as long as your samplerate is sufficient. It's madness, madness I say!
Title: Why 24bit/48kHz/96kHz/
Post by: saratoga on 2006-12-16 06:58:00
Sorry for buming old thread, but this has been lying in my head for months: someone said, that you can perfectly "capture" all frequencies up to samplerate/2. So, how do you then capture a 22049Hz sine wave with a sampling frequency of 44100Hz without a 1hz "breathing volume" ?? Sorry if this question is stupid, but I have no answer!


By using a reconstruction filter whose impulse response is based on a sin wave.  Specifically a sinc, a type of decaying, sin-like function.  If you really do add infinitely many of them together (while multiplying them times the sample value as you go) you will find that everything cancels, except a single sin wave at 22049Hz.

One of the reasons this works is that the sampled version without the DAC contains lots of jumps, where the amplitude changes very quickly between samples.  These jumps are high frequencies.  The sinc function is actually a filter with a flat passband, and perfect frequency rejection in the stop band.  So when you add and multiply those infinate number of sincs together, you're actually using a perfect filter to remove all the noise added by sampling in the first place.  With all the noise removed, the signal is perfectly reconstructed (SNR==infinity), and if you look at the waveform, you'll see that all the space between samples has been filled in by the filter.

Sorry if thats hard to follow.  Its the best I can do without lots of math.
Title: Why 24bit/48kHz/96kHz/
Post by: Kees de Visser on 2006-12-16 09:01:47
As previously mentioned, the DAC will have to do a bit of waveform reconstruction in some shape or form, because you just can't sample above the Nyquist, and I guess that's (well, it IS) where the quality of the circuitry and chips used comes into play.
The DAC's reconstruction filter has to make sure that no frequencies above 0.5 Fs come through. In practice this means that the filter has to start attenuating at a lower frequency. Some real world DAC chip numbers are (CS4340A): -0.05dB at 0.4535Fs and -3dB at 0.4998Fs. At 44.1 kHz sampling rate 0.4535 Fs corresponds to 20 kHz. Your 22.049 kHz will be reproduced, but at a lower level. (microphones, speakers and headphones are likely to have much more roll-off, be it less steep) Increasing the sampling rate will move this "problem" to a higher frequency.
Title: Why 24bit/48kHz/96kHz/
Post by: HbG on 2006-12-16 21:16:38
Sorry for buming old thread, but this has been lying in my head for months: someone said, that you can perfectly "capture" all frequencies up to samplerate/2. So, how do you then capture a 22049Hz sine wave with a sampling frequency of 44100Hz without a 1hz "breathing volume" ?? Sorry if this question is stupid, but I have no answer!


I had asked just that question, here are the answers: http://www.hydrogenaudio.org/forums/index....showtopic=47764 (http://www.hydrogenaudio.org/forums/index.php?showtopic=47764)
Title: Why 24bit/48kHz/96kHz/
Post by: putanik on 2006-12-18 00:50:36
Sorry for buming old thread, but this has been lying in my head for months: someone said, that you can perfectly "capture" all frequencies up to samplerate/2. So, how do you then capture a 22049Hz sine wave with a sampling frequency of 44100Hz without a 1hz "breathing volume" ?? Sorry if this question is stupid, but I have no answer!


being a mathematician and professionally working with DSP audio, i can fully confirm your suspicions that perfect A2D-D2A reconstruction does not exist in this world.

First, Nyquist theorem refers to signals that do not exist in this world because any signal with limited spectrum must be unlimited in duration. Second, filters that have transient (0 dB to -98 dB) bandwidth of 1/22000 exist only on paper. Third .... well, take it easy. 

Back to the initial question: you need "breathing space", both in terms of hertz, bits depth and channels. the more - the easier it will be to record and playback. being very tight on headroom requires perfection and leaves no space for mistakes whatsoever. Honestly, I think that recording classical music with 25...35 dB variations in dynamics, huge orchestra, soloists and choirs (as Mahler's 8th, etc) in 2 channel 16/44.1 without serious alterations to material is not possible. that may be reason why classical/jazz jumped to SACD - at least on recording stage.

On another hand... IMHO... currently, proportion of recordings that would benefit from SACD/16b+/48k+ is tiny.
Title: Why 24bit/48kHz/96kHz/
Post by: 2Bdecided on 2006-12-18 16:06:00
Welcome to HA!

being a mathematician and professionally working with DSP audio, i can fully confirm your suspicions that perfect A2D-D2A reconstruction does not exist in this world.

Does "perfect" anything exist in this world? (Other than mathematics, maybe?)


Quote
First, Nyquist theorem refers to signals that do not exist in this world because any signal with limited spectrum must be unlimited in duration.

True. Let us say it's "good enough"


Quote
Second, filters that have transient (0 dB to -98 dB) bandwidth of 1/22000 exist only on paper.

Well, if they can exist on paper, they can exist in the real world too. Do you have a special PC or DSP that can't do maths?

In any case, we don't need to transition in 1/22050 of the sample rate. Let us keep everything up to 20kHz intact, and make sure we have nothing above 22050kHz. Now the transition is 1/10th of the sample rate - very easy, with a perfect result up to 20kHz, some attenuation above this, and with no aliasing or distortion.

Do your ears work above 20kHz for normal amplitudes?


Quote
Third .... well, take it easy.

No, bring it on - that's the point of this thread - there are many proposed reasons why 44.1kHz isn't enough - most fall down immediately with rigorous scrutiny. Even the best ones are completely unproven.


Quote
Back to the initial question: you need "breathing space", both in terms of hertz, bits depth and channels.

You may need more of everything for ease of acquisition, clean processing etc, but that has never been doubted. I know the benefit of a greater bit depth for processing. I know the benefit of a higher sample rate (at least internally/synthetically) for some kinds of processing. The question is about the delivery method.


Quote
the more - the easier it will be to record and playback. being very tight on headroom requires perfection and leaves no space for mistakes whatsoever. Honestly, I think that recording classical music with 25...35 dB variations in dynamics, huge orchestra, soloists and choirs (as Mahler's 8th, etc) in 2 channel 16/44.1 without serious alterations to material is not possible.

Mahler requires more than 20kHz bandwidth, and more than 96dB dynamic range? I'd better buy some new ears!

To be fair, you have a point about the theoretical dynamic range requirements for a tiny proportion of the worlds recordings if peaks are to be kept, and the quietest parts are to remain noise-free (though noise shaped dither can do the job at 16-bits in practice).

Quote
On another hand... IMHO... currently, proportion of recordings that would benefit from SACD/16b+/48k+ is tiny.

Yes.

Many would benefit from good multichannel, but almost no one is proposing this!

Cheers,
David.
Title: Why 24bit/48kHz/96kHz/
Post by: Pio2001 on 2006-12-19 12:33:23
I think that recording classical music with 25...35 dB variations in dynamics, huge orchestra, soloists and choirs (as Mahler's 8th, etc) in 2 channel 16/44.1 without serious alterations to material is not possible. that may be reason why classical/jazz jumped to SACD - at least on recording stage


Do you have ABX results to backup this opinion, or is it mere conjecture ?
Title: Why 24bit/48kHz/96kHz/
Post by: jlohl on 2006-12-19 21:26:14
Quote
that may be reason why classical/jazz jumped to SACD - at least on recording stage.


In fact very very few recordings are done in DSD (the SACD format), much much more use PCM at 96KHz/24bits.
But the main reason for those "high def" format in recording, is that when you mix and edit you may loose some bits (roundings, noise,...), so better be conservative from the beginning.
Title: Why 24bit/48kHz/96kHz/
Post by: putanik on 2006-12-24 19:02:17
Quote
that may be reason why classical/jazz jumped to SACD - at least on recording stage.


In fact very very few recordings are done in DSD (the SACD format), much much more use PCM at 96KHz/24bits.
But the main reason for those "high def" format in recording, is that when you mix and edit you may loose some bits (roundings, noise,...), so better be conservative from the beginning.



Dear Jlohl,

I was not refereing to SACD specifically - just used it as an example of hi-res format. 24/96 is essentially the same for me.

The reasons for using hi-res in recordings, as far as i know, are slightly different. Recording folks absolutely hate to come up to performers and tell them something like "play it again 'cause we had too many samples clipped". They told me (most stories came from a guy who was in charge of recording Richter playing piano) that at "good old times" they often had a person per mic who would manually, during recording session, "in real time", compress sound by adjusting gain knob for this particular mic, to combat dynamic loudness variations (30 dB for Richter was reported as typical). What they want is to setup mics with approximate gains, record as is, and master it later.  For doing that, they need about 120 dB noise&distortion-free. It is still a problem - even the best mics are too noisy. Please do not ask me what other tricks recordings engineers play - ask them:-)

BTW, rounding errors during mixing are avoided by converting everything into 32bit floating point just after recording.
Title: Why 24bit/48kHz/96kHz/
Post by: putanik on 2006-12-24 19:12:41

I think that recording classical music with 25...35 dB variations in dynamics, huge orchestra, soloists and choirs (as Mahler's 8th, etc) in 2 channel 16/44.1 without serious alterations to material is not possible. that may be reason why classical/jazz jumped to SACD - at least on recording stage


Do you have ABX results to backup this opinion, or is it mere conjecture ?


Dear Pio, yes, this is my personal opinion - of course - but I am afraid I am not the only one sharing it. Could you read through what recording engineers are saying about their trade? I do not know a particular web site - but I am sure you'll find it if you really want to find out the answer. You can start with http://channelclassics.com/ (http://channelclassics.com/) - huge proportion of their CDs are winning numerous best quality awards. If they only could sign up more artists of DG, HM, etc, and more frequently...
Title: Why 24bit/48kHz/96kHz/
Post by: putanik on 2006-12-24 21:24:31
Dear David,

1. resampling and D2A/A2D.

do you have Matlab? Then try an easy script to see what kind of resampling 192->44.1 filter you are going to deal with:
[n,f0,a0,w]=firpmord([20/96 22/96],[1 0], [0.05 1e-5]);
plot(firpm(n,f0,a0,w));
n

as you see, filter duration is ~ 1.5 ms. in terms of simple math, it's ok. But from physical perspective, you can NOT use a filter longer that transient dynamic in your waveform. That's the problem!!! If you have a sharp transient in original analog, it will be accommodated by a looooong sinc with f-cut period when you do A-D-A. If you are familiar with modems, you know that the guys went to quite a trouble to overcome ISI with rised cosine filters, etc. The essence here is the same - but the solution isn't applicable. Am i clear?

2. limits of hearing.

2.1. frequency.

Limits of hearing is a controversial issue. I am an old guy, 16 kHz is my hearing limit for pure tones on 84 dB SPL in normal conditions. true. but ... ohhhh. humans can hear sounds well above the 20 k range - up to 200k, if they fed into skull bone.  in normal situations people do not report them, at least as "hearing" - but ... as something else. the question is how that contributes to perception of music. and the answer is ... i do not know. read a lot about it, experimented myself, but still can't make my mind.

For speech, things are relatively easy. 7kHz is good enough, if you care only of the meaning and intelligibility. But ... you need 12k+ . Why? those frequency are very directional and killed by distance easily. if you pass them - you end up passing a "message" of close proximity and face-to-face condition as "non-verbal";-))) clues (the funniest thing is that 7-12 kHz content is essentially noise) If you have a tele-meeting with 2 systems (one on 16 kHz sampling and another on 32), the later will result in better rapport / mutual understanding - and better outcome. Yet, no one will state that 32 kHz "sounds" so much better (if any, for most) than 16kHz on voice. Do you see the analogy?

2.2. dynamic range.

have you ever been to a good anechoic chamber? people who work there, hate it. after some time, you start hearing how blood goes through your veins, and all other unpleasant phenomenas. It seems so loud and ugly ... and when you get out to the street, you feel deafened by everything you normally ignore, and it takes some time to "recover". How long - ...  that depends.

when the sound is delivered on 70...90 dB SPL, only few people will be affected by IMDs of reasonable order lower than 0.01% (-80 dB). Now add dynamic range variations (~20 dB) and peak-to-average (~15 dB). what do we get? 115? But what do you do if dynamic range variations are >30 dB (Mahler ;-))), for example), and you are on 16 bits with mid-fi CD player having IMD of 0.007%? Yes, recordings are routinely companded for home listening ... but not down to zero (at least in classical music).

If CD format had "frame" gain embedded into their bit-stream, I would be much happier;-)))

3. delivery method.

economics, sir. $5,000 top-notch CD player sounds as good or worse comparing to a $700 SACD on the same double-layer SACD.

4. multi-channel.

I had a hope that SACD/DVD-A would bring hi-res multi-channel audio to mainstream. Format war and still prohibitive cost of good loudspeakers (x2.5) killed my hopes. Currently, only few companies still do multichannel SACDs, channel classics being a good example. right now, buying 3 more loudspeakers just to play few tens of disks ... does not sound sane to me.

I still hope that there will be a new wave of innovations in 7..15 years (THX is pushing for it), and we'll enjoy multi-channel hi-quality sound. I guess that it will be based on MP3-like compression - but i would not bet a single hair from my already balding scalp for it:-)))

best regards,
michael.

PS> do you think that the lowest octave (20-40 Hz) is more/less important for the music than the 10..20 kHz octave? :-)))
Title: Why 24bit/48kHz/96kHz/
Post by: saratoga on 2006-12-25 01:40:35
Dear David,

1. resampling and D2A/A2D.

do you have Matlab? Then try an easy script to see what kind of resampling 192->44.1 filter you are going to deal with:
[n,f0,a0,w]=firpmord([20/96 22/96],[1 0], [0.05 1e-5]);
plot(firpm(n,f0,a0,w));
n

as you see, filter duration is ~ 1.5 ms. in terms of simple math, it's ok. But from physical perspective, you can NOT use a filter longer that transient dynamic in your waveform. That's the problem!!!


Why does the filter delay matter for offline processing?
Title: Why 24bit/48kHz/96kHz/
Post by: SebastianG on 2006-12-25 16:50:05
1. resampling and D2A/A2D.
[n,f0,a0,w]=firpmord([20/96 22/96],[1 0], [0.05 1e-5]);
plot(firpm(n,f0,a0,w));
n

as you see, filter duration is ~ 1.5 ms. in terms of simple math, it's ok. But from physical perspective, you can NOT use a filter longer that transient dynamic in your waveform. That's the problem!!! If you have a sharp transient in original analog, it will be accommodated by a looooong sinc with f-cut period when you do A-D-A.

I don't know what all the fuss is about. The "ringing" is rather low (in amplitude), short (in time),  and above 20 kHz. In combination with a reconstruction filter with similar properties this "ringing" should be a great deal below the threshold of hearing in all situations -- and certainly in those more practical cases that matter (usually most of the signal energy is concentrated in the lower frequency regions).

Granted, higher sampling rates like 88.2 or 96 kHz makes life a bit easier and leaves less room for any doubts but it has yet to prove somebody (convincingly) that 44 kHz is not enough. I don't think the length of impulse responses alone is a good argument here.

SG
Title: Why 24bit/48kHz/96kHz/
Post by: Lokutus01 on 2006-12-25 17:14:34
I love to listen to 24/88,2 (pcmed SACD) or 24/96 (DVD-A) music, I really do!

And: It is actually no problem for me, that there is maybe no or a big difference... I get it for free so I don´t complain and enjoy the music.

Some SACDs or DVD-As I am owning are the best mastered medias I know... the "why" is interesting to discuss about, but not important if you like to listen to good music (which is good recorded as well).

Ciau
Andreas
Title: Why 24bit/48kHz/96kHz/
Post by: marcan on 2006-12-25 19:30:58
I love to listen to 24/88,2 (pcmed SACD) or 24/96 (DVD-A) music, I really do!

And: It is actually no problem for me, that there is maybe no or a big difference... I get it for free so I don´t complain and enjoy the music.

Some SACDs or DVD-As I am owning are the best mastered medias I know... the "why" is interesting to discuss about, but not important if you like to listen to good music (which is good recorded as well).

Ciau
Andreas

It's very kind to share your feelings here but this is not the purpose of the discussion.
The discussion is, with exactly the same source can you hear (ABX) the difference between a 16/44 (correctly dithered) and a higher definition format (24/96, …). 

Back to the topic, is the resampling from 24/96 to 16/44 completely transparent?
Title: Why 24bit/48kHz/96kHz/
Post by: Lokutus01 on 2006-12-25 20:42:29

...

It's very kind to share your feelings here but this is not the purpose of the discussion.
The discussion is, with exactly the same source can you hear (ABX) the difference between a 16/44 (correctly dithered) and a higher definition format (24/96, …). 

Back to the topic, is the resampling from 24/96 to 16/44 completely transparent?


Hey, you are welcome.... I am always trying to share my hearing-experiences with other people.

My girlfriend  btw. is my blindtesting-machine number one ;-))) She is passing by and I ask her what is better... hihi.
Title: Why 24bit/48kHz/96kHz/
Post by: Pio2001 on 2006-12-25 23:31:11
Could you read through what recording engineers are saying about their trade?


This is not the point. What I am interested in is blind listening tests comparing 44.1 kHz 16 bits with higher resolution meterial.
The 44.1 kHz limit seems very hard to overcome. Making use of more than 16 bits, on the other hand, seems to depend mostly on how loud you listen to.
Title: Why 24bit/48kHz/96kHz/
Post by: Kees de Visser on 2006-12-26 14:48:51
Could you read through what recording engineers are saying about their trade?
This is not the point. What I am interested in is blind listening tests comparing 44.1 kHz 16 bits with higher resolution meterial.
The 44.1 kHz limit seems very hard to overcome. Making use of more than 16 bits, on the other hand, seems to depend mostly on how loud you listen to.
AFAIK most recording engineers don't really care about that question. Data storage is very cheap nowadays, most professional AD/DA converters can operate in 24/96, 24/192 or DSD mode and there is probably no reason for not using the higher resolution format. At least I'm not aware of any test that proves 16/44.1 to be perceptually superior to higher bitrate formats.
In the old analog days the tape cost could eat up quite a bit of the production budget and recording at double speed would actually double the tape cost. With hard disc recording the cost of data storage is so small (probably less than the catering budget) that very few recording engineers want to be bothered with the question about audible advantages.
I agree with you that most claims about sonic advantages of hi-res audio remain unproven, which is a pity. On the other hand it's not that easy to do. Simply moving the 44.1/96/192 switch on the AD/DA equipment might introduce some new variables. IMO upsampling all material to the highest rate and listening at that rate is the fairest solution.
If there's anyone on this forum who has done some serious hi-res ABX'ing I'd love to hear about the setup and the results.
Title: Why 24bit/48kHz/96kHz/
Post by: marcan on 2006-12-26 20:58:08
AFAIK most recording engineers don't really care about that question. Data storage is very cheap nowadays, most professional AD/DA converters can operate in 24/96, 24/192 or DSD mode and there is probably no reason for not using the higher resolution format. At least I'm not aware of any test that proves 16/44.1 to be perceptually superior to higher bitrate formats.

I mix and produce pop/rock/electro/hip-hop and I have never seen hires recording. Recordings are usually 24/44 or 24/48. I use 32 bits float for bouncing. But higher sampling rate will just eat your CPU without giving you substantial benefits.
Actually, with this kind of music, we are not looking for very high frequencies. On records like the Beatles, you won't go higher than 12k and I don't think any AE will tell you it's not well balanced.
I think multi channel is much more interesting and audible for the listener than hires format.

With classical or jazz it might be different though.
Title: Why 24bit/48kHz/96kHz/
Post by: krabapple on 2007-01-03 17:51:22
Dear Pio, yes, this is my personal opinion - of course - but I am afraid I am not the only one sharing it. Could you read through what recording engineers are saying about their trade?


I often do, on www.prosoundweb.com forums.  Let's just say that the merits of and need for DSD have by no means been universally accepted.
Title: Why 24bit/48kHz/96kHz/
Post by: putanik on 2007-01-06 05:12:30
Could you read through what recording engineers are saying about their trade?


This is not the point. What I am interested in is blind listening tests comparing 44.1 kHz 16 bits with higher resolution meterial.
The 44.1 kHz limit seems very hard to overcome. Making use of more than 16 bits, on the other hand, seems to depend mostly on how loud you listen to.



Dear Pio,

you first question was: can we distinguish between life performance and 16/44? my answer was - what kind of question is it?

you ask now - is there any significant difference between 16/44.1 and 24/96|DSD? my answer is:
1) YES. absolutely, It's so obvious that I do not believe in the need for any ABX tests. but hmm .... if and only if you playing through hmm... pretty good setup. mine is  room-with-rt60=250ms-lots-of-auralex-in-strategic-places / jmlab spectral / naim / marantz8001 - chain b&k calibrated -> upto 90...95 dB SPL it's ok. then jmlab non-linearities kick in.
2) NO. unless you ensure that your total signal chain is good (in-room freq resp +- 6dB, IMD < -90dB), it should not matter at all. I doubt that a "normal-room"+b&w6xx+rotel or something alike will reveal any difference between CD and sacd.

that's my 2 c.

PS> of course, even B&W+rotel setups are not "normal", i know:-)
Title: Why 24bit/48kHz/96kHz/
Post by: putanik on 2007-01-06 05:45:21

1. resampling and D2A/A2D.
[n,f0,a0,w]=firpmord([20/96 22/96],[1 0], [0.05 1e-5]);
plot(firpm(n,f0,a0,w));
n

as you see, filter duration is ~ 1.5 ms. in terms of simple math, it's ok. But from physical perspective, you can NOT use a filter longer that transient dynamic in your waveform. That's the problem!!! If you have a sharp transient in original analog, it will be accommodated by a looooong sinc with f-cut period when you do A-D-A.

I don't know what all the fuss is about. The "ringing" is rather low (in amplitude), short (in time),  and above 20 kHz. In combination with a reconstruction filter with similar properties this "ringing" should be a great deal below the threshold of hearing in all situations -- and certainly in those more practical cases that matter (usually most of the signal energy is concentrated in the lower frequency regions).

Granted, higher sampling rates like 88.2 or 96 kHz makes life a bit easier and leaves less room for any doubts but it has yet to prove somebody (convincingly) that 44 kHz is not enough. I don't think the length of impulse responses alone is a good argument here.

SG


Dear Sebastian,

-13 dB for sidelobes is not low at all. 1.5ms is not short. Yes, ringing will be below the threshold, but when you get those transients into slightly non-linear tracts of amp and loudspeaker, IMDs result. and they won't be below the threshold for <most> materials containing sharp fronts. you may ask - how are humans affected by EXACTLY them ... I honestly do not know how to measures JUST those.

Adding later: Sebastian, I figured out that I can not explain properly what bothers me in the reconstructed waveform of short fronts even without involving non-linear distortions. bear with me. i'll try later.

yes, filter length alone is not THE proof. you know it's not alone.

PS> similarly... you know that there is on-going long-bearded mess around de-essing. it is very funny. many people believe that we do not produce sounds with f>20k because we do not hear them. well... we do. "s" and "f" spectrum spreads high. at least up to 48k - measured myself. and if the slew-rate of your front-end circuitry and its filtering abilities are not so good, then distortions come. very audible. very easily simulated. record hi-feq with fast B&K mics, and then use matlab to simulate slew-rate limiting and poorly filtered aliasing. write back to .wav. listen. enjoy, if you can:-)
Title: Why 24bit/48kHz/96kHz/
Post by: Kees de Visser on 2007-01-06 08:44:13
you ask now - is there any significant difference between 16/44.1 and 24/96|DSD? ...
hi putanik, you forgot "audible". The HA forum clearly distinguishes between technical differences (can be measured) and audible differences. There's no doubt (IMO) that higher resolution formats allow a more accurate reproduction of the original signal. The questions are:
1) is the difference useful (various opinions to be expected) and 2) is the difference audible (properly tested)
Quote
my answer is:
1) YES. absolutely, It's so obvious that I do not believe in the need for any ABX tests. but hmm .... if and only if you playing through hmm... pretty good setup...
In general when there are (measured or calculated) differences, an ABX (or equivalent) test is necessary to check if they are audible or not. Until the test proves otherwise, audible differences have to be concidered non-existant. It's ok to use the best equipment and the best pair of ears (test persons) available and take as much time as needed, but audible differences should be demonstrated with test results.

Hi res audio listening tests are hard to do without introducing new variables that might blurr the results. The test setup is very critical and IMHO most consumers and even most audio professionals don't have the facilities to do a proper (scientific) test. The AES (Audio Engineering Society) is preparing a conference (http://www.aes.org/events/31/) about hi-res audio in june 2007. "Perception" is amongst the topics. I'm very interested in the results (I might even go there ).
Title: Why 24bit/48kHz/96kHz/
Post by: SirChristof on 2007-01-06 12:37:51
Here's how I view this, mentally;

To me, 44.1khz/16-bit is "alt-preset-standard". [APS]

APS sounds great! .....most of the time.

When it does not sound great, we can re-rip the CD to lossless, and compare.

Upon comparison, we may find that APS still sounds identical to  the lossless rip.

We may also find that it does not.

Millions of people own billions of CDs with audio(or "music") on them.

A large number have been encoded to APS.

With such large-scale penetration and wide-spread listening, we found problems.  This is the main reason why LAME has tons of problem samples that can be used to improve it, where other lossy encoders may have many less "known problem samples".

----------------------------------------------------------------------------------------------------------

Now ponder this; (answer these questions for yourself)

How many audio CDs do you own?
If you encoded every song, on every CD, to APS, and listened to each and every one of them in full, do you think you would ever find one that was not transparent?  What if everyone who owned any CD's did the same?

How many "above-CD-Resolution" discs do you own?
If you converted every song, on every disc, to 44.1khz/16-bit, and listened to each and every one of them in full, do you think you would ever find one that was not transparent?  What if everyone who owned high-res discs did the same?

What if all your CDs were high-res formats?
Title: Why 24bit/48kHz/96kHz/
Post by: Axon on 2007-01-06 19:07:58
you ask now - is there any significant difference between 16/44.1 and 24/96|DSD? my answer is:
1) YES. absolutely, It's so obvious that I do not believe in the need for any ABX tests. but hmm .... if and only if you playing through hmm... pretty good setup. mine is  room-with-rt60=250ms-lots-of-auralex-in-strategic-places / jmlab spectral / naim / marantz8001 - chain b&k calibrated -> upto 90...95 dB SPL it's ok. then jmlab non-linearities kick in.

So, basically, you adamantly believe it, but you are not competent enough to actually prove it, except by inaccurately appealing to equipment quality.

Quote
2) NO. unless you ensure that your total signal chain is good (in-room freq resp +- 6dB, IMD < -90dB), it should not matter at all. I doubt that a "normal-room"+b&w6xx+rotel or something alike will reveal any difference between CD and sacd.

So, basically, you adamantly believe it, but you are not competent enough to actually prove it, except by inaccurately appealing to equipment quality.

Quote
-13 dB for sidelobes is not low at all. 1.5ms is not short. Yes, ringing will be below the threshold, but when you get those transients into slightly non-linear tracts of amp and loudspeaker, IMDs result. and they won't be below the threshold for <most> materials containing sharp fronts. you may ask - how are humans affected by EXACTLY them ... I honestly do not know how to measures JUST those.

So, basically, you adamantly believe it, but you are not competent enough to actually prove it, except by inaccurately appealing to equipment quality.

Quote
PS> similarly... you know that there is on-going long-bearded mess around de-essing. it is very funny. many people believe that we do not produce sounds with f>20k because we do not hear them. well... we do. "s" and "f" spectrum spreads high. at least up to 48k - measured myself. and if the slew-rate of your front-end circuitry and its filtering abilities are not so good, then distortions come. very audible. very easily simulated. record hi-feq with fast B&K mics, and then use matlab to simulate slew-rate limiting and poorly filtered aliasing. write back to .wav. listen. enjoy, if you can:-)

Oh! Hey! A new argument! Except that people have been trying exactly this sort of thing for, you know, the last 60+ years without any success, so I have zero expectation that this constitutes any better evidence.

I'm OK with strongly believing in the superiority of 24/96, but only when basing it on a superior argument. Which you, so far, have not demonstrated.
Title: Why 24bit/48kHz/96kHz/
Post by: putanik on 2007-01-06 20:20:23
Dear Kees,

"audible" is what i meant by "significant".  Properly "scientifically" tested ... you made me laugh.
There are as many opinions on what constitutes "proper" audio setup as the number of experts (i am one of those). what AES will come up with? another "absolute" threshold of hearing definition which turns out to be 50% quantile of a wildly scattered distribution??? do you really care?

SACD vs. CD difference is so apparent that I smile at requests of measuring it with statistics.

Yes, I agree that most consumers and audio pros do not have facilities or means to do "proper" tests, whatever is meant by them.
Title: Why 24bit/48kHz/96kHz/
Post by: putanik on 2007-01-06 20:30:43
So, basically, you adamantly believe it, but you are not competent enough to actually prove it, except by inaccurately appealing to equipment quality.


Dear Axon, I am a mathematician specializing in adaptive audio processing and an audio expert, and I've been doing it for 25 years quite successfully. Unless you can demonstrate with proper scientific arguments that I am not competent, I would like to ask you be careful with words. Just state that your opinion is different and that will be fine.
Title: Why 24bit/48kHz/96kHz/
Post by: Axon on 2007-01-06 21:36:42


So, basically, you adamantly believe it, but you are not competent enough to actually prove it, except by inaccurately appealing to equipment quality.


Dear Axon, I am a mathematician specializing in adaptive audio processing and an audio expert, and I've been doing it for 25 years quite successfully. Unless you can demonstrate with proper scientific arguments that I am not competent, I would like to ask you be careful with words. Just state that your opinion is different and that will be fine.


Fair enough - claiming a lack of competence was deliberately inflammatory and I apologize for that.

Nevertheless, I remain unconvinced of the evidence you provide. While the scientific arguments you provide for the audibility of certain things make sense (ringing from reconstruction filters, for instance, or intermodulation into low frequencies), it also makes a great deal of sense to me that such effects are not audible. As others have mentioned - these sorts of things are not universally agreed on by professionals. Appealing to your own authority, or that of others, does not really matter much to me.

What I would really like to see are quantitative descriptions of the effects you are mentioning, and/or concrete simulations of them. However, as I interpret your earlier posts, you don't have these yet. I'll stop acting like a dick by calling you out on them, but those sorts of things are what I'm most interested in, and probably others here are too.
Title: Why 24bit/48kHz/96kHz/
Post by: knutinh on 2007-01-06 21:39:28
If the difference between CD and hirez formats is so faint that it can only be discovered by 1 out of 1000 people, 1 out of 1000 times, using 1 out of 1000 rigs... Is it really that important?

On the other hand, if the differences are clearly heard in most cases, then why is it so difficult to support those views by ABX testing?

-k
Title: Why 24bit/48kHz/96kHz/
Post by: putanik on 2007-01-06 23:10:03
Nevertheless, I remain unconvinced of the evidence you provide. While the scientific arguments you provide for the audibility of certain things make sense (ringing from reconstruction filters, for instance, or intermodulation into low frequencies), it also makes a great deal of sense to me that such effects are not audible. As others have mentioned - these sorts of things are not universally agreed on by professionals. Appealing to your own authority, or that of others, does not really matter much to me.


Dear Axon, I far as I am aware there is no single theory of hearing that has been universally agreed upon, and in absence of it ... arguing is more or less just an exchange of opinions, and a honest attempt to move towards better understanding of what is the fine mechanics of audio perception.

I did quite a bit of auditory testing, routinely passing 30...50 people through blind tests in a day. People are indeed very different. once i've been in charge of choosing transducers for a wide-band headset. on measurements, one of them was beating AKG/Sennheizer heads down, another was plain bad, and 3 more were in between. there were people that spotted the difference right away. there were others who said - no difference, they are all the same. there were few who reported that the worst-on-measurements is actually the best sounding. being a scientist myself, i refused making any "scientific" arguments (in your sense of this world) out of those experiments. all I could say was to chart distributions and advise of choosing the one whose distribution was leaning right the most. I could not say to people who liked the worst transducer that they were wrong. that's the only way they hear, and it's right for them because they do not have another pair of ears:-)))

similarity, if you are convinced that all those previously discussed distortions are negligible.... you are 100% in your right, with addition - "for your ears".
Title: Why 24bit/48kHz/96kHz/
Post by: putanik on 2007-01-06 23:29:43
If the difference between CD and hirez formats is so faint that it can only be discovered by 1 out of 1000 people, 1 out of 1000 times, using 1 out of 1000 rigs... Is it really that important?

On the other hand, if the differences are clearly heard in most cases, then why is it so difficult to support those views by ABX testing?

-k


Dear K, I do not think that percentage of people to whom hi-res makes sense is as tiny as 0.1%.  but ... nowadays, for most home systems, the source is not the worst offender in terms of distortions. May I reiterate:
1) there is little if any sense to advance to hi-res if the rest of the reproducing chain is well within CD limits.
2) I will hesitate (at least:-) ) to recommend  to each one to go and buy hi-end equipment (on par with 24/96 clarity) and ugly acoustic foam while there are not enough hi-res (sa/dvd-a/etc) (and well recorded) cds to justify spendings - unless you are kind of mad/oversensitive and music _really_ matters to you.
3) "usual" ABX testing approach - like... download those samples, play them on your PC and e-mail back what you think - is not very appropriate for hi-res vs. cd comparison. if you have better ideas, I'd love to hear from you.
Title: Why 24bit/48kHz/96kHz/
Post by: knutinh on 2007-01-08 13:27:55

If the difference between CD and hirez formats is so faint that it can only be discovered by 1 out of 1000 people, 1 out of 1000 times, using 1 out of 1000 rigs... Is it really that important?

On the other hand, if the differences are clearly heard in most cases, then why is it so difficult to support those views by ABX testing?

-k


Dear K, I do not think that percentage of people to whom hi-res makes sense is as tiny as 0.1%.  but ... nowadays, for most home systems, the source is not the worst offender in terms of distortions. May I reiterate:
1) there is little if any sense to advance to hi-res if the rest of the reproducing chain is well within CD limits.

I was under the impression that most recording studios use 24/96, 24/88.2 or at least 24/44.1 as it is readily available and inexpensive and suitable for the kind of dynamic processing that they are into.

The fact that most pop music is compressed to death is something hirez isnt likely to cure once the market penetration is high enough. I cant see that this would remove any sensitivity to ultra-sound in either case.
Quote
2) I will hesitate (at least:-) ) to recommend  to each one to go and buy hi-end equipment (on par with 24/96 clarity) and ugly acoustic foam while there are not enough hi-res (sa/dvd-a/etc) (and well recorded) cds to justify spendings - unless you are kind of mad/oversensitive and music _really_ matters to you.

Do you mean that the full potential of hirez/lorez recordings cannot be exploited using a pair of headphones, a good DAC connected to a DVD-V/DVD-A/PC source and a good headphone amplifier?
Quote
3) "usual" ABX testing approach - like... download those samples, play them on your PC and e-mail back what you think - is not very appropriate for hi-res vs. cd comparison. if you have better ideas, I'd love to hear from you.

I am thinking more like large-scale, controlled, AES-type scientific tests that should quite easily be able to get positive resultat from a properly conducted ABX test. IF (and that is a big if) this really matters to human hearing. My gut-feeling is that it does not.

The sole exeption may be that the difference is simply so small that you need vastly large listening panels and very exceptional source material etc to get significant results. But in that case, what is the real gain, and what does it say about hifi-magazine-journalists that can make bold conclusions after a few hours of informal listening?

-k
Title: Why 24bit/48kHz/96kHz/
Post by: Light-Fire on 2007-01-08 14:45:53
...SACD vs. CD difference is so apparent that I smile at requests of measuring it with statistics...


Dear putanik

How do you know the contents of the SACD and CD compared were originally the same and not treated differently in the studio?

If you can't be sure about that your test is not valid.
Title: Why 24bit/48kHz/96kHz/
Post by: krabapple on 2007-01-08 16:16:17
Dear Axon, I am a mathematician specializing in adaptive audio processing and an audio expert, and I've been doing it for 25 years quite successfully. Unless you can demonstrate with proper scientific arguments that I am not competent, I would like to ask you be careful with words. Just state that your opinion is different and that will be fine.


Can you point us to some of your published work in your guise as 'audio expert'?  I'm hoping it rises to a better evidentiary standard than the 'work' you have presented here in support of audible difference between redbook and hi rez audio -- which has so far been hypothetical or anecdotal.  Since you claim to have tested a goodly number of subjects under blind conditions, surely you could report one or a few undoubtedly 'positive' results, perhaps in a JAES article?
Title: Why 24bit/48kHz/96kHz/
Post by: Woodinville on 2007-01-08 20:05:47
Dear Axon, I far as I am aware there is no single theory of hearing that has been universally agreed upon, and in absence of it ... arguing is more or less just an exchange of opinions, and a honest attempt to move towards better understanding of what is the fine mechanics of audio perception.


Regardless of what you are or are not aware of, the ultimate sensitivity of the human hearing apparatus has been tested and examined over and over again, and the results are both consistant and reliable.

Given such information, and given the knowledge of what kind of levels reproduction systems can provide, it is quite easy to figure out a maximum bit depth.

It is also easy to figure out what kind of bandwidth, but not exactly what kind of filters to use to achieve that bandwidth.

It is, further, easy to point to the clearly supported work that shows the minimum noise level possible at the ear drum in a 1-atmosphere (earth standard STP) setting, and that sets yet another lower limit for what the auditory system can possibly detect.

Such people as Fletcher, Zwicker, Stevens, and the like have provided a large body of data that has been confirmed over and over, by a host of researchers.

Do you have any evidence that more than 19 bits makes any sense whatsoever for final presentation to the human being in any standard kind of sound system? If so, let's hear it.

Furthermore, in any average listening space, do you have any evidence that more than 16 bits is at all necessary?

Thank you in advance for substantive, testable, measurable evidence.

PS> similarly... you know that there is on-going long-bearded mess around de-essing. it is very funny. many people believe that we do not produce sounds with f>20k because we do not hear them. well... we do. "s" and "f" spectrum spreads high. at least up to 48k - measured myself. and if the slew-rate of your front-end circuitry and its filtering abilities are not so good, then distortions come. very audible. very easily simulated. record hi-feq with fast B&K mics, and then use matlab to simulate slew-rate limiting and poorly filtered aliasing. write back to .wav. listen. enjoy, if you can:-)



Ok, now, please tell me something:

How much of that 40kHz signal propagates through 10 meters to your ear from the source, not counting the 1/r^2 loss, of course?

How does that compare to what travels to your ear at 4kHz? or 400Hz?

What happens at 46% RH, 20C?

What happens at 20% RH, 20C? (earth standard atmosphere)

What do you suppose the base noise level at the eardrum is?
Title: Why 24bit/48kHz/96kHz/
Post by: knutinh on 2007-01-08 20:34:32
I guess this has been said before, but what the heck... Along the chain of sound propagation:

1. I do agree that many sound sources produce a good deal of energy at infra-sound frequencies. This is a necessary condition for hirez audio to make sense, but not enough on its own.

2. What is the frequency response of the most preferred and valuable studio microphones above 20kHz?

3. What is the distance and HF loss for typical recording sessions where "hifi" argueably counts the most? According to This link (http://www.earthworksaudio.com/tech/hf_sound.pdf), the loss at 22 meter at 40kHz can be ~6dB compared to lofreq.

4. What is the trend of fletcher-munson curves at frequencies below 20kHz? If we were to make a prediction based on those, then add some dB in case we are wrong, how many dB attenuation compared to 2kHz do you think?


If instruments like violin output 1-2% of their total energy between 20kHz and 40kHz, then is attenuated by 3-6dB due to HF airloss, then 6-12dB due to microphone response curve, then we really have to have quite obvious sensitivity to be able to appreciate it in our livingroom?



The real test is if any material containing significant ultrasound content (such as white noise, sines or carefully recorded music), played back on a high-linearity high-bandwidth system (such as headphones) makes it possible for the listener to reliably identify a lowpass-filter inserted into the chain at 18, 20, 22 or some other cutoff-frequency. My gut-feeling as well as understanding of present work is that somewhere around these frequencies, even good listeners start to hear no difference.

-k
Title: Why 24bit/48kHz/96kHz/
Post by: SirChristof on 2007-01-08 20:41:24
After X years have passed, and DVDs are considered "Small"  at 4.7GB (or 9.4 dual layer), with other disc-formats supporting 25-100GB and beyond, at some point in time the distribution rate for common audio will increase.  Plenty say it's nothing more than a waste of space, which may be true in many/most cases, but if things were distributed on, say, DVDs, we would have roughly 6.5x the amount of space to store the music.

If all other things are equal (meaning similar prices, no *DRM*, etc), and the only difference is the data rate of the media, I will always opt for the higher rate.

Everyone can *pottentially* win with "high res" formats.  If you want something that is as close as possible to the original, that would be the form it is presented in.  If the source media rate is something you consider "too high", you are always free to reduce it.  Knock yourself out, downsample it, truncate it, your the boss!  Why stop at 44.1/16?  You might be able to reduce it to 32khz/12bit and still find it acceptable.  Once you get it down to 32/12, find a lossy encoder that can further package it for you.

Having said that, I personally do not want to see any new format take hold if in addition higher sampling rates & bit depths, it brings new consumer headaches (*DRM*).

Since the only high-res formats I am aware of (SACD & DVD-Audio) have iron chains everywhere---I will stick with redbook CD for the forseeable future.
Title: Why 24bit/48kHz/96kHz/
Post by: Kees de Visser on 2007-01-08 20:45:52
How much of that 40kHz signal propagates through 10 meters to your ear from the source, not counting the 1/r^2 loss, of course?
Concert halls could consider charging audiofile rates for the front row seats  .
Many recording mixing and mastering engineers are using hi-res audio like 24/96, 24/192 and dsd. Interestingly enough I don't know a single one that uses super tweeters (say >30 kHz) for monitoring. Apparently, if there is an audible advantage in using hi-res audio at all, it's probably not the extended bandwidth.
Title: Why 24bit/48kHz/96kHz/
Post by: Woodinville on 2007-01-08 20:45:56
Everyone can *pottentially* win with "high res" formats.  If you want something that is as close as possible to the original, that would be the form it is presented in.


Out of curiousity, what doyou mean by "as close as possible to the original"?

In a 2-channel CD we reproduce two of literally millions of measurements of some combination of the sound pressure and the sound velocity at a given point.

The original has all that information contained in it, not just two points that we provide. How is raising the sampling rate going to improve this?
Title: Why 24bit/48kHz/96kHz/
Post by: SirChristof on 2007-01-08 21:07:19
Out of curiousity, what doyou mean by "as close as possible to the original"?


Simply that, all other things being equal, a 96/24 recording is closer to the analog source than 44.1/16.

(and if material were distributed at that rate (96/24), and you are 100% convinced it has no benefit whatsoever, nothing is stopping you from downsampling it for your own purposes, which was my original point)
Title: Why 24bit/48kHz/96kHz/
Post by: putanik on 2007-01-09 08:31:49
I was under the impression that most recording studios use 24/96, 24/88.2 or at least 24/44.1 as it is readily available and inexpensive and suitable for the kind of dynamic processing that they are into.


I was referring mostly to amp +speakers as limiting factors.

Quote
Do you mean that the full potential of hirez/lorez recordings cannot be exploited using a pair of headphones, a good DAC connected to a DVD-V/DVD-A/PC source and a good headphone amplifier?

if you have GREAT headphones with almost any brand-name soundcard IF windows does not play with sound, and source is on the same fs as DAC in soundcard should work. just good headphones ... as you know, membrane moves as 1/f^2, so deep base+mid/hi-freqs may produce bad IMD in one-speaker cans - unless they are great.

Quote
I am thinking more like large-scale, controlled, AES-type scientific tests that should quite easily be able to get positive resultat from a properly conducted ABX test. IF (and that is a big if) this really matters to human hearing. My gut-feeling is that it does not.


My gut feeling led by my experience is different. AES-type ABX testing was many time laughed at, for example, see http://www.stereophile.com/features/113/ (http://www.stereophile.com/features/113/)

but ... why don't you walk into a hi-end audio place and ask them to show you the best speakers on a sample SACD? spend there an hour, and then we'll see if you gut feeling changes. or, have you already been there?

cheers,
putanik.
Title: Why 24bit/48kHz/96kHz/
Post by: knutinh on 2007-01-09 08:41:13
but ... why don't you walk into a hi-end audio place and ask them to show you the best speakers on a sample SACD? spend there an hour, and then we'll see if you gut feeling changes. or, have you already been there?

cheers,
putanik.

If you can produce a SACD + a CD where both are products of the exactly same hirez master, and both are purely downconverted to take full advantage of the respective format.

Of course, I will also have to have some means of gain-matching the two streams down to fractions of a dB.

If you are basing your knowledge on listening to the SACD-mixes out there vs the CD-mixes out there, shurely you must see that this can be biased information?

My gut-feeling stems from doing home-studio work and getting a real down-to-earth view on the qualities of my own hearing, combined with a MSc thesis on digital filters involving blind listening tests showing me that audiophiles are not always the best people for listening.

I have got a SACD/DVD-player, but I have no clear conclusion as to the few titles that I have heard. May be that I am victim to the reverse placebo-effect.

-k
Title: Why 24bit/48kHz/96kHz/
Post by: putanik on 2007-01-09 08:41:40



...SACD vs. CD difference is so apparent that I smile at requests of measuring it with statistics...


Dear putanik

How do you know the contents of the SACD and CD compared were originally the same and not treated differently in the studio?

If you can't be sure about that your test is not valid.


Dear Light-Fire, yes, you are right, I can not be sure. What i can say ... i have quite a few recordings of the same Ombra fedele by Vivica Genaux. HM disks include SACD and CD versions. She also recorded the same aria for EMI/virgin in Bajazet. I easily distinguish between EMI and HM versions (i can't say about artistic merits, but EMI is much worse in technical terms). I can not distinguish between SACD and CD versions of HM - except that SACD makes me loose my breath and forget of the moment being.

As a pragmatic mathematician ... I can't imagine a reason why anybody would spend his/her time on altering CD version vs stereo SACD, especially in classical music, where profits are 0.
Title: Why 24bit/48kHz/96kHz/
Post by: knutinh on 2007-01-09 08:48:30
As a pragmatic mathematician ... I can't imagine a reason why anybody would spend his/her time on altering CD version vs stereo SACD, especially in classical music, where profits are 0.

As a mathematician you seem unusually relaxed in terms of proving causality, comparing apples with apples and lacking general scepticism (towards both the industry as well as sceptical people like myself).

How do you come to the conclusion that the two signalpaths below "must" be identical except delilvery medium, and that any difference that you observe is caused by CD being inferior to SACD? How do you know that other differences are conscious, and not caused by some random mishap, or cost-cutting?
[source1]->[pre-proc1]->[delivery-medium1]->[playback-device1]
[source2]->[pre-proc2]->[delivery-medium2]->[playback-device2]

As a mathematician, can you shortly give me a mathematical analysis of the benefits of DSD vs LPCM in terms of:
1. Technical/functional terms
2. Subjective/perceptual terms
3. As an effective means of producing high-quality modern music

-k
Title: Why 24bit/48kHz/96kHz/
Post by: putanik on 2007-01-09 09:16:18
Regardless of what you are or are not aware of, the ultimate sensitivity of the human hearing apparatus has been tested and examined over and over again, and the results are both consistant and reliable.


o-o.

Do you have any evidence that more than 19 bits makes any sense whatsoever for final presentation to the human being in any standard kind of sound system? If so, let's hear it.


I do not understand your question. At the moment, I am not aware of any A2D/D2A having more that 19 TRUE bits. 20 bits -> 122 dB ~ .7e-4% IMD/linearity error. nobody claimed it - yet. AKM's AK4394 claims only 100 dB THD, and that's quite typical now. Could you please explain yourself?

Furthermore, in any average listening space, do you have any evidence that more than 16 bits is at all necessary?


That' what I wrote to K: in average listening space 16 bits are ample.


Ok, now, please tell me something:
How much of that 40kHz signal propagates through 10 meters to your ear from the source, not counting the 1/r^2 loss, of course? How does that compare to what travels to your ear at 4kHz? or 400Hz? What happens at 46% RH, 20C? What happens at 20% RH, 20C? (earth standard atmosphere) What do you suppose the base noise level at the eardrum is?


Why asking what we both know? why don't read again what I wrote, please - when you use "slow" mic (close distance assumed) then...



As a pragmatic mathematician ... I can't imagine a reason why anybody would spend his/her time on altering CD version vs stereo SACD, especially in classical music, where profits are 0.

As a mathematician you seem unusually relaxed in terms of proving causality, comparing apples with apples and lacking general scepticism (towards both the industry as well as sceptical people like myself).

How do you come to the conclusion that the two signalpaths below "must" be identical except delilvery medium, and that any difference that you observe is caused by CD being inferior to SACD? How do you know that other differences are conscious, and not caused by some random mishap, or cost-cutting?
[source1]->[pre-proc1]->[delivery-medium1]->[playback-device1]
[source2]->[pre-proc2]->[delivery-medium2]->[playback-device2]

As a mathematician, can you shortly give me a mathematical analysis of the benefits of DSD vs LPCM in terms of:
1. Technical/functional terms
2. Subjective/perceptual terms
3. As an effective means of producing high-quality modern music

-k


Dear K, as a pro mathematician, i do not believe in math, and quite aware of its limitations. don't you know the anecdote about mathematicians, ending with "their answers are as precise as useless"? what about my skepticism ... you run to conclusions a bit too fast:-)))

I did not came to any conclusion.  I wrote - I do not hear the difference that may be attributed to different mastering/mixing. that's it.

I already wrote that for me DSD/PCM of sufficient bit depth/speed are essentially the same.

best regards,
putanik.



If you are basing your knowledge on listening to the SACD-mixes out there vs the CD-mixes out there, shurely you must see that this can be biased information?


I readily admit I am biased.  (am i the only one?) Yes, most of audiophiles ... I agree with all you said and politely avoided to state directly.
Title: Why 24bit/48kHz/96kHz/
Post by: knutinh on 2007-01-09 09:19:37
Dear K, as a pro mathematician, i do not believe in math, and quite aware of its limitations. don't you know the anecdote about mathematicians, ending with "their answers are as precise as useless"? what about my skepticism ... you run to conclusions a bit too fast:-)))

I believe that mathematics can give quite precise answers - but that mathematicians are not always the right people to ask the question that is to be answered ;-)
Quote
I did not came to any conclusion.  I wrote - I do not hear the difference that may be attributed to different mastering/mixing. that's it.

"SACD vs. CD difference is so apparent that I smile at requests of measuring it with statistics"

That is exactly the same statement that I frequently get from the audiophile crowd when they dont want critical questions about their magic pyramids, green markers, voodo-power-cables etc.
Quote
I already wrote that for me DSD/PCM of sufficient bit depth/speed are essentially the same.

Is that for you as a subjective listener in a blind experiment, for you as a listening conusmer, or as a mathematical analysis of the dsp that occurs in dsd vs lpcm systems?

-k
Title: Why 24bit/48kHz/96kHz/
Post by: putanik on 2007-01-09 09:29:43
The real test is if any material containing significant ultrasound content (such as white noise, sines or carefully recorded music), played back on a high-linearity high-bandwidth system (such as headphones) makes it possible for the listener to reliably identify a lowpass-filter inserted into the chain at 18, 20, 22 or some other cutoff-frequency. My gut-feeling as well as understanding of present work is that somewhere around these frequencies, even good listeners start to hear no difference.

-k


Dear K, most people will not distinguish the sound of the music sampled at 192 and it, filtered with 12 kHz LPF. It sounds pretty much the same. it feels different, that's the problem. and here I am as clueless as you - why and how???


How much of that 40kHz signal propagates through 10 meters to your ear from the source, not counting the 1/r^2 loss, of course?
Concert halls could consider charging audiofile rates for the front row seats  .
Many recording mixing and mastering engineers are using hi-res audio like 24/96, 24/192 and dsd. Interestingly enough I don't know a single one that uses super tweeters (say >30 kHz) for monitoring. Apparently, if there is an audible advantage in using hi-res audio at all, it's probably not the extended bandwidth.


Please excuse me for replaying instead of Woodinwille... concert halls do charge a lot for premium sitting (mid of 5...10'th row). LaScala : 170 e vs. 12 e. (http://www.teatroallascala.org/public/LaScala/EN/venditaBiglietti/prezzi/AcquistoBigliettiOpera/prezzi_opera_scala/index.html)

some of very good recording engineers use poor radioshack speakers. what does it prove?
Title: Why 24bit/48kHz/96kHz/
Post by: knutinh on 2007-01-09 09:34:03

The real test is if any material containing significant ultrasound content (such as white noise, sines or carefully recorded music), played back on a high-linearity high-bandwidth system (such as headphones) makes it possible for the listener to reliably identify a lowpass-filter inserted into the chain at 18, 20, 22 or some other cutoff-frequency. My gut-feeling as well as understanding of present work is that somewhere around these frequencies, even good listeners start to hear no difference.

-k


Dear K, most people will not distinguish the sound of the music sampled at 192 and it, filtered with 12 kHz LPF. It sounds pretty much the same. it feels different, that's the problem. and here I am as clueless as you - why and how???

I believe that I can pick out 12kHz lpf filtered material in a blind ABX. I have never tried though.

But I can (or could) track sines up to about 17kHz. From there on I could not tell if the tester had the amplitude on or off.

My thing is that even if the difference manifests itself as a "feeling", then logically, an ABX test should reveal it. If a well designed ABX test gives a 0-result, then the most likely mechanism in my view really is that we are fooling ourselves.

The notion that humans are some kind of objective measurement devices always astonish me. Throughout our lives, we are making all kinds of desitions based on emotions, "gut-feeling", believes and hopes. There are countless stories in the hifi industry about cheap components being wrapped into "ultra-hifi-clothing" receiving standing ovations from the hifi journalists.

This is what makes it difficult to figure out what is "really going on", and probably why the hifi crowd cannot accept scientific findings - they dont match their belief system. Try explaining a religious person that according to this or that empirical evidence, certain statements in their religious book seems to be imprecise. You will never win that discussion because the core of religion is believing, not proving.

The goal of science the way that I see it (and I was never any good at scientific philosophy at uni) is to design systems for observing, describing and modelling the world that minimize the subjective biasing from the researchers. Eliminating is probably impossible, but if the goal is to understand how stuff works, then we should strive to remove bias from researchers trying to make a name of themselves.

A great tool then is to make every step repeatable so that anyone with the right equipment can try to repeat the experiment. Of course, there may be hidden factors that cause the outcome to be more or less randomised, but this should make it a lot harder to form a scientific career upon falsifying tests. As long as the empirical evidents can be trusted, then anyone reading a paper are quite free to read, calculate and think about the logical conlusions of the author and debate his findings.

The difference to hifi-stuff is so obvious. Everyone is claiming large differences, measurable differences, plots showing differences (but no axes...). But no one ever brings proofs to the table. Hardly anyone of the
"big names" in mainstream audiophile circles are represented in AES or some other reputable journals. And everyone is argueing over what is better. In between we have all kinds of snake-oil producers making profit on people with less critical view of the world.

regards
k
Title: Why 24bit/48kHz/96kHz/
Post by: putanik on 2007-01-09 09:40:20
I believe that mathematics can give quite precise answers - but that mathematicians are not always the right people to ask the question that is to be answered ;-)

please do not tell that to other mathematicians - so many of us believe that we are the closest to truth under the sun, and they will be so offended by you:-)))

Quote
That is exactly the same statement that I frequently get from the audiophile crowd when they dont want critical questions about their magic pyramids, green markers, voodo-power-cables etc.


oops... sorry. yes, you are right, my degree of self-confidence [in this case, in my hearing abilities] is not a proof for you. but ... i trust and will trust my ears, even if the rest of the world calls me moron;-)))

Quote
Is that for you as a subjective listener in a blind experiment, for you as a listening conusmer, or as a mathematical analysis of the dsp that occurs in dsd vs lpcm systems?


as a mathematician. SACD/DVD-A differences are minor, effect unsure.
Title: Why 24bit/48kHz/96kHz/
Post by: knutinh on 2007-01-09 09:50:03
as a mathematician. SACD/DVD-A differences are minor, effect unsure.

I agree. The difference between CD, DVD-A and SACD are minor and the effects are unshure.

My view is that a system that try to revolutionise the industry by effectively changing all the rules, should have a clear benefit. Sony and Philips are asking everyone (producers as well as consumers) to change all of their gear to gain some obscure benefit. While giving Sony and Philips a new income source after they lost the CD hegemony to Toshiba. They blatantly ignore all of the critical voices from the tech world. "We are stronger than you, our propaganda division will crush any scientific evidents against us, any needed proofs will be made up if we cant find them, either you are with us or against us"

Kind of reminds me of a certain other "CEO"...

(http://avforum.no/forum/attachments/generelt-audio-hifi/13507d1161780469-sacd-vs-dvd-dsd-vs-pcm-commercial.jpg)


A LPCM-system on the other hand, while its benefits are dubious, will easily integrate with current production gear. Also, even DVD-V is able to take full benefit of the current AD and DA technology and then some. I agree with the notion that whatever is needed, the world will move forwards. I dont know if 3d gui menus and transparent windows makes me more productive at work, but Windows Vista will be that way, and this will benefit hardware producers and the circle keeps spinning.

-k
Title: Why 24bit/48kHz/96kHz/
Post by: ImAlive on 2007-01-09 14:03:05
(http://avforum.no/forum/attachments/generelt-audio-hifi/13507d1161780469-sacd-vs-dvd-dsd-vs-pcm-commercial.jpg)

The catch about this diagram is that, due to the way our hearing works, the neat 10k square-wave will sound exactly the same as the 10k sine. If you could hook an oscilloscope at the acoustic nerve, you would get exactly the sine, no matter which signal was fed. There is no way for energy beyond 20kHz to be coupled into the 'human sum signal', because there are no perception cells for above this frequency.

But yeah, in terms of signal processing, the 10k square is reproduced more exactly. It's just that a human will hear a 10k square as a 10k sine.

The filters needed for 44k can be a problem, as this is the point where manufacturers tend to 'save money', especially with crappy soundcards (non-flat frequency response, ripple, and aliasing may be the consequences, but even this will be near unaudible).
Title: Why 24bit/48kHz/96kHz/
Post by: Axon on 2007-01-09 14:09:38
I do not understand your question. At the moment, I am not aware of any A2D/D2A having more that 19 TRUE bits. 20 bits -> 122 dB ~ .7e-4% IMD/linearity error. nobody claimed it - yet. AKM's AK4394 claims only 100 dB THD, and that's quite typical now. Could you please explain yourself?


Lavry AD122-96MKIII claims 126dB THD (0.00005%), and 127dB SNR unweighted.
Title: Why 24bit/48kHz/96kHz/
Post by: knutinh on 2007-01-09 15:09:58
I do not understand your question. At the moment, I am not aware of any A2D/D2A having more that 19 TRUE bits. 20 bits -> 122 dB ~ .7e-4% IMD/linearity error. nobody claimed it - yet. AKM's AK4394 claims only 100 dB THD, and that's quite typical now. Could you please explain yourself?


Lavry AD122-96MKIII claims 126dB THD (0.00005%), and 127dB SNR unweighted.

I am not into current DAC specs. Is this done using a single chip, or stacking several?

-k
Title: Why 24bit/48kHz/96kHz/
Post by: Axon on 2007-01-09 16:06:03

I do not understand your question. At the moment, I am not aware of any A2D/D2A having more that 19 TRUE bits. 20 bits -> 122 dB ~ .7e-4% IMD/linearity error. nobody claimed it - yet. AKM's AK4394 claims only 100 dB THD, and that's quite typical now. Could you please explain yourself?


Lavry AD122-96MKIII claims 126dB THD (0.00005%), and 127dB SNR unweighted.

I am not into current DAC specs. Is this done using a single chip, or stacking several?

-k

I think that unit is all-discrete, actually.
Title: Why 24bit/48kHz/96kHz/
Post by: Axon on 2007-01-09 16:22:11
As a pragmatic mathematician ... I can't imagine a reason why anybody would spend his/her time on altering CD version vs stereo SACD, especially in classical music, where profits are 0.

If SACD really has absolutely no benefit to audio quality, as I believe, then the only way to show a difference to the customer is to cripple the redbook mix. This has, in fact, happened in the past - see Stereophile's analysis of the DSOTM SACD release. The RB layer was vastly inferior to both the SACD layer and an early RB release from a few years back.
Title: Why 24bit/48kHz/96kHz/
Post by: Axon on 2007-01-09 16:56:42
My gut feeling led by my experience is different. AES-type ABX testing was many time laughed at, for example, see http://www.stereophile.com/features/113/ (http://www.stereophile.com/features/113/)

Stereophile, of course, being the center of academic and thoughtful debate on the matter. Seriously, can you do any better than that? At least point to an academic paper newer than Lipshitz.

Quote
but ... why don't you walk into a hi-end audio place and ask them to show you the best speakers on a sample SACD? spend there an hour, and then we'll see if you gut feeling changes. or, have you already been there?

I've been to a hifi store twice: once to audition turntables (and buy one), and once for a Head-Fi meet. I can confidently say that in both situations, hires (either via vinyl or SACD/DVD-A) didn't do anything at all special for me, compared to redbook. So your experience runs contrary to my own.
Title: Why 24bit/48kHz/96kHz/
Post by: putanik on 2007-01-09 20:17:16
I do not understand your question. At the moment, I am not aware of any A2D/D2A having more that 19 TRUE bits. 20 bits -> 122 dB ~ .7e-4% IMD/linearity error. nobody claimed it - yet. AKM's AK4394 claims only 100 dB THD, and that's quite typical now. Could you please explain yourself?


Lavry AD122-96MKIII claims 126dB THD (0.00005%), and 127dB SNR unweighted.


very interesting... do they have data sheet detailed description how and what they measured, like BurrBrown has for pcm4202 http://focus.ti.com/lit/ds/symlink/pcm4202.pdf (http://focus.ti.com/lit/ds/symlink/pcm4202.pdf) and pcm1792a http://focus.ti.com/lit/ds/symlink/pcm1792a.pdf (http://focus.ti.com/lit/ds/symlink/pcm1792a.pdf)? a minor problem ... front sheet for those specifies "marketing" specs of 118/132 dB dynamic range, but when you look inside, you find numbers like 0.0004% (-107 dB) and -105 for THD+N (which both mean 17.5...18 true bits). I have no problems coming up to our CFO asking for 1792 /4202 EVMs so that I can analyze them myself in whatever details i please... but I can't come to him asking for this Larvy ADC bearing pricetag of $7500.
Title: Why 24bit/48kHz/96kHz/
Post by: Axon on 2007-01-09 20:27:33
I'm sure Dan will candidly answer all the technical questions posed for him by posting on his forums.
Title: Why 24bit/48kHz/96kHz/
Post by: krabapple on 2007-01-09 20:32:33
I guess this has been said before, but what the heck... Along the chain of sound propagation:

1. I do agree that many sound sources produce a good deal of energy at infra-sound frequencies. This is a necessary condition for hirez audio to make sense, but not enough on its own.


Infra actually means 'below' or 'low' , so you probably mean 'ultra'.
Title: Why 24bit/48kHz/96kHz/
Post by: saratoga on 2007-01-09 20:36:33

I do not understand your question. At the moment, I am not aware of any A2D/D2A having more that 19 TRUE bits. 20 bits -> 122 dB ~ .7e-4% IMD/linearity error. nobody claimed it - yet. AKM's AK4394 claims only 100 dB THD, and that's quite typical now. Could you please explain yourself?


Lavry AD122-96MKIII claims 126dB THD (0.00005%), and 127dB SNR unweighted.


very interesting... do they have data sheet detailed description how and what they measured, like BurrBrown has for pcm4202 http://focus.ti.com/lit/ds/symlink/pcm4202.pdf (http://focus.ti.com/lit/ds/symlink/pcm4202.pdf) and pcm1792a http://focus.ti.com/lit/ds/symlink/pcm1792a.pdf (http://focus.ti.com/lit/ds/symlink/pcm1792a.pdf)? a minor problem ... front sheet for those specifies "marketing" specs of 118/132 dB dynamic range, but when you look inside, you find numbers like 0.0004% (-107 dB) and -105 for THD+N (which both mean 17.5...18 true bits). I have no problems coming up to our CFO asking for 1792 /4202 EVMs so that I can analyze them myself in whatever details i please... but I can't come to him asking for this Larvy ADC bearing pricetag of $7500.


NI also makes industrial ADCs that are rated for >120dB over the entire 0 to .45 fs range.  They're a pleasure to work with.
Title: Why 24bit/48kHz/96kHz/
Post by: krabapple on 2007-01-09 20:36:46
Since the only high-res formats I am aware of (SACD & DVD-Audio) have iron chains everywhere---I will stick with redbook CD for the forseeable future.


There are two new ones -- Dolby True HD lossless and DTS HD.  You can bet they are chained. 
Title: Why 24bit/48kHz/96kHz/
Post by: putanik on 2007-01-09 20:39:04
I agree. The difference between CD, DVD-A and SACD are minor and the effects are unshure.

My view is that a system that try to revolutionise the industry by effectively changing all the rules, should have a clear benefit. Sony and Philips are asking everyone (producers as well as consumers) to change all of their gear to gain some obscure benefit. While giving Sony and Philips a new income source after they lost the CD hegemony to Toshiba. They blatantly ignore all of the critical voices from the tech world. "We are stronger than you, our propaganda division will crush any scientific evidents against us, any needed proofs will be made up if we cant find them, either you are with us or against us"

Kind of reminds me of a certain other "CEO"...


1). SACD vs DVDA. look at http://focus.ti.com/lit/ds/symlink/pcm1792a.pdf (http://focus.ti.com/lit/ds/symlink/pcm1792a.pdf). this is a typical modern DAC. supports both PCM-24 and DSD. look at the fig 19 (PCM) ans 22 (DSD). i fail to see why DSD is better than PCM. what do you think? re CD: we are going in circles.

2) imho, recording industry does not consists of idiots, nor sony/phillips. toshiba has been a part of both SACD and DVD-A consortium's, if i am not mistaken, as well as sony/etc.

best regards,
putanik.
Title: Why 24bit/48kHz/96kHz/
Post by: krabapple on 2007-01-09 20:40:50
Dear K, most people will not distinguish the sound of the music sampled at 192 and it, filtered with 12 kHz LPF.


Yes, but some people can, and can document the results with an ABX test.  That's a key difference.
Title: Why 24bit/48kHz/96kHz/
Post by: putanik on 2007-01-09 20:52:51
NI also makes industrial ADCs that are rated for >120dB over the entire 0 to .45 fs range.  They're a pleasure to work with.


which one? i am not very familiar with NI products. I found pxi4461 and alike. yes, they claim 120 dynamic range, but the same -107 dB for THD.
Title: Why 24bit/48kHz/96kHz/
Post by: Woodinville on 2007-01-09 23:57:29
Simply that, all other things being equal, a 96/24 recording is closer to the analog source than 44.1/16.


How are they any different, if the total dynamic range of the input is limited to 90dB, and the bandwidth in a practical sense to 15kHz to 20kHz by the capture arrangements?

o-o.


Sorry, that's not an answer.

Now, what's the noise level of the actual, STP atmosphere, at your ear drum. Let's start there, ok?
Title: Why 24bit/48kHz/96kHz/
Post by: putanik on 2007-01-10 02:24:11
Sorry, that's not an answer.
Now, what's the noise level of the actual, STP atmosphere, at your ear drum. Let's start there, ok?


Dear Woodinville, you are asking good questions - but not here, please. I am already tired of going in circles on this forum. sometimes i come south to seattle area to meet friends. whereabout can we meet there? or, do you came to vancouver often? I'll be glad to meet you and discuss things in a normal way.

best regards,
icassp1996@yahoo.ca

PS> btw, are you with mackie, headquartered in "you"? :-)))
Title: Why 24bit/48kHz/96kHz/
Post by: knutinh on 2007-01-10 10:18:18
The catch about this diagram is that, due to the way our hearing works, the neat 10k square-wave will sound exactly the same as the 10k sine. If you could hook an oscilloscope at the acoustic nerve, you would get exactly the sine, no matter which signal was fed. There is no way for energy beyond 20kHz to be coupled into the 'human sum signal', because there are no perception cells for above this frequency.

But yeah, in terms of signal processing, the 10k square is reproduced more exactly. It's just that a human will hear a 10k square as a 10k sine.

The filters needed for 44k can be a problem, as this is the point where manufacturers tend to 'save money', especially with crappy soundcards (non-flat frequency response, ripple, and aliasing may be the consequences, but even this will be near unaudible).

My reason for showing this diagram was to show how Sony/Philips was counting on people being simple-minded.

The funny thing is, those that claim that measurements does not count, only perceptual experience... Those are the first to claim "hah! look at that diagram, SACD can reproduce 0.00000001myS rise-time" :-)

-k


I guess this has been said before, but what the heck... Along the chain of sound propagation:

1. I do agree that many sound sources produce a good deal of energy at infra-sound frequencies. This is a necessary condition for hirez audio to make sense, but not enough on its own.


Infra actually means 'below' or 'low' , so you probably mean 'ultra'.

Ahh.. :-)

-k
Title: Why 24bit/48kHz/96kHz/
Post by: m0rbidini on 2007-01-10 12:23:06
Quote
I am already tired of going in circles on this forum.


You're not the only one. Really.
Title: Why 24bit/48kHz/96kHz/
Post by: 2Bdecided on 2007-01-10 13:28:00
Quote
I am already tired of going in circles on this forum.


You're not the only one. Really.


Well I haven't been here since before Christmas, so can I circle just one more time please?


You haven't said anything that hasn't been said before putanik. Let's be bluntly clear, there are two dead simple questions...

1. Is there an audible difference between 44.1kHz/16bits and "higher resolution" formats in a fair test?
2. If so, why?

This is Hydrogen Audio. If you're not willing to accept ABX tests as a way of determining whether a difference is audible or not, you probably shouldn't have joined!

An ABX test of this possible difference is dead easy: Take a hi-res master, convert it to 44.1/16, and convert it back to the orginal format. Then simply compare the original with the double conversion.


You said, very provocatively, that the difference was so obvious that it wasn't worth ABXing. What a silly thing to say. If it was possible (never mind easy - just possible!) to reliably ABX the difference with a test budget of, say, $1,000,000, don't you think Sony or Philips would have done so, and published the results?


(I do have MATLAB, that's not a great example you provided with 0.84dB total passband ripple! I took the point anyway...)

You made a sensible suggestion about IMD and ultrasonic ringing. That's clearly possible with some content, and has been proposed already. It's more of a problem with bad anti-image filters, but the ringing of an ideal filter could have an audible impact if the following equipment adds distortion.

_If_ a difference was detected in a fair test, I think this would be the most likely explanation. _If_ a difference was detected in a fair test, you could investigate this by changing the transducers. IMD is easily measurable, and you could see if it correlated with the audibility of the difference.


These are all great, real world experiments that could be carried out. Yet hi-fi magazines are too scared of revealing the Emperor's new clothes to jump in and make some sense of the explanation.


Two other points in recent pages caught my eye...

The first was "if we can capture and deliver audio at 24/96 (or whatever) why not?". Why not go to the effort of doing something that (may) make absolutely no audible improvement? Because time, money, effort, marketing, and consumer cash could be better expended on things with real benefits! It's a distraction. Just like most other things in the audiofool world.

To be clear: if we were talking about 24/96 _and_ great multichannel (for example), I'd be all for it. Yet largely we're talking about 24/96 stereo or really mediocre 5.1. It's no wonder consumers aren't interested.


The second was "people can feel a difference (with hi-res), just like 12kHz low pass, so don't dismiss it". People don't "feel" a difference with a 12kHz low pass filter. They either don't hear a difference (due to age or hearing damage) or they do hear a difference. If they don't have the vocabulary to express what that difference is, they may, just, say that the music "feels" different - but the reality is a very clear audible difference which can be picked up in an ABX test, even by those without the language to explain what it is.

This is a world away from the placebo "feels" different experience which vanish as soon as blind testing is invoked.


Finally, you might like this post...

http://www.hydrogenaudio.org/forums/index....ost&p=96338 (http://www.hydrogenaudio.org/forums/index.php?s=&showtopic=9311&view=findpost&p=96338)

...where you see I'm not anti-hi-res - I'm just (increasingly) frustrated at the lack of ABX evidence and scientific explanation. And the (I believe related) decline in the "hi-fi" audio industry as a whole. The part of the industry that concentrates on difference that people can actually perceive (today: convenience!) is doing a damn site better than the part of the industry that concentrates on what people have to image for themselves. As someone who loves music, and recorded music, I think it's a crying shame that the hi-fi industry is all but dead, and the chances of the quality reproduced music increasing further in my lifetime seem to have died with it.

I don't blame the 90% of people who don't give a damn. I blame the snake oil salesmen who have all but taken over the industry and driven away the 10% of people who care about what they listen to.

Cheers,
David.
Title: Why 24bit/48kHz/96kHz/
Post by: Light-Fire on 2007-01-10 17:29:01




...SACD vs. CD difference is so apparent that I smile at requests of measuring it with statistics...


Dear putanik

How do you know the contents of the SACD and CD compared were originally the same and not treated differently in the studio?

If you can't be sure about that your test is not valid.

...As a pragmatic mathematician ... I can't imagine a reason why anybody would spend his/her time on altering CD version vs stereo SACD, especially in classical music, where profits are 0.


I can think of a couple:

Pushing a new format where you have copy protection.

Pushing a new format that people will believe has a better audible quality and will be encouraged to upgrade their collections.
Title: Why 24bit/48kHz/96kHz/
Post by: PKI on 2007-01-10 17:37:40
Is feeling the music a placebo response?

Part of the the original question, '...is there any real need for the higher bit depth and sampling rate?', could be answered in how & what we do 'feel' about the music we're listening to and not just hearing. Those hairs on the back of you neck and arms standing up aren't just down to how the music is encoded, compressed, digitised or even the lyrics and harmonies etc... it could be down to those impercetpible aritfacts that do that 'cetain something' to the music. Call it the X-factor if you will.

This is something that can't be quantified, measured, analysed or ABX'd. Maybe those inaudible higher frequency artifacts and harmonics are subtly vibrating & affecting parts that simply make you feel good.

Don't get me wrong, I like detailed technical analysis as much as the next man or woman, but I believe we're trying to analyse this to the nth degree where analysis can't or doesn't work. Could the extended dynamic range, depth or colouring that comes with higher order encoding just make it 'feel' good?

Surely this is the beauty of music that simply can't be put on a chart. But puts a smile on your face...

Thanks for reading my first post after several motnhs of 'lurking'.
Title: Why 24bit/48kHz/96kHz/
Post by: pepoluan on 2007-01-10 18:11:16
Is feeling the music a placebo response?

Part of the the original question, '...is there any real need for the higher bit depth and sampling rate?', could be answered in how & what we do 'feel' about the music we're listening to and not just hearing. Those hairs on the back of you neck and arms standing up aren't just down to how the music is encoded, compressed, digitised or even the lyrics and harmonies etc... it could be down to those impercetpible aritfacts that do that 'cetain something' to the music. Call it the X-factor if you will.

This is something that can't be quantified, measured, analysed or ABX'd. Maybe those inaudible higher frequency artifacts and harmonics are subtly vibrating & affecting parts that simply make you feel good.

Don't get me wrong, I like detailed technical analysis as much as the next man or woman, but I believe we're trying to analyse this to the nth degree where analysis can't or doesn't work. Could the extended dynamic range, depth or colouring that comes with higher order encoding just make it 'feel' good?

Surely this is the beauty of music that simply can't be put on a chart. But puts a smile on your face...

Thanks for reading my first post after several motnhs of 'lurking'.
Neurologists will say: Your 'feelings' depend on the inputs of your sensory organs and your previous memory(-ies) of the situation involved.

So, with regards to 'feeling a music', there are some conjectures.

* If you can't hear, you can't 'feel' music.
* If you can 'feel' music, you can hear.
* If you 'feel differently' for a music, then you *are* hearing differently.

Now, those "inaudible high frequency artifacts" you're saying... if they are inaudible, by the previous conjectures they are also incapable of inducing any 'feeling' in you.

Other possible sources for 'feeling': vision, memory, and ambient emotion (i.e. emotions you are having while you're listening to a music).

For instance, listening to "I Swear" gives me warm feeling... as it is the 'theme song' of my relation with my very first girlfriend back in highschool.

Some songs did sound 'warmer' (whatever that's supposed to mean) when I listened to them while my current girlfriend was sitting right beside me, reading her book, leaning on my shoulder.

Some songs may sound 'scarier' when played alongside a horror movie (though I have to admit I have lost all sense of 'scare' nowadays. I'm not a fun guy to take to a horror movie... OTOH girls will cling to me hiding their faces behind my arm, while I just sit straight-backed there, being cool  )
Title: Why 24bit/48kHz/96kHz/
Post by: PKI on 2007-01-10 18:42:22
Now, those "inaudible high frequency artifacts" you're saying... if they are inaudible, by the previous conjectures they are also incapable of inducing any 'feeling' in you.


I'm aware that this is a scientific forum so I'm going to stay away from the tenuous 'I can feel something' argument, but we are talking about sonic resonances that may be having an affect on certain people, that when mixed with other stimuli you mention (memory, smell, touch etc), creates that feeling.

Just because we can't hear these (super, infra) artifacts, it doesn't mean that they're not there playing their part in the overall signature of what you're experiencing.
Title: Why 24bit/48kHz/96kHz/
Post by: Woodinville on 2007-01-10 19:20:25

Sorry, that's not an answer.
Now, what's the noise level of the actual, STP atmosphere, at your ear drum. Let's start there, ok?


Dear Woodinville, you are asking good questions - but not here, please. I am already tired of going in circles on this forum. sometimes i come south to seattle area to meet friends. whereabout can we meet there? or, do you came to vancouver often? I'll be glad to meet you and discuss things in a normal way.

best regards,
icassp1996@yahoo.ca



Dear Putanik, the questions I'm asking you are very germane to what you could actually experience in a real venue.

The answers are actually rather surprising.

And, no, I don't work for Loud Technologies.
Title: Why 24bit/48kHz/96kHz/
Post by: pepoluan on 2007-01-10 20:00:51
Now, those "inaudible high frequency artifacts" you're saying... if they are inaudible, by the previous conjectures they are also incapable of inducing any 'feeling' in you.
I'm aware that this is a scientific forum so I'm going to stay away from the tenuous 'I can feel something' argument, but we are talking about sonic resonances that may be having an affect on certain people, that when mixed with other stimuli you mention (memory, smell, touch etc), creates that feeling.

Just because we can't hear these (super, infra) artifacts, it doesn't mean that they're not there playing their part in the overall signature of what you're experiencing.
Since we are arguing the difference of a specific method of sampling to audio quality, all other stimuli except the audio itself must be removed. Unfortunately, we can't really remove other stimuli (e.g. smell; suppose my gf juuuuust making me a nice pancake and I smell the nice aroma wafting through the door). Hence, ABX. It randomizes the hidden track. Do it often enough (e.g. 16 run) and the effect of other stimuli is averaged, and for practical reasons nullified. (e.g. I may start smelling the smell of pancake from run #5 and onwards, thus affecting the rest of the session uniformly). Furthermore, the concentration necessary to carry out the ABX test might mask inputs from other stimuli.

Now let's see. The difference between higher sampling rates will be only the existence of ultrasonic frequencies in the recording. These frequencies will be pumped out via either (1) headphones or (2) speaker sets.

With (1), the energy pumped into your ears are low enough that your ear's skin will not be able to discern them. Thus, only the sound received by your ears is registered as stimulus. Your cochlea is finely tuned by nature to discern only up to 20kHz, and most people can only discern up to 16kHz. All higher frequencies are lost.

With (2), the ultrasonic frequencies can be regenerated only by the small tweeters. The larger midranges and woofers (and also subwoofers) just don't have the physical characteristics required to reproduce them. Further, such multispeaker setup uses crossovers (either passive or active), ensuring that higher frequencies go into the tweeters only. Due to its size, there is just not enough energy coming out of the tweeters to induce other stimuli other than hearing.

So as you can see, from a practical P.O.V., there is no way for ultrasonic frequencies to affect your 'feeling', as sound waves can only be detected as input stimulus via the ears, which already perform a physical (not to mention physiological) Low Pass Filter.

On a lighter note, I think I can intuit why 96kHz sampling rate might give me a *worse* experience than a 44.1kHz sampling rate: My dogs may start howling or barking madly, thus totally ruining my audio enjoyment
Title: Why 24bit/48kHz/96kHz/
Post by: benski on 2007-01-10 20:34:34
In my opinion, the fact that even modern recording and playback gear tends to filter out frequencies above 20khz renders this debate useless.  Most of the 96khz tracks I've seen have steep fallout above 20khz, and the shape of the curve suggests filtering (mechanical filtering from the microphone, analog filtering from the circuits).

Perhaps a testing using Earthworks mics (they claim (http://www.earthworksaudio.com/26.html) flat frequency response up to 50khz) and their corresponding mic pre-amp recording the cymbals or hats of a drum set (which have significant energy past 20khz) and listening on a good set of headphones (beyerdynamics dt 880 claims (http://www.beyerdynamic.com/cms/download.php?filename=DT880PRO_DB_E.pdf) response to 35khz) would be a better test than some 96khz "remaster" that was likely recorded with SM57's and condensor vocal mic
Title: Why 24bit/48kHz/96kHz/
Post by: crimsontide on 2007-01-10 23:54:56
This is something that can't be quantified, measured, analysed or ABX'd. Maybe those inaudible higher frequency artifacts and harmonics are subtly vibrating & affecting parts that simply make you feel good.

Don't get me wrong, I like detailed technical analysis as much as the next man or woman, but I believe we're trying to analyse this to the nth degree where analysis can't or doesn't work. Could the extended dynamic range, depth or colouring that comes with higher order encoding just make it 'feel' good?

Surely this is the beauty of music that simply can't be put on a chart. But puts a smile on your face...

Thanks for reading my first post after several motnhs of 'lurking'.



I was posting earlier in this thread but have been lurking ever since.

I agree entirely.

Without meaning any disrespect, nor to direct this at anyone in particular, the whole point of trying to scientifically examine the listening of a musical recording is like trying to digitally sample the aesthetics (for want of a better word) of a good mature glass of wine.

Those of you who appreciate wine will know that this simply cannot be done, and computer equipment must be at least 3 decades to a century short on development for this particular application (wine).

While the wine experience has more dimensions than audio frequencies, the audio frequencies interact in ways which are far more than simply mathematical in the same way as the flavour of a wine develops as your buds respond etc.....

I should know, I have a PRS (guitar), with a 1980-something marshall guvnor, running through a JCM900 valve head, all plugged into a 1960 Marshall 4*12 which i bought off Genesis' ex-roadie, who "acquired" it from the band.

And i can tell you, no amount of technical jiggery pokery can reproduce that sound, nor explain it, not the best mic or the most powerful amplifier - or the flattest speaker response. As of 2007 - it still cant be done. ABX is not the be all and end all of audio comparisons. To form an analogy, ABX is to sonic comparison, what a postcode is to an address - an engineering necessity, but with wholly inaccurate results.

It gets a general result, but does it actually answer the question....? I think its time to chat with some audiophiles, who would concur with my wine analogy.....again no disrespect, i know this is an engineers forum.......but science, is now. The answer, comes from tomorrow. Tomorrows engineers break the rules, break new ground and create the new rules......
Title: Why 24bit/48kHz/96kHz/
Post by: Mercurio on 2007-01-11 00:08:12
Just because we can't hear these (super, infra) artifacts, it doesn't mean that they're not there playing their part in the overall signature of what you're experiencing.


If you experience something, you can measure it, at least qualitatively 
I don't understand why this something can't be abxed. If you "feel" better (happier, colder or whatever), then there should be some effect, and you should be able to test and reproduce them... maybe for some subconscious effects it is more difficult, and it requires a bit of statistic, and maybe is impossible to measure the effects of listening "hi-res" audio to the destiny of the soul after the death...

Luckily we are speaking about a way to store the information carried by an electrical signal (this is what the sound becomes after a mic) and I don't think this can have any religious implication.

Do you think you have a better experience when you listen hi-res audio, PKG? Why do you think it is an effect of 24 bit/96khz? Maybe it can be an effect of the light in your room ^^

btw I can't help in this discussion, since my not-so-untrained ears can't listen any difference (feeling or whatever) between a 44khz and a 32khz sampled song  . I want to try again that test with better equipment, and then, maybe, I will start to think about 96khz and other mythical resolutions.
Title: Why 24bit/48kHz/96kHz/
Post by: Woodinville on 2007-01-11 00:29:57
Without meaning any disrespect, nor to direct this at anyone in particular, the whole point of trying to scientifically examine the listening of a musical recording is like trying to digitally sample the aesthetics (for want of a better word) of a good mature glass of wine.

Your example is improper. We drink the wine, there is nothing to encode. There is no transmission method, unless you include the bottle and the glass.

And, we all know that formal wine tastings are done blind, so we can't see the bottle, and the glasses are all the same.

"digitally sample the aesthetics of ... wine" is a meaningless statement.

Whereas, in audio, we record the audio, be it digitally or not, for later delivery.  When we record it, we radically change the aesthetic value just by the act of recording it into a few channels.
Quote
Those of you who appreciate wine will know that this simply cannot be done, and computer equipment must be at least 3 decades to a century short on development for this particular application (wine).

Your argument is completely invalid. Wine is not conveyed by computer. Music is.

What's more, we can analyze wine, we can tell what flavour constituants go with what chemicals, we can tell what aging does, we can (and do) do various kinds of analysis from vapor spectroscopy through mass spectrometry that can directly guide the knowledgeble modern winemaker as to what kind of product they are making, how the aging is going, and so on.

This kind of analysis is testable, verifiable, material, and repeatable.

In audio, we do not have any verification of the alleged artifacts in the first place, so:

1) We can not repeat the test, since we have not yet verified it.
2) We have no evidence of a material difference
3) We have yet to even understand a way to verify the alleged phenominon.
Quote
While the wine experience has more dimensions than audio frequencies, the audio frequencies interact in ways which are far more than simply mathematical in the same way as the flavour of a wine develops as your buds respond etc.....

None the less, how wine is tasted, how it tastes, etc, is in fact reduced to a rather analytical science. The "target taste" varies from place to place, simply because different people prefer different tastes, hence we have a Hedges 3-vineyard, a Terra-Blanca Cab, and a DeLille Harrison Hill, all of which are different, acknowledged to be different, and built to different tastes.
Quote
I should know, I have a PRS (guitar), with a 1980-something marshall guvnor, running through a JCM900 valve head, all plugged into a 1960 Marshall 4*12 which i bought off Genesis' ex-roadie, who "acquired" it from the band.

Indeed, and in fact the differences between this and a Stratocaster running through a 1970's Peavey solid state head are easily measured. Your point?
Quote
And i can tell you, no amount of technical jiggery pokery can reproduce that sound, nor explain it, not the best mic or the most powerful amplifier - or the flattest speaker response.

Nonsense. Stuff and nonsense. In fact, the characteristics of that system can be measured to a what-for, but in fact the easiest way to "synthesize" that is probably to build it. Particular kinds of signal-dependent nonlinearities (which is what you're on about) are hard to synthesize. No doubt about that, but they can be measured, evaluated, etc.

ABX testing, done properly, especially used as part of a signal-detection test design, has been shown to provide auditory thresholds right on down to the levels possible from physics. There is no doubt that a proper double-blind test, ABX or not, can detect anything that the human organism can detect.  Adding things like test controls ensures this, as the sensitivity can be verified in-situ.

Something that many people, including musicians for sure, do not realize, is that what is a "subtle" musical effect is actually very large in terms of what human beings can detect.

Sorry, but there is more than a century of work behind this, starting with Helmholtz if not earlier, and the results are plain as day.

Now, people can run an insensitive ABX test. People can drive their car poorly. People can sink boats, and crash airplanes, too, but we don't argue they are ineffective at their purpose because people make mistakes.

But the good news is that if the insensitive ABX test includes proper controls, the insensitivity will stand out like a 1kw search light in an otherwise dark cave.
Title: Why 24bit/48kHz/96kHz/
Post by: putanik on 2007-01-11 08:40:58
This is Hydrogen Audio. If you're not willing to accept ABX tests as a way of determining whether a difference is audible or not, you probably shouldn't have joined!


Dear David, I came in to answer a question of 22049 sine wave perfect reconstruction, and said that this kind of reconstruction exists only on paper. a filter getting from 0 to -98 in 1Hz shall be at least 5s long, what means 44.1^2*1e6*5 Flops ~ 10 G flops, and my gut feeling is that double precision is not enough for all sinewaves from 1hz to 22049 with amplitudes from -80 dBm and up. but people here are completely sure it's a piece of cake. I saw no evidence that anybody having such opinion actually did filters of the kind and is familiar with accompanying difficulties - yet they talk with confidence reminding of an anecdote:"what is the difference between a lawyer and a G-d?". "G-d does not think he is a lawyer".

you are definitely right about "you probably shouldn't have joined"!

that's all, folks.
Title: Why 24bit/48kHz/96kHz/
Post by: knutinh on 2007-01-11 10:19:45
I play the Hammond organ, and I can tell you that it is a great experience. It weighs about 200kg, makes all kinds of noises and smells like old sewing machine oil and hot tubes. Oh, and you need a loan as well as hefty, frequent repairs. No digital simulator has ever given me the same feeling as playing the Hammond organ. Does this mean that those simulators cannot recreate the sound with sufficient accuracy for my ears and abilities? Does that mean that Nyquist, Shannon, ABX etc is all B.S.?

Of course not. Just that we are comparing apples with... something else. The experience of playing that instrument consists of sound as well as all kinds of sensatory and emotional influences. It would be correct to say that no digital simulator can recreate the entire package leading to my positive subjective response. But that is no guarantee that the sound itself is lacking. The sound _could_ be proven to be inferior if an ABX test was done by removing all other influences. If this is not done (properly), then it might as well be that the digital simulators are lacking in simulating burned sewing-machine oil, something that the designers never tried to simulate in the first place.

One could argue that the smell and looks of an instrument never appears on a record and therefore are "luxury" stuff. On the other hand, the musician will probably perform differently if he is satisfied with the sound (even if it is all in his head), and that will most certainly change what is recorded. In much the same way, the listener experience in practical hifi is intertwined with sound, vision, knowledge etc. So why would we want to tell a hifi-listener that his experience is based on superstition and that he cant hear anything at below -80dBFs, or above 22kHz? I feel no need for doing this, but I think that when discussing these matters it is important to get the facts as right as possible.

Most people will state - directly or indirectly that their children are the smartes, prettiest children to ever walk on the face of the earth. Does this make it an undisputable fact? If so, how can every child be "best"? Isnt it a positive thing that we as humans are constructed to be subjective, even if that makes us bad measurement devices? It is just a problem for discussing "the fact" in web forums and courts and a few other fora. For the most important facets of life, "subjectivity" is probably highly beneficial and necessary to lead a good life and have offspring.

-k
Title: Why 24bit/48kHz/96kHz/
Post by: Mercurio on 2007-01-11 11:14:16
Dear David, I came in to answer a question of 22049 sine wave perfect reconstruction, and said that this kind of reconstruction exists only on paper. a filter getting from 0 to -98 in 1Hz shall be at least 5s long, what means 44.1^2*1e6*5 Flops ~ 10 G flops, and my gut feeling is that double precision is not enough for all sinewaves from 1hz to 22049 with amplitudes from -80 dBm and up. but people here are completely sure it's a piece of cake.


I think you made a great service for hydrogenaudio  providing these data, putanik.

However it is more interesting to calculate also how much is the error - and if this affect the sound quality -, not only what you need to have a perfect reconstruction. I think i will start to do it in my spare time ^^

Do you think people need a perfect reconstruction of a wave sampled at 16bit/44khz?

(as I said I can't express opinion about this topic, I'm still trying to discover the limits of 32khz)
Title: Why 24bit/48kHz/96kHz/
Post by: Mercurio on 2007-01-11 12:11:43
I should know, I have a PRS (guitar), with a 1980-something marshall guvnor, running through a JCM900 valve head, all plugged into a 1960 Marshall 4*12 which i bought off Genesis' ex-roadie, who "acquired" it from the band.

And i can tell you, no amount of technical jiggery pokery can reproduce that sound, nor explain it, not the best mic or the most powerful amplifier - or the flattest speaker response.


You are very off-topic crimsontide: the discussion about 24bit etc if a whole different story. Also why these strange 2007' SACD should sound better than the good old CDs? ^^.... nobody can reproduce the feeling of a great, traditional, bad filter used for the 44khz.

[VERY offtopic]
However after reading your post I started to do some interesting (anthropological) thoughts. To understand you, I need to think about some acoustic instrument (a guitar... a violin...) since I have no respect for the electronic sound, while for me acoustic-instruments craftsmen are a sort of wizards, and each instrument is unique and irreproducible -nobody can rebuild a Stradivari-, and it changes and lives each day.

I still could not record the sound and the feeling I have when I play my acoustic and classical guitars. Indeed It seems it is more simple to synthesize them than to record! You can easily check what I said if you try to looking  for the "best" system to amplifying an acoustic guitar to achieving a *natural* sound. They are a sort of synthesizers!

These kind of thoughts are not strange: music was born as a form of magic, and each instrument has a myth  that tells you about a god that invented it. And all craftsmen were ever viewed as wizards. Even a blacksmith, Hephaestus, is a god!

But while I'm attending the courses of electronic engineering, I haven't still seen Harry Potter among my classmates  . Maybe it is the same for a craftsman.

The industrial production has, at least partially, destroyed the myth...

Quote
no amount of technical jiggery pokery can reproduce that sound

Can't you rebuild all your equipment exactly as it is, with the same components? I think there is not "lost knowledge" about it. It was an industrial product. Then I have a doubt: why nobody build it now, in 2007?
Title: Why 24bit/48kHz/96kHz/
Post by: knutinh on 2007-01-11 12:41:22
[Can't you rebuild all your equipment exactly as it is, with the same components? I think there is not "lost knowledge" about it. It was an industrial product. Then I have a doubt: why nobody build it now, in 2007?

I am sure that some part of the music industry is based on faith, just like the hifi industry.

It has been claimed that certain vintages of Fender and Gibson electric guitars sound different from others, and people are trying to build instruments that offer the same qualities in 2007. Who knows what percentage is placebo and what percentage is actual sound/mechanics?

In the case of Hammond organs, at least, its a question of wages. The amount of manual labor needed to build a new one today without capitalising on more advanced (and cost cutting) production methods would ensure that noone could buy it. Back in the 60s people were put i debt to buy Fenders and Gibsons etc here in Europe. Today, you can do it with no sweat...

The question of hifi is a lot simpler because we are building a reproduction chain that isnt supposed to make artistic statements on its own, it is supposed to recreate the soundwaves found in some recording venue (or in the head of some record producer).

-k
Title: Why 24bit/48kHz/96kHz/
Post by: 2Bdecided on 2007-01-11 12:47:14
Mercurio (and crimsontide),

You may have thought you were way off topic, but you're actually very on topic in a way.

There are loads of things about being somewhere which just listening to a recording (even a "perfect" one) cannot re-create. Even if the exact same sound waves hit our ears, all the other senses, experience, and memories could be completely different.

That is interesting, but irrelevant in terms of doing the best we can to reproduce the sound itself.


However, what you said is so important because, all the extra stuff aside, there is lots of information in the sound field of a real acoustic event which isn't captured or reproduced. The same sound waves do not hit our ears. Surely that's a bad thing? Surely we should try to do that at least!

The whole high-res audio debate has concentrated on delivering sound waves to our ears which are more accurate in ways which human ears simply cannot detect, while neglecting to improve the accuracy of the sound waves in ways which almost anyone can detect.

That's what's so tragic. It's not even as if re-creating accurate sound fields is a black art or magic - the science is well known, it's just more difficult that doubling the sampling rate of a DAC! Plus it wouldn't do much for all the old recordings which are just stereo. I'm sure we all realise that any new audio format is partly an excuse for selling us all the same recordings all over again!


The reasons why modern CDs can be so disappointing compared to older releases are well covered here on HA. However, more striking is just how bad any recording (made with conventional microphone layouts) is at reproducing real acoustic instruments in a real space. This is most apparent with instruments which send out different sound waves in different directions, which them bounce around the room and hit the listener from all different directions. We've got used not having this in a recording, but you only have to try to make a realistic recording of a violin or an acoustic guitar to realise that we're dealing in pale, flat imitations. It's no wonder some people like things which distort the sound (e.g. vinyl, valve/tube amplifiers etc) - there is so much lost that people want to put something back in, even if its wrong or artificial. FWIW I don't think 5.1 help much at all, though 6.0 or ambisonics is a step in the right direction.


If ever I have the equipment to perform the test, I want to compare 24/96 stereo with 6.0 driven from 3 256kbps mp3s! I'm willing to bet the latter sounds better.

Cheers,
David.
Title: Why 24bit/48kHz/96kHz/
Post by: Mercurio on 2007-01-11 13:03:40
Mercurio (and crimsontide),
The whole high-res audio debate has concentrated on delivering sound waves to our ears which are more accurate in ways which human ears simply cannot detect, while neglecting to improve the accuracy of the sound waves in ways which almost anyone can detect.


I agree - this is why I smile to the whole discussion. It seems to me 24bit etc are not the weak ring of the audio chain.

I'm just looking for a good (cheap) headphone, and so I'm reading about distortions, strange frequency response and so on, so big that thinking about 24 bit, 130db of  dynamic range etc etc is simply (yet?) beyond my visual.
Title: Why 24bit/48kHz/96kHz/
Post by: knutinh on 2007-01-11 13:15:38
Now we are onto something interesting :-)

Question is, how many users are willing to do the considerable spending and home-redecoration that is necessary for most multi-channel delivery systems?

Why arent dolby etc more focused on delivering general systems that can be "resynthesized" in terms of soundfield, speaker number and placement according to customer needs?

How many technicians will be able to record music in a way that suits such systems?

Even in a perfectly echo-free chamber with many, many speakers rendering the spatial soundfield with very high precision, are we able to fool our brain into analysing this as a room 30 times larger than what our vision is telling us?

-k
Title: Why 24bit/48kHz/96kHz/
Post by: PKI on 2007-01-11 14:08:20
Do you think you have a better experience when you listen hi-res audio, PKG? Why do you think it is an effect of 24 bit/96khz? Maybe it can be an effect of the light in your room ^^

Yes. I've atteneded many auditions where a fair sweep of price range, resolution and sources was used. The environmental variables across this are simply vast. Moreover, as I never took notes of the audition and equipment, I can't make a valid comment on specifics and I''ve never conducted a back-to-back ABX or test of the higher res format.

But what I can say with certaintly is that I've been moved more by the higher end equipment. I currently own some mid-range Sugden valve o/p equipment and the way it reproduces Sara Maclachlan's 'Angel's' sends shivers all over me and the first time I heard it in a friends' audio room moved me to tears.

Lighting, cables, stands, positioning, tuning & the equipment all played their part. All these variables affected (and affect) the reproduction of the standard CD format we heard.

There were probably no artifacts present in this home based situation (reproduction would have changed with a variable being altered) but was my real world affect by the higher end & res audio just down to better reproduction?

This improved reproduction (from some of this equipment) would have added in better transitions and dynamic range, so I would say yes. But that's measurable.

What's catching us out here is our lack of measured evidence to support what output is produced by higher res sources. This obviously must inlcude the effect (or lack of) the reproduction equipment has.


At the moment, it's just a gut feeling. But it's a feeling I like..
Title: Why 24bit/48kHz/96kHz/
Post by: 2Bdecided on 2007-01-11 18:04:13
Even in a perfectly echo-free chamber with many, many speakers rendering the spatial soundfield with very high precision, are we able to fool our brain into analysing this as a room 30 times larger than what our vision is telling us?


I bet it works with your eyes closed!

Until you cough, or just sniff loudly - and then the lack of sound coming back to you makes your brain realise it's all a trick.

FWIW I found normal stereo sounded truly awful in an anechoic chamber.

Cheers,
David.
Title: Why 24bit/48kHz/96kHz/
Post by: PKI on 2007-01-11 19:13:26
I think this has broadened into a seriously good debate.



I find it fascinating that the subject touches people in so many different ways and is really what I love about music and the technology of Audio...
Title: Why 24bit/48kHz/96kHz/
Post by: Woodinville on 2007-01-11 21:08:23
Dear David, I came in to answer a question of 22049 sine wave perfect reconstruction, and said that this kind of reconstruction exists only on paper. a filter getting from 0 to -98 in 1Hz shall be at least 5s long, what means 44.1^2*1e6*5 Flops ~ 10 G flops, and my gut feeling is that double precision is not enough for all sinewaves from 1hz to 22049 with amplitudes from -80 dBm and up. but people here are completely sure it's a piece of cake.



Really now?

I haven't seen anyone say "that's a piece of cake". Could you please provide a citation?

You do notice Redbook CD has a 2.05 kHz transition band, don't you? Nobody tries to put a 22049 Hz signal through a Redbook CD.

So, how is this straw man any sort of contribution at all to the discussion?

Btw, with a 5*44100 tap filter, you'll need about log base 2 of that in extra bits at the bottom of the filter resolution.  Why is this news?  It's old hat.
Title: Why 24bit/48kHz/96kHz/
Post by: crimsontide on 2007-01-11 23:27:45
Can't you rebuild all your equipment exactly as it is, with the same components? I think there is not "lost knowledge" about it. It was an industrial product. Then I have a doubt: why nobody build it now, in 2007?



I concur my wine tasting analogy was probably a bad move.

However I'm not as off topic as you may think. My point is - that even WITH 24bit, 96khz, through the very best and stupidly powerful amplifier, with incredible speakers, it's just not going to sound the same hearing that riff chugging out of those 4 * 12" greenback cones. At present we do not have the technology to make it sound like that through synthesis (i.e. VST plugins etc.) - nor does recording it reproduce this faithfully. The question even stands - can our current hifi speakers reproduce that kind of noise, and vibration, can they really disturb the air in that way?

I know that there will be cries of "microphone" now, and rightly so.....but assuming the micorophone is perfect........one day........perhaps we will get speakers and microphones which make 24bit 96khz worthwhile, even if it cannot be abxed with todays sources.

I agree with you about acoustics too - my dads has a beautiful 12 string which i dearly love to play (although tuning is a right pain!), and it really is this intangible "feeling" which I'm not convinced I've ever heard coming from a digital sample (from any source).

Well thats my last on the matter, I'll probably lurk for another 6 to 12 months now.....I will read any replies however.

I guess I wanted to say, that while abxing the digital samples can't be done at the moment, perhaps if we keep developing this higher resolutionm technology, we will find that once the other weak points in the chain are improved, we actually CAN abx between these audio resolutinos and frequencies. Food for thought anyway. Not wine, but food.
Title: Why 24bit/48kHz/96kHz/
Post by: m0rbidini on 2007-01-12 20:28:33
Quote
Quote
You're not the only one. Really.


Well I haven't been here since before Christmas, so can I circle just one more time please?


I just want to clarify what I meant, cause I'm also not sure if I passed the message correctly. To avoid a somewhat inflammatory comment I contained myself. What I really wanted to say was (in a less dubious way): "I'm also tired of you [putanik] going around in circles". I mean, putanik gets all touchy when his competence is questioned but refuses to provide scientific evidence of discernible difference (ABX tests). He refused to answer simple/concrete questions related to the technical arguments he raised and in the end (before my post) he invites Woodinville to meet and "discuss things in a normal way". I mean, WTF?! Not to sound like a drama queen, but to me this is called disrespecting everyone involved in this discussion, including the ones just reading.

Regarding the topic: no, I don't have reasons to believe that anything more than 44.1 kHz/16bit (properly mastered and dithered) is necessary for listening purposes. On the contrary, from human physiology limits (hard to circumvent) to restrictions on almost all recording and playback devices, everything seems to point that 44.1 kHz/16 bit is enough. Unfortunately, my knowledge is not that great and that's why I enjoy reading threads like this.

Would I have problems with 96 kHz/24bit becoming the minimum choice in generally available consumer formats? No, not really, as long as I could still rip and encode it in whatever way I need/want.
Title: Why 24bit/48kHz/96kHz/
Post by: Woodinville on 2007-01-12 20:52:11
I agree with you about acoustics too - my dads has a beautiful 12 string which i dearly love to play (although tuning is a right pain!), and it really is this intangible "feeling" which I'm not convinced I've ever heard coming from a digital sample (from any source).


(*this applies to the electric issue as well*)

Consider.

When you hold the guitar, you feel the vibration as well as hear it. The guitar has nothing like a uniform radiation pattern, it is different to the two ears, and different to the surroundings that reflect.

There is, in short, a lot more information than can be conveyed in two channels. Even if you use binaural miking, you can't move your head the same way twice (for the binaural miking and the playback) and you wouldn't feel the guitar vibrating in your hands, against your chest, etc.

So there's no way the experience is going to seem similar.

The point, of course, is that all of these effects are way, way above any kind of auditory threshold or sensory threshold, and the effects from 16/44 are, at best, at the very teeny-tiny-barely-perceptable even if they are.

Much information is lost in recordings. That is the major issue with verisimilitude in recordings.  This information is primarly spatial, and is not due to the flaws of the system, except in that the system is only two channel.  Yes, multichannel works better.

FWIW I found normal stereo sounded truly awful in an anechoic chamber.

Cheers,
David.


It's possible to make a recording that does sound good played back in stereo in an anechoic chamber.

It is, however, not a normal recording, and it wouldn't sound so hot in a normal room.

He refused to answer simple/concrete questions related to the technical arguments he raised and in the end (before my post) he invites Woodinville to meet and "discuss things in a normal way".


Well, the question of bit depth is directly related to the peak level one can get from one's system and the noise level of both the room and the atmosphere.  At best, the noise level of the atmosphere at the eardrum is circa-6dB SPL white noise. So 6+96=102, if you listen below 102dB peak level, 16 bits has to suffice.

Most equipment will go somewhat beyond that, though. So you could need a bit more, until you consider the noise level in a good, quiet listening room, which is well above that 6dB number. Even in very good, custom-built commercial isolated settings, 9dB is an extraordinarily good number to achieve.
Title: Why 24bit/48kHz/96kHz/
Post by: Firon on 2007-01-12 21:46:49
Don't forget that with proper dithering and noise shaping you can get some extra dB with 16-bit audio.
Title: Why 24bit/48kHz/96kHz/
Post by: Woodinville on 2007-01-12 21:59:24
Don't forget that with proper dithering and noise shaping you can get some extra dB with 16-bit audio.



At redbook standards, not a lot, though.

Now a 4th order shaper at 64kHz...
Title: Why 24bit/48kHz/96kHz/
Post by: krabapple on 2007-01-13 03:54:28
Quote

...As a pragmatic mathematician ... I can't imagine a reason why anybody would spend his/her time on altering CD version vs stereo SACD, especially in classical music, where profits are 0.


I can think of a couple:

Pushing a new format where you have copy protection.

Pushing a new format that people will believe has a better audible quality and will be encouraged to upgrade their collections.



A third:  making the CD version more listenable (louder) in a noisy (automobile, portable) environment because you can play the CD tracks in a car/discman/ipod, whereas you can't play the SACD layer in any of them (without first recapturing analog output to digital PCM).  Nothing especially 'shady' about that.  It's simply a decision about the market.

Is feeling the music a placebo response?


It can be, if you are 'responding' to something extramusical -- like a subconscious bias that a 'high resoution' format will sound better.




Now, those "inaudible high frequency artifacts" you're saying... if they are inaudible, by the previous conjectures they are also incapable of inducing any 'feeling' in you.


I'm aware that this is a scientific forum so I'm going to stay away from the tenuous 'I can feel something' argument, but we are talking about sonic resonances that may be having an affect on certain people, that when mixed with other stimuli you mention (memory, smell, touch etc), creates that feeling.


We cannot hear 'resonances' above ~22 kHz, nor can we differentiate bitdepths above ~15 bits.  So if the sonic input is already encompassing these limits, then the 'other stimuli' MUST be what is causing the different 'response'.  Not the sound!  Why is this so hard to grasp?


Quote
Just because we can't hear these (super, infra) artifacts, it doesn't mean that they're not there playing their part in the overall signature of what you're experiencing.


Actually, given what is known, it DOES mean that.  Unless you have some good evidence to the contrary.
Title: Why 24bit/48kHz/96kHz/
Post by: krabapple on 2007-01-13 04:07:25
Without meaning any disrespect, nor to direct this at anyone in particular, the whole point of trying to scientifically examine the listening of a musical recording is like trying to digitally sample the aesthetics (for want of a better word) of a good mature glass of wine.


BULL.  Really.  I'm tired of this particular , anti-scientific, perennial non-argument from woo.  I move that its ilk be added to the list of TOS violations for RAHE.


Quote
Those of you who appreciate wine will know that this simply cannot be done,


Bull.  The reality of wine differences CAN BE and ARE routinely validated via blind tests.    THAT is all that we are asking: evidence that differences that are reportedly 'heard' , are really due to the SOUND.

Quote
and computer equipment must be at least 3 decades to a century short on development for this particular application (wine).


Actually wine qualities are already being decoded into chemical correlates.  There's a chemist making very good money selling his chemistry-based recommendations to the vintners of the world. 

Quote
I should know, I have a PRS (guitar), with a 1980-something marshall guvnor, running through a JCM900 valve head, all plugged into a 1960 Marshall 4*12 which i bought off Genesis' ex-roadie, who "acquired" it from the band.

And i can tell you, no amount of technical jiggery pokery can reproduce that sound, nor explain it, not the best mic or the most powerful amplifier - or the flattest speaker response. As of 2007 - it still cant be done.


I'd bet it could actually be simulated pretty well digitally, if someone cared to.

Quote
ABX is not the be all and end all of audio comparisons.


They're a damn sight more reliable than sighted methods.

Quote
To form an analogy, ABX is to sonic comparison, what a postcode is to an address - an engineering necessity, but with wholly inaccurate results.


THat's a rather bad analogy.  Here's another slightly better one: assuming sighted listening is 'accurate' is like depending on someone to ALWAYS write the correct address.

Quote
It gets a general result, but does it actually answer the question....? I think its time to chat with some audiophiles, who would concur with my wine analogy.


Yeah, it is, for you...I suggest you skidaddle to Audio Asylum or some other den of woo, where standards of evidence are considered bad form.
Title: Why 24bit/48kHz/96kHz/
Post by: Light-Fire on 2007-01-13 06:23:23
...Lighting, cables, stands... All these variables affected (and affect) the reproduction of the standard CD format we heard...



This is one of the most absurd things I've ever heard!
Title: Why 24bit/48kHz/96kHz/
Post by: Kees de Visser on 2007-01-13 08:49:17
Quote
Those of you who appreciate wine will know that this simply cannot be done,
Bull.  The reality of wine differences CAN BE and ARE routinely validated via blind tests.    THAT is all that we are asking: evidence that differences that are reportedly 'heard' , are really due to the SOUND.
I agree. It's just a shame that those who can afford the >€100 per glass wines are seldom the ones who can identify them in a blind test
Quote
Actually wine qualities are already being decoded into chemical correlates.  There's a chemist making very good money selling his chemistry-based recommendations to the vintners of the world.
My father-in-law is a wine lover and tried to analyse wine in his chemical lab years ago. That worked very well, but re-synthesizing the wine, using exactly the same components was a great disillusion.
Quote
Quote
ABX is not the be all and end all of audio comparisons.
Perhaps ABX isn't, but double-blind is, IMHO.
Title: Why 24bit/48kHz/96kHz/
Post by: knutinh on 2007-01-13 09:54:08
I think that wine is an excellent example that:
1. Science cannot do everything. Simply analysing and resynthesizing the best wines is either impossible or prohibitively expensive.

2. The above does not matter in establishing "facts" as we can simply do a blind test to find the best wine.



It astonishes me that presumably intelligent audiophiles keep hammering that "science is not good enough", "the ear has more resolution than any measurement", when logically, that does not matter. What matters is that the moment you remove the info on what is playing, the audiophile cannot "hear" the improvements reported by using hoax cables, green marker, and hirez audio.

The single, intelligent model for explaining the above is that humans are subjective devices that responds to other stimuli as well as sound. Once agreed upon, it is evident that you cannot take the claims from a sighted test ("improved soundstage") to mean anything about the sound alone. It can only be seen as a total subjective rating on anything from looks, tactile feel, marketing etc.

-k
Title: Why 24bit/48kHz/96kHz/
Post by: knutinh on 2007-01-13 10:08:41
From a socialogical (?) perspective, there is one thing that astonishes me:
Belonging to a "objectivist" school of thought myself, I have no strong feelings about the capabilities of human hearing. We might have really bad capabilities, or we might be able to circumvent any measurement devices. What matters to me is that in all probability, it is limited. Just because most things in nature are limited, the number of processing units in our brain is limited etc.

But for those that subscribe to the "subjective" school of thought, it seems like there is a mission to actively suppress any info that counters their initial believe. That there is no need for good arguements because the truth should be self-evident. In fact, if you are trying to make an arguement, then you are a cold, life-less engineer not able to appreciate the music. Oh, and by the way, that music "flows" better in a 96kHz path because the electrone-particles are smaller and therefore more analog-like :-)

I actually think it would be very cool if the hearing was proven to be more than the scientists have proven so far. Interesting. But as long as the alternative/subjectivist camp is incapable of providing coherent lines of thought that would pass the scruitiny of a 5-year-old, I suspect that any such revelations will appear from elsewhere.

This is sad because it is always more interesting to discuss with smart people than angry people, even if you disagree on everthing.

-k
Title: Why 24bit/48kHz/96kHz/
Post by: crimsontide on 2007-01-13 10:49:32
I'd bet it could actually be simulated pretty well digitally, if someone cared to.


Well if you had any knowledge, experience or indeed any valid opinion whatsoever on this subject you'd be well aware that this is not possible yet. Have you ever actually heard a combination of equipment like this in real life? Have you ever tried using that ridiculous cabinet simulation equipment, or VST valve emulators?

I know the answer - NO.

because they all sound absolutely rubbish - tried pretty much all of them, as my friend works in turnkey in london.

I'm not sure you really know what a guitar sound should really be like... so you're using a logic gate of blind comparisons to work out whats best.....it's ridiculous. What happened to flavour......my point was that you really can't record that, and if you COULD, maybe we would be able to ABX it, its just that the microcphones can't record the raw power, and miss a lot of texture.

Quote
Yeah, it is, for you...I suggest you skidaddle to Audio Asylum or some other den of woo, where standards of evidence are considered bad form.


you failed to provide any evidence yourself back there......

I'm beginning to think this forum is full of pale skinned sun-dodgers who've never left their house but read a lot of forum bullshit, and all concur, since it is the latest scientific opinion in their favourite circles of pseudo social groups, that in the fact the earth IS flat, because noone has provided any "evidence" that it is is in fact round.

I'm suggesting there are other possibilities, since you missed my second post to follow up the one that just tickled your hate buds, you don't seem the kind of person to cover all angles anyway.

This forum appears to be the audio-scientific equivalent of the Arian Brotherhood, except, ain't nobody got your back.

There really is no discussion here....just fascism.

I'm pleased to say you dont have to add me to any list, because I am already gone.
Title: Why 24bit/48kHz/96kHz/
Post by: [JAZ] on 2007-01-13 11:03:58
I vote for this topic to be closed.

The nonsense starts to be too high.  And there has been no proove that high-resolution is needed, so point closed.

Just stupid references to simulators..
Title: Why 24bit/48kHz/96kHz/
Post by: ImAlive on 2007-01-13 13:03:20
This forum appears to be the audio-scientific equivalent of the Arian Brotherhood, except, ain't nobody got your back.

There really is no discussion here....just fascism.

YMMD! 

Since this thread is growing long I guess it was about time, right? All hail Godwin (http://en.wikipedia.org/wiki/Godwin%27s_Law)!
Title: Why 24bit/48kHz/96kHz/
Post by: bug80 on 2007-01-13 15:09:23

I'd bet it could actually be simulated pretty well digitally, if someone cared to.


Well if you had any knowledge, experience or indeed any valid opinion whatsoever on this subject you'd be well aware that this is not possible yet. Have you ever actually heard a combination of equipment like this in real life? Have you ever tried using that ridiculous cabinet simulation equipment, or VST valve emulators?

I know the answer - NO.

because they all sound absolutely rubbish - tried pretty much all of them, as my friend works in turnkey in london.

That's true. The main reason for this is that guitars and amplifiers are highly non-linear systems. Most digital systems try to simulate them using LTI theory or, I believe, coupled interpolated LTI systems to simulate some degree of non-linearity. The problem is, that non-linearities are one of the main reasons guitar purists prefer valve amps ("warm", "smooth", etc). Furthermore, things like amp <-> speaker feedback and speaker <-> air response are hard to simulate.

There are some quite good simulators on the market (I actually use one myself, which emulates a Vox amp quite nicely). However, when I plug my guitar into the Fender Twin of our other guitarist I'm in guitar heaven for at least half an hour... 

By the way I don't think this topic should be closed. I see nice postings from both "camps". I'm currently deeply into human auditory and cognitive research, so I find it highly interesting.
Title: Why 24bit/48kHz/96kHz/
Post by: PKI on 2007-01-13 17:22:32


...Lighting, cables, stands... All these variables affected (and affect) the reproduction of the standard CD format we heard...



This is one of the most absurd things I've ever heard!


You're missing the point and I'm only asking you to look beyond the purely scientific viewpoint for a few seconds.

You've taken that stement out of a wider context. I was saying that there are many variables that will affect the way the sound is reproduced... Are you telling me that the various dymanic ranges and electrical characteristics of the equipment won't make any difference? But this isn't about what we can hear or measure in an ABX or lab..

Of course we can't 'hear' above 22k.. we all agree that bit depths above 15 or 16 become hard(er) to differentiate.. Science and good mearuring has proved as much.. but just because we can't hear them, doesn't mean that frequencies greater than 22k or lower that 18 Hz aren't playing their part in the 'experience'... 

You can remove various influences to lean or focus the measuring, but you simply can't discount all other known or unknown factors and state they have no effect...

I may have bettered the most absurd thing you've ever heard, as It doesn't suit the black and white mathematical model, but the phrase you're looking for is it just might..

If it stays a friendly, then I agree that this should stay open, as it is nice to see some behavioural input.
Title: Why 24bit/48kHz/96kHz/
Post by: knutinh on 2007-01-13 17:30:45
Still no-one that has ever been able to pick out a ripped DVD-A from a ripped DVD-A properly converted to 44.1/16?

-k
Title: Why 24bit/48kHz/96kHz/
Post by: Woodinville on 2007-01-13 20:33:48
Well if you had any knowledge, experience or indeed any valid opinion whatsoever on this subject you'd be well aware that this is not possible yet. Have you ever actually heard a combination of equipment like this in real life? Have you ever tried using that ridiculous cabinet simulation equipment, or VST valve emulators?



Well I have a lot of knowledge, experience, and if there is a "valid opinion" on this planet I'm one of them, and yes, it is possible. It takes more processing than most people imagine (and by the way, LTI is not the approach taken, either, but you start by oversampling by 4 or 8...).

I don't quite understand your hostility. You might expect some people to be annoyed when you barge in and say a collection of inaccurate things.

Of course we can't 'hear' above 22k.. we all agree that bit depths above 15 or 16 become hard(er) to differentiate.. Science and good mearuring has proved as much.. but just because we can't hear them, doesn't mean that frequencies greater than 22k or lower that 18 Hz aren't playing their part in the 'experience'...

How does something that nobody can detect, ever, in a decently designed test, affect anything?

You're arguing for something that looks frankly parapsychological. If it affects the person in any way, it will show up in a decently designed test.

If you've ever run such a test, you will know just how annoyingly true that is, how hard it is to keep influences OUT of a test, even when you know about them and want to exclude them.

Those of us who have run extensive well-designed blind tests (not all DBT, some were computer-administered and were effectively DBT, some were signal detection, etc) have quite a bit of experience in detecting things like "was the door shut all the way".

Any bit of any information, via any sensory modality, will creep in if you're not careful to avoid it. If ultrasonics were affecting someone in any way, you'd see it in the test results. "Hearing" is not required.
Title: Why 24bit/48kHz/96kHz/
Post by: [JAZ] on 2007-01-13 20:38:20
If the topic is not closed, either we keep on topic, or continue this nonsense.

The discussion is about the need,  *for listening pleasure* , of high resolution formats, (currently namely DVD-A and SACD, but extendable to higher sampling rates/bitdepths in general).


To this point,

A) The "look" of the waveform does not correlate to what a DAC outputs, so "the 96Khz is more similar to the analog counterpart" points is useless. Anyone is free to show me a 1Khz sine at 44Khz samplerate being worse than 1Khz sine at 96Khz, being both DAC's of the same quality, and having no resampling issues. (1Khz square wave is out of the question, for obvious reasons).

B) It doesn't really matter if i can simulate an electric guitar, or a microwave heating my milk. We are talking about reproduction of recorded material, and how the sampling Rate/bitdepth affects it.

C) Hardware will definitely play a role into differentiating a high resolution format from standard CD Audio, but only because most hardware today is made to play audio in the "Human range", which is quite near to the CD Audio standard.  Then, a blind test is needed, else, we are just blindly *believeing*.


Let me say it again... We, at Hydrogenaudio, *only* care about what is audible, and verifiable.
Title: Why 24bit/48kHz/96kHz/
Post by: foxyshadis on 2007-01-14 10:52:15

Of course we can't 'hear' above 22k.. we all agree that bit depths above 15 or 16 become hard(er) to differentiate.. Science and good mearuring has proved as much.. but just because we can't hear them, doesn't mean that frequencies greater than 22k or lower that 18 Hz aren't playing their part in the 'experience'...

How does something that nobody can detect, ever, in a decently designed test, affect anything?

You're arguing for something that looks frankly parapsychological. If it affects the person in any way, it will show up in a decently designed test.

The point made is that 44/16 is fine, discussion dead, about 10 page ago, and now moving to a discussion away from the benefits of doubling rate/depth into a broader discussion on recording and replaying experiences. PKI simply isn't nearly as good at explaining himself as 2Bdecided. But his assertion is also rather pointless - it's that anything that affects your psychological state will color your perception of sound, which should be self-evident, since listening is a function of the brain and not just the ears. Anyone who's listened to a song while mopey and later while distracted and irritated could tell you that.

I'm totally with those who feel this is not the thread to be discussing this, the original topic has long run its course. And without some better science, probably not the forum to be discussing it. I'm personally interested in more science behind immersive music reproduction (beyond "crank it to 11" and "add two more channels"), though, if someone will start a new thread.
Title: Why 24bit/48kHz/96kHz/
Post by: knutinh on 2007-01-14 11:22:10
I'm totally with those who feel this is not the thread to be discussing this, the original topic has long run its course. And without some better science, probably not the forum to be discussing it. I'm personally interested in more science behind immersive music reproduction (beyond "crank it to 11" and "add two more channels"), though, if someone will start a new thread.

Thought experiment:
Is there some way to do comparisions between actually being in a soundfield, and reproducing that soundfield using PCM audio and binaural reproduction?

If PCM and AD/DA technology can outperform our 1-dimensional hearing capabilities, and we only have two ears that can be properly stimulated using headphones... How can one practically put a person into a live concert hall, let him hear live music, then playback music recorded in the same spot using binaural tech, and do some kind of ABX?

-k
Title: Why 24bit/48kHz/96kHz/
Post by: crimsontide on 2007-01-14 12:23:14
Well I have a lot of knowledge, experience, and if there is a "valid opinion" on this planet I'm one of them, and yes, it is possible. It takes more processing than most people imagine (and by the way, LTI is not the approach taken, either, but you start by oversampling by 4 or 8...).

I don't quite understand your hostility. You might expect some people to be annoyed when you barge in and say a collection of inaccurate things.


Interesting.....

Firstly I based my opinion on the number of experiments i have done in hardware and software emulation. I've never even come close to the aforementioned guitar "heaven" without using cables, and valves that have been nicely warmed up. Please note that I said it's not possible _yet_ and I stand by my comment, until i have seen evidence of anything else.

Which brings me to the question - have you abxed the two? recorded a riff through your cab and head, and abxed it with the emulated equivalent? could it be done? and if so, have you got an yresults anywhere? I amn highly intersted in this area, and would dearly love to be proven wrong.

My hostility was because a certain user was downright rude, and patronising. In a personal social situation his address to me would have caused a conflict without question, when if you read my posts it was not deserved. I had already addressed my "inaccurate" comments, and admitted that it was a poor analogy anyway.

Hence the Godwins moment as well.

Moving on however....and back on topic...can you use some of these samples of emulated speaker cabs and abx those between 44/16 and 96/24 and see if you can tell the difference - has anyone tried doing this yet?

I'm not convinced I have many samples (if any) that are worth trying to abx between these frequencies. I might ask my friend to record some seriously expensive kit at 96/24 (he has logic audio and nice equipment) through a nice expensive mike, and perhaps i could post it up for an abxing experiment.

It wont answer the BIG question of whether the emulation sounds as good (one which will likely forever continue as per CD/vinyl debate), but it might shed new light on the practicality of 44/16 vs 96/24.

M<ore importantly can you post up a sample of the emulation as it currently stands so i can have a listen , and form a wholly unscientific thought in woo about it?   

I like living in woo......I'm roadieing at lark in the park in london in 2 weeks, where I can woo all day long.
Title: Why 24bit/48kHz/96kHz/
Post by: Woodinville on 2007-01-14 14:20:18
Which brings me to the question - have you abxed the two? recorded a riff through your cab and head, and abxed it with the emulated equivalent?


Do you really think you can play the riff the same way, exactly, to permit this test?

I doubt it.

That aside, yes, I have seen good emulators, but I frankly don't care a lot about that, because it's irrelevant to the question in the OP. You're asking a question of how to model things that are far, far above audible thresholds. The question in the OP is at best at audible threshold.  Different problem, different question.
Title: Why 24bit/48kHz/96kHz/
Post by: PKI on 2007-01-14 16:32:59
PKI simply isn't nearly as good at explaining himself as 2Bdecided.


Harsh. But that's because I'm trying to put across cognitive & behavioural concepts I'm not entirely familiar with into a scientific debate that I am familiar with...

The problem we have here is that;

We, at Hydrogenaudio, *only* care about what is audible, and verifiable.


This rules out any more 'what if' debate then..

Edited: Correct quote box error.
Title: Why 24bit/48kHz/96kHz/
Post by: knutinh on 2007-01-14 16:35:49
Why mix guitar amp rigs into this discussion?

The subjectively "best" electric guitar 12" speaker cabinets start to rolloff above 12kHz, and the most common microphone used for this application is the Shure sm-57 with very limited hf response.

Moreover the SNR of most speaker setups is low enough that 24 bits recording presicion would be a waste.

24bits may however have valid use if one is to simulate tube distortion on the dry electric guitar signal due to the gain used in the simulation.


There is no doubt in my mind that inaccurate simulation of such instruments is due to algorithmic problems and lack of processing power, not the DAC converters used. Also, when doing non-blind testing, one really is testing more than the pure audio capabilities.

-k
Title: Why 24bit/48kHz/96kHz/
Post by: [JAZ] on 2007-01-14 16:46:43
24bits may however have valid use if one is to simulate tube distortion on the dry electric guitar signal due to the gain used in the simulation.


Just adding that:
Digital signal processing with software is done, at least, with 32bits floating point precision, if not 64bits (double precision). Hardware is a different story, but floating point processors are appearing more and more on these too (no longer limited to integer arithmetics).
Title: Why 24bit/48kHz/96kHz/
Post by: knutinh on 2007-01-14 16:50:59


We, at Hydrogenaudio, *only* care about what is audible, and verifiable.


This rules out any more 'what if' debate then..

Edited: Correct quote box error.

If I understand the rules correctly, you are free to suggest _possible_ mechanisms as long as you dont state them as a fact? Moreover, you are free to discuss mothods of verifying audible differences.

If I cannot pick out a hirez sample from a downconverted lorez sample in a blind test using a verified playback chain.... What is your explanation for this? How do you suggest that I test to maximise the chance of me noticing any difference at all?

If the difference can be "felt" but not heard, why cant I "feel" it in a blind-test? What is it about regular listening that suddenly makes me sensitive to feeling such high frequencies?


regards
k

Quote
' date='Jan 14 2007, 17:46' post='464121']

24bits may however have valid use if one is to simulate tube distortion on the dry electric guitar signal due to the gain used in the simulation.

Just adding that:
Digital signal processing with software is done, at least, with 32bits floating point precision, if not 64bits (double precision). Hardware is a different story, but floating point processors are appearing more and more on these too (no longer limited to integer arithmetics).

The internalt precision of a DSP may be du to the specific design of algorithms to avoid accumulating errors. By careful algorithmic design, one may be able to work around limited internal precision. Of course this is tedious work, so everyone prefers being able to consider multiplications as "infinite precision".

I was talking about the need to capture the source with as much precision as possible due to digital gainstages of perhaps 40dB attenuation.

A subtle difference I believe.

-k
Title: Why 24bit/48kHz/96kHz/
Post by: krabapple on 2007-01-15 05:31:39
My hostility was because a certain user was downright rude, and patronising. In a personal social situation his address to me would have caused a conflict without question, when if you read my posts it was not deserved. I had already addressed my "inaccurate" comments, and admitted that it was a poor analogy anyway.



This certain user has seen arguments just like yours ad nauseam on audio forums.  They aren't new.  They aren't fresh.  They tend to rely heavily on subjective impressions gathered in bias-prone situations, and often veer off into demisequiturs like the guitar gear sim thing.  They are perennial outcroppings of the dominant paradigm of subjectivist audiophile culture. HA.org's is the *minority view* in audiophile-land -- a bizarre land whose inhabitants sadly include more than a few audio/sound professionals with an apparent suspicion of scientific methods.

That you'd present such arguments as they were news here on HA.org, given its rather explicit TOS and history of threads about hi rez formats, sampling rates, blind tests -- and present them in the Scientific/R&D subforum no less --  might itself be interpreted as an aggressive act, perhaps even rude and patronizing...a suspicion supported by your subsequent Mel Gibson moment.  It was by me.
Title: Why 24bit/48kHz/96kHz/
Post by: crimsontide on 2007-01-15 13:23:58
That you'd present such arguments as they were news here on HA.org, given its rather explicit TOS and history of threads about hi rez formats, sampling rates, blind tests -- and present them in the Scientific/R&D subforum no less --  might itself be interpreted as an aggressive act, perhaps even rude and patronizing...a suspicion supported by your subsequent Mel Gibson moment.  It was by me.


I suggested that the sound samples we all have available aren't truly representative of the sounds they were recording, and that if they were better one day - perhaps 96/24 would come into its own.

That is not aggressive, rude or patronising, and to suggest it is, somehow because you're opinion is a minority in a scene I know nothing of, is the audio technological equivalent of "is it cos I is black"?

you seem so blindly committed to the direction of your scientific research, that it is no longer scientific research. "lets find a way to prove what i already believe" is not science.

I made a fair suggestion, that is all....to expand the research before you draw your conclusion.

Anyways.....I don't care anymore. I still think we should just use 96/24 anyhow - we've got plenty of space on our media. But thats for another thread, on another forum.
Title: Why 24bit/48kHz/96kHz/
Post by: Axon on 2007-01-15 17:13:23
To summarize crimsontide's post, with my own spin:

Unless you are an audiologist who has found new, concrete evidence of ultrasonic audibility, which is peer-reviewed, do not post in this thread.
Title: Why 24bit/48kHz/96kHz/
Post by: pepoluan on 2007-01-15 18:20:28
I play the Hammond organ, and I can tell you that it is a great experience. It weighs about 200kg, makes all kinds of noises and smells like old sewing machine oil and hot tubes. Oh, and you need a loan as well as hefty, frequent repairs. No digital simulator has ever given me the same feeling as playing the Hammond organ. Does this mean that those simulators cannot recreate the sound with sufficient accuracy for my ears and abilities? Does that mean that Nyquist, Shannon, ABX etc is all B.S.?
What's playing got to do with sampling rate or sampling bitdepth?

You're not only comparing apples to oranges. You're comparing apples to ladybugs, which are not only different kinds of fruits, heck one is not even a fruit!

Your satisfaction of playing a real Hammond obviously cannot be replaced by playing a digital keyboard. BUT, record the Hammond you played, and record the digital keyboard capable of reproducing the Hammond's sound exactly... AND LISTEN to them, using ABX. Now that's comparing apple to apple.

One could argue that the smell and looks of an instrument never appears on a record and therefore are "luxury" stuff.
Exactly.

On the other hand, the musician will probably perform differently if he is satisfied with the sound (even if it is all in his head), and that will most certainly change what is recorded.
Which means that it will be easily ABX-able, no?

Take me, playing a piano hesitantly. The notes will sound hesitant, as I do not dare to bang the keys with too much force for fearing I'd make a mistake. Then get Harry Connick Jr. or Barry Manilow to bang out a tune on the piano. And they don't fear a mistake and you'll hear the keys more distinct, more... fortissimo.

But they're easily ABX-able.

In much the same way, the listener experience in practical hifi is intertwined with sound, vision, knowledge etc.
So if we want to limit the scope to whether higher sampling rate and/or higher bitdepths affect the sound quality, we must remove all those variables.

So why would we want to tell a hifi-listener that his experience is based on superstition and that he cant hear anything at below -80dBFs, or above 22kHz? I feel no need for doing this, but I think that when discussing these matters it is important to get the facts as right as possible.
No, we don't say that his experience is based on superstition, but too often the hifi-listener "thought" that he/she hears a difference. Sometimes using strong words like "it lost its warmth, the stereo separation collapses, blah blah blah". While in fact there is no difference whatsoever between the first and the second tracks (i.e. the tester purposefully did not switch input, but just faked switching).

That's bias. That's placebo. And ABX will remove that bias/placebo.

For the most important facets of life, "subjectivity" is probably highly beneficial and necessary to lead a good life and have offspring.
So why do we have SAT's, GRE's, entrance exams, Cisco Certification Exams, Microsoft Certification Exams, etc?

Because maybe to have a good life and have offsprings one needs to be subjective; but to progress and excel one need, one must, be objective and measurable.
Title: Why 24bit/48kHz/96kHz/
Post by: knutinh on 2007-01-15 21:04:29
I play the Hammond organ, and I can tell you that it is a great experience. It weighs about 200kg, makes all kinds of noises and smells like old sewing machine oil and hot tubes. Oh, and you need a loan as well as hefty, frequent repairs. No digital simulator has ever given me the same feeling as playing the Hammond organ. Does this mean that those simulators cannot recreate the sound with sufficient accuracy for my ears and abilities? Does that mean that Nyquist, Shannon, ABX etc is all B.S.?
What's playing got to do with sampling rate or sampling bitdepth?

You're not only comparing apples to oranges. You're comparing apples to ladybugs, which are not only different kinds of fruits, heck one is not even a fruit!

Your satisfaction of playing a real Hammond obviously cannot be replaced by playing a digital keyboard. BUT, record the Hammond you played, and record the digital keyboard capable of reproducing the Hammond's sound exactly... AND LISTEN to them, using ABX. Now that's comparing apple to apple.

If you read my post once more, I think you will find that I was trying to make a point. Hint: I was not argueing for higher sampling rates or bit-depths...

My point was exactly that people bring all kinds of other mechanisms into the discussion to proveobscure points about how "magic" the hifi experience is. My point is that there are many music-related processes that are complex, but the simple operation of replaying a digitized waveform can be investigated in a perceptual fashion quite easily.

-k
Title: Why 24bit/48kHz/96kHz/
Post by: Woodinville on 2007-01-15 22:57:55
I suggested that the sound samples we all have available aren't truly representative of the sounds they were recording,

Indeed they are not, there is a huge quantity of spatial information and soundfield information that is lost when something is reduced to either 2 or 5 channels.
Quote
and that if they were better one day - perhaps 96/24 would come into its own.


So, why use twice the data in a way that is well understood and known to be very close to threshold if not below it, when you could double the number of channels, and attack a known problem?
Title: Why 24bit/48kHz/96kHz/
Post by: Woodinville on 2007-01-16 01:26:44
I actually think it would be very cool if the hearing was proven to be more than the scientists have proven so far. Interesting. But as long as the alternative/subjectivist camp is incapable of providing coherent lines of thought that would pass the scruitiny of a 5-year-old, I suspect that any such revelations will appear from elsewhere.

This is sad because it is always more interesting to discuss with smart people than angry people, even if you disagree on everthing.

-k


And, if such were discovered, it would provide new avenues for research that is pretty much settled on an understanding of how the Human Auditory System works.

If PCM and AD/DA technology can outperform our 1-dimensional hearing capabilities, and we only have two ears that can be properly stimulated using headphones... How can one practically put a person into a live concert hall, let him hear live music, then playback music recorded in the same spot using binaural tech, and do some kind of ABX?

-k



Some things to consider. 

Ears/heads/bodies have HRTF's. So they are not "1-dimensional" in very real ways.

We have two ears. They move around a lot.

If you do a binaural recording, the playback head position will not match the recording head position.

When you move around, you sample a lot of the soundfield around you. Next time you're in a non-rock, mostly acoustic concert, watch people's heads, watch how they hold them, move them, hold their bodies, etc.

Try it yourself. Watch what happens when instruments change, even with the same mono'ed PA.
Title: Why 24bit/48kHz/96kHz/
Post by: greynol on 2007-01-16 07:15:57
Quote
and that if they were better one day - perhaps 96/24 would come into its own.

So, why use twice the data in a way that is well understood and known to be very close to threshold if not below it, when you could double the number of channels, and attack a known problem?

To add to this, I don't care if you model a guitar and amp combination using 192/48.  When it comes to playing it back, 44/16 will not sound any poorer than "hi-rez" ESPECIALLY with a precious PRS through an over-driven Marshall.

I too have been playing guitar for a very long time and have some very nice equipment, just as good as anything you'll get from a roadie from Genesis, I assure you.  This does not give me any more or any less credibility when it comes to discussing things in this thread that are actually ON-TOPIC.
Title: Why 24bit/48kHz/96kHz/
Post by: knutinh on 2007-01-16 09:26:27


If PCM and AD/DA technology can outperform our 1-dimensional hearing capabilities, and we only have two ears that can be properly stimulated using headphones... How can one practically put a person into a live concert hall, let him hear live music, then playback music recorded in the same spot using binaural tech, and do some kind of ABX?

-k



Some things to consider. 

Ears/heads/bodies have HRTF's. So they are not "1-dimensional" in very real ways.

We have two ears. They move around a lot.

If you do a binaural recording, the playback head position will not match the recording head position.

When you move around, you sample a lot of the soundfield around you. Next time you're in a non-rock, mostly acoustic concert, watch people's heads, watch how they hold them, move them, hold their bodies, etc.

Try it yourself. Watch what happens when instruments change, even with the same mono'ed PA.

Well, at the entry-point to the ear they are one-dimensional. The movement in 3d space, interactions with body etc are of course as 3d as sound itself.

A proper 2x 1-dimensional headphone, fed the right signal, should be enough to simulate any physical soundspace that we may encounter except sound traveling through our body.

I still think that it would be a fascinating experiment to try to recreate the in-room sound.... well... in-room :-) Even if the user was forced to not move, and one had to modell his exact body and head physically.

I think that binaural sound is a highly underestimated way of simulating reality. I am aware that there is considerable research in synthesizing binaural sound (from eg movement sensors), calibrating real head vs artificial head differences (from pictures) etc.

-k
Title: Why 24bit/48kHz/96kHz/
Post by: Woodinville on 2007-01-16 10:15:20
Well, at the entry-point to the ear they are one-dimensional. The movement in 3d space, interactions with body etc are of course as 3d as sound itself.

And you move around.
Quote
A proper 2x 1-dimensional headphone, fed the right signal, should be enough to simulate any physical soundspace that we may encounter except sound traveling through our body.



Except that your head moves around. The "headphone" would have to capture that from a very complex original source. There are some examples of this, they are presently a wee bit complex.

And you'd have to have a lot of information about the original soundfield, not two channels.
Title: Why 24bit/48kHz/96kHz/
Post by: crimsontide on 2007-01-17 13:22:24
Quote
and that if they were better one day - perhaps 96/24 would come into its own.

So, why use twice the data in a way that is well understood and known to be very close to threshold if not below it, when you could double the number of channels, and attack a known problem?

To add to this, I don't care if you model a guitar and amp combination using 192/48.  When it comes to playing it back, 44/16 will not sound any poorer than "hi-rez" ESPECIALLY with a precious PRS through an over-driven Marshall.

I too have been playing guitar for a very long time and have some very nice equipment, just as good as anything you'll get from a roadie from Genesis, I assure you.  This does not give me any more or any less credibility when it comes to discussing things in this thread that are actually ON-TOPIC.



I made a postulation. There is no evidence. Thats why its a postulation. But it's still a part of science - without postulation you can't define a range for your samples. You have to do further tests to discern if there _is_ any evidence. The tests i have seen being done have not imho been conclusive or varied enough to draw a solid conclusion as yet. IF you read back inthe thread you'll see that I have abxed some samples, and i did conclude that i could not tell the difference, and now I'm suggesting a couple of reasons why more research should be done.  I'm not challenging anyone or any person, just the samples used, and potentially the equipment used to abx them.....(but then thats also another thread).

My subject matter is actually on topic - sure I went off a bit to demonstrate why i think the conclusion you (collective) seem to have drawn may not be valid (in the long run), but my reasons were always on topic. I'm sorry I poked holes in your baby. You may not be able to see their faults, but they are there.

Your response has ignored my suggestion about microphones and recording methods developing, for example - providing no evidence. Would you be so short sighted to think they (mics and recording techniques) are as good as they will ever be? That is not the case.....I can assure you from life experience.

I agree with the idea about adding more channels instead of bitrate and depth though - can't argue with that....but then its already been done, and lets be honest, the live CD doesnt sound as good as the 5.1AC3 soundtrack on the DVD - probably due to immersion from 3 dimensional sources.

BUT has anyone abxed the CD versus the AC3 file, after mixing them both down/up to 96/24 stereo?

Suggest fleetwood mac - the dance live. I dont have the DVD anymore, i lost it.
Title: Why 24bit/48kHz/96kHz/
Post by: 2Bdecided on 2007-01-17 15:05:32
A proper 2x 1-dimensional headphone, fed the right signal, should be enough to simulate any physical soundspace that we may encounter except sound traveling through our body.


Except that your head moves around. The "headphone" would have to capture that from a very complex original source. There are some examples of this, they are presently a wee bit complex.

And you'd have to have a lot of information about the original soundfield, not two channels.


But knutinh was suggesting a kind of ABX, and how you'd do it - not a practical recording idea.

Clamp the listener's head. Use in-ear microphones to record the content (captures the listener's HRTFs), and in-ear headphones to replay it.

You can't do a blind test, never mind ABX - you need the in-ear headphones present to hear the recorded version, but absent to hear the live version.

However, even a sighted comparison would be interesting. I suggest it will tell you nothing about the parameters of the recording (44.1kHz is probably enough) but lots about the transducers (which will be near impossible to get right).


You're right about the sound field of course. That's where we should be going. To think Michael Gerzon was doing this stuff 30 years ago - how little we learn!

Cheers,
David.
Title: Why 24bit/48kHz/96kHz/
Post by: Woodinville on 2007-01-17 19:14:53
I'm sorry I poked holes in your baby.



Show me evidence of that, please. I haven't seen anything relevant from you to speak of, only a huge misunderstanding of what kinds of effects are near or below audible thresholds, and what kinds of effects are way, way above audible threasholds, and an even bigger failure from you to understand why comparing something in your hands, that you feel, smell, etc, with something that you only experience from a two-channel recording is just silly. Of course that's different. It says nothing about 24/192 or anything of the sort.

To think Michael Gerzon was doing this stuff 30 years ago - how little we learn!

Cheers,
David.


However, he only captured the details of the soundfield at one point in the soundfield. This is where some other approaches differ with his.
Title: Why 24bit/48kHz/96kHz/
Post by: dekkersj on 2007-01-19 19:55:18
Quick update.

I am replying to the original question: "Why 24bit/48kHz/96kHz".

A few weeks ago I visited a member of another forum and as a sort of sanity check I played a sweep at -89 dBFS. Normally it stays silent, meaning that 16 bits is enough. But in his case I could hear a very small fraction of the sweep. That was because of the quite large playback level. We could not talk with each other when listening to his set. Funny thing was that he could not hear a thing of the sweep.

Conclusion: at normal levels, 16 bits is enough.

Regards,
Jacco
Title: Why 24bit/48kHz/96kHz/
Post by: Axon on 2007-01-19 20:02:06
Here's another reason to record at high sampling rates. However, it is rather contrived and currently doesn't exist in the real world AFAIK.

If you have an accurate numerical model of a nonlinear distortion, and you are able to invert the model, then you can remove the distortion from a signal. The signal will intermodulate with itself in the presence of that distortion, generating ultrasonics. Those ultrasonics are probably going to be necessary in order to run the distortion cleaner accurately. (Or, if you know more than that, perhaps you can autodetect the distortion model based on the ultrasonics and work from there.)

This could apply both to recording (to correct nonlinearities in analog media) or to playback (to correct nonlinearities in speakers). But you'd need scarily accurate simulations.
Title: Why 24bit/48kHz/96kHz/
Post by: ilo on 2007-01-20 00:44:45
I'm kind of late to this thread, and I haven't read it in its entirety, but generally some intuitive thoughts supportive of high resolution audio formats would be:

1. Headroom: I would like to see recordings with 90dB nominal SNR where there was still 40dB headroom for loud dynamic passages. The nominal recording level should be far, far below 0dBfs for real tangible dynamic headroom, but still be perceptually transparent in the "nominal mode".

2. Why not? It's no problem with modern computer technology.

3. It is always much harder to improve analog performance than digital, in modern submicron processes more so than ever, thus the error contribution from the format itself should preferably be negligible, even when budgeting for something on the limit of human hearing capabilities. Then, the entire error budget could be allocated to the analog parts, instead of being eaten up by the digital encoding and processing.
Title: Why 24bit/48kHz/96kHz/
Post by: uart on 2007-01-20 08:40:40
Quick update.

I am replying to the original question: "Why 24bit/48kHz/96kHz".

A few weeks ago I visited a member of another forum and as a sort of sanity check I played a sweep at -89 dBFS. Normally it stays silent, meaning that 16 bits is enough. But in his case I could hear a very small fraction of the sweep. That was because of the quite large playback level. We could not talk with each other when listening to his set. Funny thing was that he could not hear a thing of the sweep.

Conclusion: at normal levels, 16 bits is enough.

Regards,
Jacco



That's probably because his already damaged his hearing by listening at such high volume levels.
Title: Why 24bit/48kHz/96kHz/
Post by: 2Bdecided on 2007-01-22 12:30:14
I'm kind of late to this thread, and I haven't read it in its entirety, but generally some intuitive thoughts supportive of high resolution audio formats would be:

1. Headroom: I would like to see recordings with 90dB nominal SNR where there was still 40dB headroom for loud dynamic passages. The nominal recording level should be far, far below 0dBfs for real tangible dynamic headroom, but still be perceptually transparent in the "nominal mode".


Have you heard/got the Hi-Fi news test CD, which includes a single fire cracker / firework being set off, and the crowd's reaction to it? It's a great example of dynamic range. Here is a short extract:

[attachment=2827:attachment]

That's from a CD, so is "only" 16-bits.

The problem I have reproducing this realistically (i.e. to sound as if I was stood in the crowd) isn't the 16-bit limit, but my amplifier, speakers, and neighbours!

Cheers,
David.
Title: Why 24bit/48kHz/96kHz/
Post by: knutinh on 2007-02-04 13:01:11
It is possible to create digital content with practically any snr wanted. I could take a regular cd recording at 16 bits, decide that I wanted to pad down the middle section by 96dB, and effectively have a 32bit dynamic range mix. I cant see that this is anything that anyone wants to do. Otherwise, you are practically limited by the microphones, pre-amps and rooms of this world. I dont think that it is possible to capture a live acoustic performance with 32bits of precision at a reasonable samplerate today.

Of course, if you listen to techno or some other purely synthetic music, then all usic could be rendered as a mathematical model inside a vst plugin as 32bit, 64bit or even 128bit floating poit precision numbers. But how important is that in the big picture?


As long as adc boxes capture the signal from microphone/preamps exceeding their specs, I see no place for "headroom increase" as mentioned above. The problem lie in a signal chain of limited SNR (mike, preamp, dynamics processors, amplifiers, rooms etc). And even that chain seems to satisfy our limited hearing senses.

It is interesting with "extreme dynamic" reproduction. But I think that we need better microphone/loudspeaker technology before blaiming the poor CD player. And given the choice, sound engineers hired for their subjective tastenearly always _reduce_ dynamics on live and recorded music, at least partially because that is what sounds better.

-k
Title: Why 24bit/48kHz/96kHz/
Post by: hushypushy on 2007-02-06 23:55:54
The problem I have reproducing this realistically (i.e. to sound as if I was stood in the crowd) isn't the 16-bit limit, but my amplifier, speakers, and neighbours!


This is a great quote! Every time I read/hear people talking about how 24/96 is necessary for sound quality, I'm constantly reminded of your post
Title: Why 24bit/48kHz/96kHz/
Post by: Danny Kaey on 2007-02-11 18:33:53
the problem isn't so much the 16bit dynamic limit, rather music being mastered to have 0.5db of dynamic range and headroom.  You could have 24bit dynamic limits, if the music is mastered loud w/ no headroom and dynamic range, it will still sound like crap!
Title: Why 24bit/48kHz/96kHz/
Post by: wildnewt on 2007-02-21 13:08:41
Yes, I have searched the forum.
Yes, maybe I am dumb.

But it seems I cannot find the answer.

Why do we need 24bit/48kHz/96kHz/192kHz if 16bit/44.1kHz is good enough? Are there any situations that 16bit/44.1kHz simply cannot satisfy? In other words, is there any real need for the higher bit depth and sampling rate?

Thanks for answering.


Simple: 16bits gives you: 20log (2^16bits) = 96.32dB dynamic range, fine for listening, and mixed materials (mixed CD output).
24bits gives you: 20log (2^24bits) = 144.49dB dynamic range which covers the dynamic range (difference of quietest and loudest points of PGM) of stuff like an orchestra (with up to 120db dyn range), so to capture all nuances, we need high dynamic range.

higher dynamic range means lower noise floor, and when mixing multiple signals, multiple low noise floors combine to give increased noise, so we want to make this as low as possible when mixing.

Higher khZ generally means higher quality - better image representation through shifting the Nyquist frequency higher up (well beyond audible). However, if you have shit-hot (eg apogee) DACs, you don't need super high sample rates (due to their oversampling functions). A great DAC at 44.1 is as good as an average DAC at 192khz. Ask Bob Ludwig about this.
Title: Why 24bit/48kHz/96kHz/
Post by: knutinh on 2007-02-21 13:33:11

Yes, I have searched the forum.
Yes, maybe I am dumb.

But it seems I cannot find the answer.

Why do we need 24bit/48kHz/96kHz/192kHz if 16bit/44.1kHz is good enough? Are there any situations that 16bit/44.1kHz simply cannot satisfy? In other words, is there any real need for the higher bit depth and sampling rate?

Thanks for answering.


Simple: 16bits gives you: 20log (2^16bits) = 96.32dB dynamic range, fine for listening, and mixed materials (mixed CD output).
24bits gives you: 20log (2^24bits) = 144.49dB dynamic range which covers the dynamic range (difference of quietest and loudest points of PGM) of stuff like an orchestra (with up to 120db dyn range), so to capture all nuances, we need high dynamic range.

How many systems are actually able to represent that dynamic range, given inevitable toom-noise and the pain-threshold?
Quote
higher dynamic range means lower noise floor, and when mixing multiple signals, multiple low noise floors combine to give increased noise, so we want to make this as low as possible when mixing.

So by how many dB do you think that the noise-floor increases if I mix 8 16-bit streams into one 16-bit mix at high internal precision? Assuming that the noise is white uncorrelated, and that signals are uncorrelated.
Title: Why 24bit/48kHz/96kHz/
Post by: wildnewt on 2007-02-21 14:02:41
Quote
How many systems are actually able to represent that dynamic range, given inevitable toom-noise and the pain-threshold?

higher dynamic range means lower noise floor, and when mixing multiple signals, multiple low noise floors combine to give increased noise, so we want to make this as low as possible when mixing.

So by how many dB do you think that the noise-floor increases if I mix 8 16-bit streams into one 16-bit mix at high internal precision? Assuming that the noise is white uncorrelated, and that signals are uncorrelated.


don't confuse dynamic range and physical loudness which are both measured in dB. PGM with 144dB dynamic range doesn't necessarily mean the loudest parts of the it will be represented/reproduced at (physical loudness) 144dB.

Without numbers or measuring the 16bit streams, what type of dithering used etc it's hard to say exactly. but in a well designed studio it's possible to hear the noise-floor on one 16bit channel (and a noisefloor of -80dB). never mind when you mix 8. the standard of listening environment required increases dramatically if you want to hear the noise-floor in a 24bit stream.

Unless anyone understands the maths and reasonings behind precision and dB, it's easy to get confused or follow misconceptions.
Title: Why 24bit/48kHz/96kHz/
Post by: 2Bdecided on 2007-02-21 14:32:24
Unless anyone understands the maths and reasonings behind precision and dB, it's easy to get confused or follow misconceptions.


Oh, I don't think there's anyone in HA who understands maths 


Seriously, at least read some of the thread before jumping in with a half baked answer to a question posed on Dec 29 2005!!!

All your points have been rebutted already. 16-bits is sufficient for delivery to almost any conceivable home listening situation. Mixing can and should take place at higher resolution, and no one has argued against this.

Cheers,
David.
Title: Why 24bit/48kHz/96kHz/
Post by: w1L50n on 2007-02-21 15:57:25
I've been trying to learn about all this stuff from reading and starting at the fundamentals.  In this pursuit I read all 17 pages....

Until a couple of posts ago, there was something that was confusing me:

'don't confuse physical loudness with dynamic range...both in db'

Thanks for that!

eeesh...17 pages...that was a hard fought piece of important intel however...now I can continue learning.
Title: Why 24bit/48kHz/96kHz/
Post by: greynol on 2007-02-21 18:07:54
'don't confuse physical loudness with dynamic range...both in db'

Thanks for that!
 
I hope you're kidding!
Title: Why 24bit/48kHz/96kHz/
Post by: krabapple on 2007-02-21 19:14:14


Simple: 16bits gives you: 20log (2^16bits) = 96.32dB dynamic range, fine for listening, and mixed materials (mixed CD output).
24bits gives you: 20log (2^24bits) = 144.49dB dynamic range which covers the dynamic range (difference of quietest and loudest points of PGM) of stuff like an orchestra (with up to 120db dyn range), so to capture all nuances, we need high dynamic range.

How many systems are actually able to represent that dynamic range, given inevitable toom-noise and the pain-threshold?


And how many audience members at orchestral concerts actually experience 120 dB of dynamic range?  Answer: none. That figure (probably from Alton Everest's book) comes from close-microphone placement.

Heres's from Everest, 4th edition, citing Fielder:

(condition: base SPL/dynamic range/peak SPL )

piano solo  13dB/90dB/103dB
typical classical symphony (tcs), w/audience: 13dB/100 dB/113dB
tcs, w/o audience:  8dB/105dB/113dB
tcs, close mic: 4dB/109dB/113dB
percussive classical, close mic : 4dB/118dB/122dB

see discussion of these numbers at

http://groups.google.com/group/rec.audio.h...ource&hl=en (http://groups.google.com/group/rec.audio.high-end/msg/b8bf2ad97451e238?dmode=source&hl=en)
Title: Why 24bit/48kHz/96kHz/
Post by: Woodinville on 2007-02-21 20:32:21
I've been trying to learn about all this stuff from reading and starting at the fundamentals.  In this pursuit I read all 17 pages....

Until a couple of posts ago, there was something that was confusing me:

'don't confuse physical loudness with dynamic range...both in db'

Thanks for that!

eeesh...17 pages...that was a hard fought piece of important intel however...now I can continue learning.



What is "physical loudness"?

Loudness, technically, refers to the intensity of the signal AS PERCIEVED BY THE LISTENER. It is a physiological variable, not a physical variable.

Intensity is the "SPL" or what-have-you.
Title: Why 24bit/48kHz/96kHz/
Post by: wildnewt on 2007-02-22 00:08:37
sorry, shouldn't have replied with my 17 nanoseconds of lunch break.    Yes, SPL, not physical loudness.

Anyway - I was answering the original question, not trying to summarise the entire thread of arguments. The end listener is fine with 16 bits, but it doesn't hurt to have more bits. More bits seem to sell more units of whatever hardware you're trying to sell 
Title: Why 24bit/48kHz/96kHz/
Post by: 2Bdecided on 2007-02-22 10:55:42
see discussion of these numbers at

http://groups.google.com/group/rec.audio.h...ource&hl=en (http://groups.google.com/group/rec.audio.high-end/msg/b8bf2ad97451e238?dmode=source&hl=en)


Fair discussion. "Audience noise, 13dB eqv" is so laughable that I can't believe anyone can take it seriously.

In a very nice large and well insulated anechoic chamber, where the one compromise was having to leave the (very large, padded motorised) door open by a few inches, I couldn't get lower than 20dB SPL A-weighted. Very few people can, anywhere. Never mind in a concert hall with an audience!!!.

Cheers,
David.
Title: Why 24bit/48kHz/96kHz/
Post by: drumroll57 on 2007-04-08 05:10:27
First post for me on here, and I guess reading the 17-page + argument is quite enlightening. There are obviously a lot a widely varied opinions on these topics, and although perhaps not backed by instrumentation and a lab environment, would like to report some of my own findings, which have taken a slow poke like myself years to really understand.

Let me state right upfront that I have been involved in recording, production, mixing and DJ activities for over a quarter of a century. Somehow this means experience, but also probably hearing damage, LOL!

The scope of my comment should be relating to one thing in particular: size of listening environment.

Here is the analogy: If I look at this site's graphic logo in its intended resolution for example, it looks perfectly fine, smooth and sharp. However, if I suddenly take the same logo and blow it up to fit a 2.5 meter-wide billboard poster (approx. 10 ft) then its edges will look ugly, jagged and just plain wrong....

My experiences have afforded me the opportunity to work on a weekly basis with very large-scale sound systems, the range going from 5,000 to 60,000 Watts of amplification, in rooms that can sometimes accommodate  as much as  10,000 people.

The one thing which has become immediately apparent to me is that the bigger the room's volume and its  sound system, the more easily I was able to distinguish artifacts which were totally inaudible in a smaller space. I also have access to studio-grade gear in vast quantity, and must confess that it is really hard for me to pick out the kinks between an MP3 file and its 16-bit 44.1 kHz source when listening on a pair of big Tannoy monitors and Bryston mono blocks in a studio, it seems that the sound fails to develop enough to really make a difference. Forget about headphones.

But surely I have done a number of tests on very large-scale listening systems, and that is where it has quickly become apparent that all of the extra information contained in a properly-encoded 24-bit / 96 kHz file does help make a noticeable difference in how it sounds compared to a 16-bit / 44.1 sibling.

Also experimented with up-sampling existing CD's via open-reel recording, simultaneously re-recording the result into a workstation at 24/96 to amazing results on quick A/B tests, which no doubt can be attributed to tape compression and other usual artifacts.

But there is no question in my mind that many of the arguments presented in the thread need to be illustrated within certain parameters, and that in my mind room size happens to be one of the most important ones.

The rough description of the type of tests I am basing my experiences on would be as below, in a room that holds about 1,500 people, roughly 30 meters wide by 25 meters deep with a ceiling height of 5 meters, with adequate amplification to bring sound to about 105 dB (?) without audible distortion, would imagine in the neigborhood of 15,000 Watts with plenty to spare, barely hitting '3' on the master volume:

-1) CD Red-Book Audio version of song  DAC from Pioneer CDJ-1000 Mk 2 player.
-2) vinyl version of same. (obviously, depending on cartridge, tonearm calibration, etc.. used on test was standard DJ-grade Ortofon cart + Technics SL-1200 turntable)
-3) File up-sampled via 15 ips tape recording to 24/96 from workstation. AD from Metric Halo 2882 - DAC from RME Multiface.
-4) recording of vinyl at 24/96.  AD from Metric Halo 2882 - DAC from RME Multiface

I can hear that the tape transfer has more bottom and smoother highs in pretty much any size room. Never mind that part, it's the 'tape-effect' for sure.

The big story for me was is how fond of COHERENCE my ears were.

When it comes to pure ear-pleasing power (punchiness, coherence and smooth presentation) the vinyl usually won as long as mastering wasn't horrible, even though there were many high frequency bits just missing in the very high upper register, and many percussive details in the midrange also a bit smeared.

The Red-Book audio CD sounded edgy and generally grating, screechy hi-hats, jagged brass and typical lack of bass, including what I'd call no 'woof' (percussive bass factor?), but certainly had a fair amount of extra detail and pleasant overall imaging/separation compared to vinyl.

The vinyl transfer to 24/96 PCM audio was certainly subjectively much nicer sounding than its CD counterpart, but obviously subject to the same limitations as its vinyl source.

The up-sampled tape to 24/96 seemed to have the best of all worlds: the cleanliness of its original CD source, along with the 'glue' that passing things through tape seem to bring, overall effortlessly smooth sound, and the kind of dynamic range that really made listening to the music a joy.

I realize that I should have included 16-bit / 44.1 kHz versions of the tape up-sample and vinyl transfers to really make this more of a valid test, but this is all I did. Not having access to DSD recording gear, this was obviously not included in the test. (let me not fail to thank Sony for its continued lack of support for the very formats it invents)

Quite empirical, for sure, but at least it left me convinced that there is a place for all of those extra bits, and that somehow judicious analog re-mastering can improve just about any sterile-sounding 16-bit digital source.

The last bit I really noticed is that when passing the pure digital source through a tube line amp, (didn't have the luxury to try a stand-alone tube D/A where I did this) it predictably smoothed things a bit more, and made the Red-Book CD at least listenable. The up-sampled 24/96 via tape sounded absolutely glorious through tubes on playback.

My little moral of the story also seems to be that anything that 'reconstitutes' the signal on playback, even if totally adding distortion and other artifacts appears to be the real important factor to pleasing the human ear in large-scale listening situations. The bigger the space, the more you notice it. For most people's everyday home listening situations, all of this stuff would probably make no difference whatsoever.

There you go, I hope some of you can see beyond my probable failure to follow some of the site's TOS for the bit of interesting data it may hold for those who care about its applications in similar large acoustic spaces.

I would of course welcome hearing any comments on this...

D.
Title: Why 24bit/48kHz/96kHz/
Post by: Axon on 2007-04-08 06:12:02
I will freely admit to having roughly a millionth of the production experience that you have, but from what I would expect, wouldn't the listening environment for a PA system involve far more distortion than for a high-quality nearfield monitor, or high-quality headphones? At least from a THD/frequency response point of view?

If so, that points to a highly theoretical and abstract mechanism of audibility. Extremely large variations in frequency response (and THD) could do crazy things with the ATH, so that certain changes to the sound that are inaudible with a perfect audio system become audible in distorting system. The "usual" example around here is that certain LAME encoder settings are quite transparent when using stereo material, but are substantially worse on Dolby Pro Logic stuff. Another example is that the presence of even order harmonic distortion could increase the audibility of absolute polarity reversal.

That said, unless your room with a 15kw PA system has a +40db high-Q peak somewhere, or intermodulates 10-100% at ultrasonic frequencies, I'm a little skeptical of the testing approach. Intuitively, the numeric distortions needed to bring high-res PCM accuracy to the audible range would have to be a little wild.

Quote
-1) CD Red-Book Audio version of song  DAC from Pioneer CDJ-1000 Mk 2 player.
-2) vinyl version of same. (obviously, depending on cartridge, tonearm calibration, etc.. used on test was standard DJ-grade Ortofon cart + Technics SL-1200 turntable)
-3) File up-sampled via 15 ips tape recording to 24/96 from workstation. AD from Metric Halo 2882 - DAC from RME Multiface.
-4) recording of vinyl at 24/96.  AD from Metric Halo 2882 - DAC from RME Multiface


Out of curiousity, do you know the mastering changes between the the CD and vinyl releases of this record?
Title: Why 24bit/48kHz/96kHz/
Post by: mdmuir on 2007-04-08 10:00:40
"The last bit I really noticed is that when passing the pure digital source through a tube line amp, (didn't have the luxury to try a stand-alone tube D/A where I did this) it predictably smoothed things a bit more, and made the Red-Book CD at least listenable. The up-sampled 24/96 via tape sounded absolutely glorious through tubes on playback."

Which from my (admitted) laymens knowledge means, you prefer the sound of higher harmonic distortion.
Title: Why 24bit/48kHz/96kHz/
Post by: Madman2003 on 2007-04-08 11:10:58
"The last bit I really noticed is that when passing the pure digital source through a tube line amp, (didn't have the luxury to try a stand-alone tube D/A where I did this) it predictably smoothed things a bit more, and made the Red-Book CD at least listenable. The up-sampled 24/96 via tape sounded absolutely glorious through tubes on playback."

Which from my (admitted) laymens knowledge means, you prefer the sound of higher harmonic distortion.


Specifically, even order harmonics. Only tube amps have this. Not a 100% sure, so don't take this as a fact.
Title: Why 24bit/48kHz/96kHz/
Post by: tommypeters on 2007-04-08 23:35:21
Here is some food for thought: http://www.meridian-audio.com/w_paper/Coding2.PDF (http://www.meridian-audio.com/w_paper/Coding2.PDF)

...though since it doesn't agree with what seems to be the stipulated view here, I guess it will be attacked on all fronts

And no, I don't think 16 bits are enough.
And yes, I think 192kHz is total overkill.
Title: Why 24bit/48kHz/96kHz/
Post by: pdq on 2007-04-09 01:19:08
When you have the same signal being reproduced through multiple speakers at different distances from the listener, obviously you will have very complex combining of the audio, increasing some frequencies and canceling others. Clearly the normal psychacoustics of a lossy encoder are not meant to compensate for this.

As for the OP's other observations, I have little interest in how adding this or that distortion makes the sound more or less pleasing to him. And comparing vinyl to CD of the same material when nothing is known about how the two were mastered is pointless.

My interest in this forum was and continues to be how lossy codecs can be made to reproduce stereo material played in a normal listening environment with the least audible difference from the original.
Title: Why 24bit/48kHz/96kHz/
Post by: singaiya on 2007-04-09 05:30:03
I agree totally - especially on the last sentence. 
Title: Why 24bit/48kHz/96kHz/
Post by: krabapple on 2007-04-09 05:58:27
Here is some food for thought: http://www.meridian-audio.com/w_paper/Coding2.PDF (http://www.meridian-audio.com/w_paper/Coding2.PDF)

...though since it doesn't agree with what seems to be the stipulated view here, I guess it will be attacked on all fronts

And no, I don't think 16 bits are enough.
And yes, I think 192kHz is total overkill.


That paper's been linked to several times on HA in the past few years.

Now, find us the published double blind listening tests results that support Stuart's assertions.
Title: Why 24bit/48kHz/96kHz/
Post by: Woodinville on 2007-04-09 18:22:05

"The last bit I really noticed is that when passing the pure digital source through a tube line amp, (didn't have the luxury to try a stand-alone tube D/A where I did this) it predictably smoothed things a bit more, and made the Red-Book CD at least listenable. The up-sampled 24/96 via tape sounded absolutely glorious through tubes on playback."

Which from my (admitted) laymens knowledge means, you prefer the sound of higher harmonic distortion.


Specifically, even order harmonics. Only tube amps have this. Not a 100% sure, so don't take this as a fact.


Depends on the kind of tube amp. Single ended amps have primarily 2nd order, with all orders present. Push-pull amps have all even-order harmonics very nearly cancelled, it's almost all odd-order distortions.
LP's have very strong assymetric (i.e. even order and odd order both) nonlinearities.
Title: Why 24bit/48kHz/96kHz/
Post by: nrand on 2008-05-01 12:20:45
[/quote]
Most studios know that recording at > 44.1 kHz is pretty useless, unless it is a recording for DVD-A or SACD purposes. All studios I know record simply at 44.1 kHz / 24 bit. There are no artists that I know of that record at 192 kHz to give the end user a fuzzy feeling
[/quote]

OK so with my home recording gear limitations I basically have two options:
44.1/16 bit or 48/24 bit
Which would you opt for and why?
Title: Why 24bit/48kHz/96kHz/
Post by: pdq on 2008-05-01 12:24:25
Quote

Most studios know that recording at > 44.1 kHz is pretty useless, unless it is a recording for DVD-A or SACD purposes. All studios I know record simply at 44.1 kHz / 24 bit. There are no artists that I know of that record at 192 kHz to give the end user a fuzzy feeling


OK so with my home recording gear limitations I basically have two options:
44.1/16 bit or 48/24 bit
Which would you opt for and why?

If it is going straight to CD as-is then 44.1/16. If you plan to do any kind of manipulation first then 48/24.

Edit: fixed quites.
Title: Why 24bit/48kHz/96kHz/
Post by: greynol on 2008-05-01 17:50:52
The video world works at 48kHz.

If you were just planning on dealing with CDDA, it would be better to work at 44.1/24 than 48/24 (EDIT: if given the option), no?
Title: Why 24bit/48kHz/96kHz/
Post by: andrew & david on 2008-05-02 15:02:20
Linn Records offer 24bit 96Khz recordings

http://www.linnrecords.com/linn-help-downl...-downloads.aspx (http://www.linnrecords.com/linn-help-downloads-studio-master-quality-album-downloads.aspx)

A lot of concerts on DVD have a 24bit 48 or 96 khz LPCM stereo audio track.

Is this just marketing or maybe when you need to do some post manipulation or processing it degrades the sound less at higher bit rates prior to downsampling.

I have done some AB comparisions with 24/96 vs 16/44.1 and the difference are subtle on certain music and not noticable at all on other music.
Title: Why 24bit/48kHz/96kHz/
Post by: Slipstreem on 2008-05-02 16:05:40
Was this AB testing or ABX testing? ABX testing eliminates the placebo effect. AB testing doesn't.

Cheers, Slipstreem. 
Title: Why 24bit/48kHz/96kHz/
Post by: andrew & david on 2008-05-02 16:20:15
Was this AB testing or ABX testing? ABX testing eliminates the placebo effect. AB testing doesn't.

Cheers, Slipstreem. 


This was ABX, with 2 people accessing the differences and a 3rd changing source. 

Was using a wonderfull recording on David Gilmore Live DVD which has a 24/96 LPCM stereo of the concert. Ripped from the DVD and converted to 16/44.1 using Cakewalk and left as 24/96. It's one of the best recordings i have ever heard, highly recomended. Started me off in the interest of looking for LPCM music off DVD's.
Title: Why 24bit/48kHz/96kHz/
Post by: Slipstreem on 2008-05-02 16:35:27
Fair enough. I guess that the point of my post was to remind you to state clearly when ABX testing has been carried out in preference to AB testing. This forum can be a foreboding place to a newcomer so I thought I'd nudge you politely before any less forgiving members 'stuck the boot in', so to speak. We like to deal in certifiable facts. 

Cheers, Slipstreem. 
Title: Why 24bit/48kHz/96kHz/
Post by: pdq on 2008-05-02 17:56:52

Was this AB testing or ABX testing? ABX testing eliminates the placebo effect. AB testing doesn't.

Cheers, Slipstreem. 


This was ABX, with 2 people accessing the differences and a 3rd changing source. 

You also need to provide numerical results to show that it is statistically significant.
Title: Why 24bit/48kHz/96kHz/
Post by: KikeG on 2008-05-02 22:43:54


Was this AB testing or ABX testing? ABX testing eliminates the placebo effect. AB testing doesn't.

Cheers, Slipstreem. 


This was ABX, with 2 people accessing the differences and a 3rd changing source. 

You also need to provide numerical results to show that it is statistically significant.

Also, were the sources level-matched within 0.1 dB? And properly time-aligned?