HydrogenAudio

Hydrogenaudio Forum => General Audio => Topic started by: shakeshuck on 2017-04-23 10:21:17

Title: PCM, DSD - Trying to get my head round some basics
Post by: shakeshuck on 2017-04-23 10:21:17
I've recently revived an interest in music/equipment, but am only slowly putting the pieces together regarding the PCM/DSD discussions.

My main query is regarding the value of a sampled bit. There are plenty of references that state things like '24 bits has a higher dynamic range than 16 bits'. Why does a bit sample always have to have the same amplitude value? Why can't you expand (say) 90dB into 24 bits instead of 16 bits, and get higher resolution?

With DSD, is a 0/1 bit switch the same amplitude as one of the 65535 16 bit bits?
DSD is quoted as having a better impulse response than PCM. Surely though (at least in theory) PCM can go from 0-65535 in one sample, whereas DSD would take 65535 samples to do the same? So DSD would have to be at 65535 times the sample rate to get the same response?
Obviously in practice you wouldn't get these extreme values, but what is the reality of the speed of a transient? DSD would only be good with transients of low amplitude.

Or have I got everything completely bass-ackwards?
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: andy o on 2017-04-23 10:37:59
I'm far from one of the experts here, but as I understand bit depth and dB:

I think you are confusing dB and dB SPL (https://en.wikipedia.org/wiki/Sound_pressure#Sound_pressure_level). The former is a relative relationship between two values, the latter indicates absolute sound pressure level, as measured in dB relative to the lowest threshold of hearing. So regarding your question, it's just a matter of mathematics. the 144dB theoretical number for 24-bit is just 2 different ways of saying the same thing. You double the dynamic range with each bit, and you double the dynamic range every roughly 6 dB. Hence, 24-bit translates to 144 dB. I might be using dynamic range instead of signal-to-noise ratio interchangeably, and it might be wrong (I can never seem to grasp the subtle difference), but you get the idea.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: shakeshuck on 2017-04-23 10:52:32
[...] You double the dynamic range with each bit [...]

But why? Isn't it just some arbitrary scale?
Why can't it be 1.6x (or whatever) per bit?
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: tedsmith on 2017-04-23 11:38:38
You misunderstand DSD.  DSD, like PCM can go from one rail to the other as fast as the reconstruction filter allows.

In (16 bit) PCM you'd go from -32768 to 32767 in one sample and there'd be a reconstruction filter at 1/2 the sample rate after that: e.g. for 44.1k there'd be a 22.05k filter that would limit the speed that that step could happen.

In DSD you'd go from a series of 0's to a series of 1's in one sample and then thru the low pass filter, which for DSD is usually a shallower filter than PCM but starting at, say, 50k or 80k instead of 22.05k for 16/44.1k PCM.  So the transition is faster in DSD.  DSD can transition faster than 24/96 but about the same as 24/192k.  (a maximum value in DSD isn't really all ones, but the slightly more complicated complete picture doesn't affect the above explanation.)

You could invent many different encodings of the bits, PCM and DSD are but two.  They represent the extremes - in PCM each bit of a sample represents twice the value of the previous bit, the maximum value for just two symbols.  In DSD all bits have exactly the same value.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: dhromed on 2017-04-23 12:18:28
It's basic counting in a base-2 system.
1-bit -> 2 possible values: 0, 1
2-bit -> 4 possible values: 00, 01, 10, 11
3-bit -> 8 possible values: 000, 001, 010 etc etc
4-bit -> 16 possible values
etcetera

If you use more bits for the signal, you have more resolution to express smaller, quieter signals. A potentially larger difference between the loudest and quietest signal means a larger dynamic range. But increasing the resolution of an existing signal obviously does not magically increase the resolution of its contents. It's just being stretched across the new space.

Decibel is a relative scale that doubles values every time you add 6. That number comes from the logarithmic math, so it's not a random number someone chose. So if you have an ampitude of x, then adding 6 dB means your amplitude is now 2x. Conversely, increasing the bit-depth by 1 means you make room for small signals that have half the amplitude of the previous smallest signal, and lo and behold, half the amplitude is an extra 6dB range.

My knowledge of the actual mathematics of dB is shallow at best, so I can't go much deeper than that.

> Why does a bit sample always have to have the same amplitude value?
It doesn't. It's all relative to 1, also called fullscale, also called 100% volume. The only absolute sense of amplitude is in the real world with pressure levels in air and voltages in the wire. When we say 16 bits has X dB dynamic range, it means the difference between the largest and smallest storeable signal is X dB. If you want to put that in an amplifier and blast it or play it quietly, sure, that doesn't change the relative amplitudes of what's coming out of your speakers.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: shakeshuck on 2017-04-23 13:02:20
You misunderstand DSD.  DSD, like PCM can go from one rail to the other as fast as the reconstruction filter allows.
[...]
In DSD you'd go from a series of 0's to a series of 1's in one sample [...]

So one sample can contain many 0's or many 1's? Now my head really hurts.
I thought it was 1 bit x sample rate?
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: tedsmith on 2017-04-23 13:16:23
No, DSD is really simple: a stream of ones and zeros.  A one means to head for the positive voltage rail as fast as possible and a zero means head for the negative rail as fast as possible.  The speed of movement is restricted by a low pass filter.  - That's it.  All you need to play DSD is a low pass filter on the stream of bits.

For example 0101010101 for ever represents a 0 output (as does 1100110011001100...)  The stream 0001000100010001... represents a value 1/2 of the way from a zero output to the maximum negative output...  Tho these patterns flop around, by the time they go thru the low pass filter that noise is filtered out.

Calculating those bits is very weird, but one method that's very inefficient but really does work is to search thru all streams of bits looking for the one that (after being low pass filtered) gets you closest to the signal you want.  Usually a sigma delta modulator is used: one way of thinking of a SDM is to compare the low pass filtered stream of bits generated so far to the input signal and produce a one if the stream is too low and a zero if the stream is too high - then repeat forever.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: Arnold B. Krueger on 2017-04-23 13:29:56
I've recently revived an interest in music/equipment, but am only slowly putting the pieces together regarding the PCM/DSD discussions.

There is really very little to discuss. It is basically a PCM world, and DSD is a little placebophile side show that you can safely ignore.

If properly implemented, PCM  and DSD sound exactly alike

For about the last 20 years just about all of our DACs have been converting PCM to DSD under the covers prior to turning the data into analog, and nobody but a few technical specialists knew or cared.  The code name for this process is Delta-Signal or Sigma-Delta.

DSD was taken out from under the covers in order to provide IP protection. Some marketing genius decided to mine the same old reliable but fraudulent vein of gold that the placebophiles have been mining for decades, and claim that DSD "Sounded Better".  It doesn't and if you take what I said about almost all of our DACs turning PCM into DSD as a stardard part of our modern way of ding business, to heart, you can see why.

Quote
My main query is regarding the value of a sampled bit. There are plenty of references that state things like '24 bits has a higher dynamic range than 16 bits'.

That is because it is true. But it does not follow that any music sampled with 24 bits has higher resolution than music sampled with 16 bits.  Both processes depend on the same analog data, and the same acoustical event. In general live music has less than 16 bits of resolution.

Quote
Why does a bit sample always have to have the same amplitude value?


It doesn't.  For example the most signfiicant bit of any PCM word has twice the value of the next one.

Quote
Why can't you expand (say) 90dB into 24 bits instead of 16 bits, and get higher resolution?

Because PCM words represent data, they don't define it. On a good day, they don't limit it, either.  

Data is what it is. You can reduce its resolution or preserve it, but you can't increase it without going back to the original physical process that it represents, and changing it.  That's why many of us have a hearty laugh about all those DSD and 24 bit PCM files that are based on analog tapes and legacy PCM files.  Interestingly enough, about half the SACD and DVD-A recordings released through about 2006 were this way, and no golden ear ever noticed it. Many of those recordings are still being sold, and they have been joined by a host of newer recordings that had the same limitations simply because of the usual forces of nature.

Virtually every so-called "Hi Rez" recording has the bits but not the data to be true to its name.

Quote
With DSD, is a 0/1 bit switch the same amplitude as one of the 65535 16 bit bits?

No.

In DSD there is only one bit and it is switched very rapidly on and off to provide the same kind of handling of the various signal levels as we have with 16 or 24 or whatever bits in PCM.

Quote
DSD is quoted as having a better impulse response than PCM.

...As if impulse response was that relevant to sound quality, which in general it is not...

If you had exactly comparable DSD and PCM systems, the impulse response would be the same. But, as implemented, the systems are not the same. So, they have different impulse responses.

Quote
Surely though (at least in theory) PCM can go from 0-65535 in one sample, whereas DSD would take 65535 samples to do the same?

Well, a sample in DSD is not just one bit.  The parity you suggest is real, so a DSD data stream has to have a whole lot more bits in it to do the identical same thing as a comparable PCM data stream.  The PCM data stream weights the bits with binary values, and the DSD stream does not.

Quote
So DSD would have to be at 65535 times the sample rate to get the same response?

True if all things were equal but they are not all equal in the commercial implementations.

Quote
Obviously in practice you wouldn't get these extreme values, but what is the reality of the speed of a transient? DSD would only be good with transients of low amplitude.

Like I said, if all things were equal, but...

Quote
Or have I got everything completely bass-ackwards?

More like comparing apples with oranges than bass ackwards.  If you really are a newbie, you would do well to concentrate on other issues, such as the ones that actually matter for sound quality. Things like acoustics...

Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: pelmazo on 2017-04-23 13:37:07
Ouch, my entire post was somehow lost. I don't feel like typing this all over again... :-(
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: bennetng on 2017-04-23 13:41:33
Ouch, my entire post was somehow lost. I don't feel like typing this all over again... :-(

Try My Account -> Account Settings -> Profile Info -> Show Drafts
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: shakeshuck on 2017-04-23 14:04:11
If you use more bits for the signal, you have more resolution to express smaller, quieter signals. A potentially larger difference between the loudest and quietest signal means a larger dynamic range. But increasing the resolution of an existing signal obviously does not magically increase the resolution of its contents. It's just being stretched across the new space.

Decibel is a relative scale that doubles values every time you add 6. That number comes from the logarithmic math, so it's not a random number someone chose. So if you have an ampitude of x, then adding 6 dB means your amplitude is now 2x. Conversely, increasing the bit-depth by 1 means you make room for small signals that have half the amplitude of the previous smallest signal, and lo and behold, half the amplitude is an extra 6dB range.

My knowledge of the actual mathematics of dB is shallow at best, so I can't go much deeper than that.

> Why does a bit sample always have to have the same amplitude value?
It doesn't. It's all relative to 1, also called fullscale, also called 100% volume. The only absolute sense of amplitude is in the real world with pressure levels in air and voltages in the wire. When we say 16 bits has X dB dynamic range, it means the difference between the largest and smallest storeable signal is X dB. If you want to put that in an amplifier and blast it or play it quietly, sure, that doesn't change the relative amplitudes of what's coming out of your speakers.

I'm afraid I'm having problems interpreting this post.  :(

Surely, either the scale is logarithmic (and double sound pressure is double on the scale), and therefore the folks saying that "anything over 20-bits is a waste of time as it is beyond human endurance" are correct, OR we are capturing an amplitude of -1 to 1 in n resolution, regardless of the amplitude max sound level, then we get a higher resolution to "to express smaller, quieter signals".

Are these two things not contradictory?
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: shakeshuck on 2017-04-23 14:07:28
No, DSD is really simple: a stream of ones and zeros.  A one means to head for the positive voltage rail as fast as possible and a zero means head for the negative rail as fast as possible.  The speed of movement is restricted by a low pass filter.  - That's it.  All you need to play DSD is a low pass filter on the stream of bits.

For example 0101010101 for ever represents a 0 output (as does 1100110011001100...)  The stream 0001000100010001... represents a value 1/2 of the way from a zero output to the maximum negative output...  Tho these patterns flop around, by the time they go thru the low pass filter that noise is filtered out.

Calculating those bits is very weird, but one method that's very inefficient but really does work is to search thru all streams of bits looking for the one that (after being low pass filtered) gets you closest to the signal you want.  Usually a sigma delta modulator is used: one way of thinking of a SDM is to compare the low pass filtered stream of bits generated so far to the input signal and produce a one if the stream is too low and a zero if the stream is too high - then repeat forever.

Ah, I see now. That's not how I thought it worked.
Thanks.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: shakeshuck on 2017-04-23 14:21:14
I've recently revived an interest in music/equipment, but am only slowly putting the pieces together regarding the PCM/DSD discussions.

There is really very little to discuss. It is basically a PCM world, and DSD is a little placebophile side show that you can safely ignore.

If properly implemented, PCM  and DSD sound exactly alike

For about the last 20 years just about all of our DACs have been converting PCM to DSD under the covers prior to turning the data into analog, and nobody but a few technical specialists knew or cared.  The code name for this process is Delta-Signal or Sigma-Delta.

DSD was taken out from under the covers in order to provide IP protection. Some marketing genius decided to mine the same old reliable but fraudulent vein of gold that the placebophiles have been mining for decades, and claim that DSD "Sounded Better".  It doesn't and if you take what I said about almost all of our DACs turning PCM into DSD as a stardard part of our modern way of ding business, to heart, you can see why.

Quote
My main query is regarding the value of a sampled bit. There are plenty of references that state things like '24 bits has a higher dynamic range than 16 bits'.

That is because it is true. But it does not follow that any music sampled with 24 bits has higher resolution than music sampled with 16 bits.  Both processes depend on the same analog data, and the same acoustical event. In general live music has less than 16 bits of resolution.

Quote
Why does a bit sample always have to have the same amplitude value?


It doesn't.  For example the most signfiicant bit of any PCM word has twice the value of the next one.

Quote
Why can't you expand (say) 90dB into 24 bits instead of 16 bits, and get higher resolution?

Because PCM words represent data, they don't define it. On a good day, they don't limit it, either.  

Data is what it is. You can reduce its resolution or preserve it, but you can't increase it without going back to the original physical process that it represents, and changing it.  That's why many of us have a hearty laugh about all those DSD and 24 bit PCM files that are based on analog tapes and legacy PCM files.  Interestingly enough, about half the SACD and DVD-A recordings released through about 2006 were this way, and no golden ear ever noticed it. Many of those recordings are still being sold, and they have been joined by a host of newer recordings that had the same limitations simply because of the usual forces of nature.

Virtually every so-called "Hi Rez" recording has the bits but not the data to be true to its name.

Quote
With DSD, is a 0/1 bit switch the same amplitude as one of the 65535 16 bit bits?

No.

In DSD there is only one bit and it is switched very rapidly on and off to provide the same kind of handling of the various signal levels as we have with 16 or 24 or whatever bits in PCM.

Quote
DSD is quoted as having a better impulse response than PCM.

...As if impulse response was that relevant to sound quality, which in general it is not...

If you had exactly comparable DSD and PCM systems, the impulse response would be the same. But, as implemented, the systems are not the same. So, they have different impulse responses.

Quote
Surely though (at least in theory) PCM can go from 0-65535 in one sample, whereas DSD would take 65535 samples to do the same?

Well, a sample in DSD is not just one bit.  The parity you suggest is real, so a DSD data stream has to have a whole lot more bits in it to do the identical same thing as a comparable PCM data stream.  The PCM data stream weights the bits with binary values, and the DSD stream does not.

Quote
So DSD would have to be at 65535 times the sample rate to get the same response?

True if all things were equal but they are not all equal in the commercial implementations.

Quote
Obviously in practice you wouldn't get these extreme values, but what is the reality of the speed of a transient? DSD would only be good with transients of low amplitude.

Like I said, if all things were equal, but...

Quote
Or have I got everything completely bass-ackwards?

More like comparing apples with oranges than bass ackwards.  If you really are a newbie, you would do well to concentrate on other issues, such as the ones that actually matter for sound quality. Things like acoustics...

There are proponents of both sides.
I like to try to gather information so I can make an informed decision, instead of blindly following one herd or the other - and understanding how these things work hopefully goes a long way to cut through the misinformation.

From the few samples I've downloaded so far, I can understand where the two camps come from, but I'm not sure the difference is worth fighting over.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: Arnold B. Krueger on 2017-04-23 14:45:42
There are proponents of both sides.

Right, but this isn't politics, it is science.  This isn't new controversial science, it is established science. And the confirming or denying experiments aren't rocket science or take a million dollars worth of equipment to do. The science for testing these things is well known, commonly practiced, and can be done for almost no money if you are interested.

Quote
I like to try to gather information so I can make an informed decision, instead of blindly following one herd or the other - and understanding how these things work hopefully goes a long way to cut through the misinformation.

The science says, PCM and DSD is compared on an apples-to-apples basis must sound the same. You'll probably never find a DSD proponent admitting that in public, but you will find a lot of people are hip to audio and are well-informed about science saying so. Go figure.

Quote
From the few samples I've downloaded so far, I can understand where the two camps come from, but I'm not sure the difference is worth fighting over.

Oh, you'll hear or at least you'll think you hear a lot of differences. The DSD proponents have a number of ways to trick you into doing bad listening experiments where there are hidden and not-so-hidden audible variables like differences in mastering.  There is big money on the table or at least some people think so.

Fact is just about every high-rez initiative fails to deliver on its perceived promise. Notice that SACD and DVD-A aren't in what's left of local record stores like Best Buy. Note that Pono's music site went bankrupt and only used an NOS players are all that seems to be available.

Note that there are about 400 different DAC on the market, many featuring DSD or other hi-rez support but they  are languishing in the marketplace and none of them really went mainstream. Hint: The Emperor has no clothes and many people quietly see it and just keep their money where it was or spend it on something else.

All by yourself, as relatively easy as it really is, most people resist doing proper listening tests. I just got called a bunch of dirty names for politely pointing that out on another thread. But know it or not, you signed up to agree to that when you registered here. Please see TOS 8.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: bennetng on 2017-04-23 15:12:05
I'm afraid I'm having problems interpreting this post.  :(

Surely, either the scale is logarithmic (and double sound pressure is double on the scale), and therefore the folks saying that "anything over 20-bits is a waste of time as it is beyond human endurance" are correct, OR we are capturing an amplitude of -1 to 1 in n resolution, regardless of the amplitude max sound level, then we get a higher resolution to "to express smaller, quieter signals".

Are these two things not contradictory?

It relates to the operating voltage range of the recording and playback device, it is on the analog side.

Let's ignore the digital part, now I have an analog tape recorder, mixer or anything which are purely analog. If you look at the manual and find the specs, it will say something like max level: +20dBu, 8V RMS and so on. If you feed a +30dBu, or a 10V signal to these devices, will result in distortion. Anything, unless digitally synthesized and will never be played on speakers or headphones, are limited by the analog part.

Now the bit depth part is simple, it simply refer to how this +20dBu or 8V signal can be divided into 65536 steps, for example, when recorded in 16-bit. Now if you play the recorded file from a consumer device, like a mobile phone, which is less powerful, for example, only 0.5V, and the phone has a 16-bit DAC, then this 65536 steps are just remapped to a less powerful state.

It is just like a 1080p mobile phone vs a 1080p TV.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: shakeshuck on 2017-04-23 15:33:28
All by yourself, as relatively easy as it really is, most people resist doing proper listening tests. I just got called a bunch of dirty names for politely pointing that out on another thread. But know it or not, you signed up to agree to that when you registered here. Please see TOS 8.

I didn't realise that I'd taken sides. I thought I'd deliberately been libertarian about the whole thing; not offending anyone or poking any cages by sitting on the fence. Now I'm being chastised for it?

I once (years ago) brought up the subject of CD burning, and that in an experiment I had done the newly burned CD sounded better than the stamped original. I didn't know why, it just did as far as I was concerned. I was pounded for my efforts, as "science says it isn't so", quoted Reed-Solomon etc, etc.
Recently I found out that Reed-Solomon is only effective if the disc is tracking properly, and mistracking could easily have been the reason for the difference in sound. I have seen many discs where the punched centre hole is completely misaligned with the data layer on the disc.

Because of my own experiences, I don't try to tell anyone which side of the fence to sit on, that's up to them. I am happy for anyone to believe what they like until science changes its mind and proves otherwise.

That's why I'm asking for help in understanding the processes.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: shakeshuck on 2017-04-23 15:43:55
I'm afraid I'm having problems interpreting this post.  :(

Surely, either the scale is logarithmic (and double sound pressure is double on the scale), and therefore the folks saying that "anything over 20-bits is a waste of time as it is beyond human endurance" are correct, OR we are capturing an amplitude of -1 to 1 in n resolution, regardless of the amplitude max sound level, then we get a higher resolution to "to express smaller, quieter signals".

Are these two things not contradictory?

It relates to the operating voltage range of the recording and playback device, it is on the analog side.

Let's ignore the digital part, now I have an analog tape recorder, mixer or anything which are purely analog. If you look at the manual and find the specs, it will say something like max level: +20dBu, 8V RMS and so on. If you feed a +30dBu, or a 10V signal to these devices, will result in distortion. Anything, unless digitally synthesized and will never be played on speakers or headphones, are limited by the analog part.

Now the bit depth part is simple, it simply refer to how this +20dBu or 8V signal can be divided into 65536 steps, for example, when recorded in 16-bit. Now if you play the recorded file from a consumer device, like a mobile phone, which is less powerful, for example, only 0.5V, and the phone has a 16-bit DAC, then this 65536 steps are just remapped to a less powerful state.

It is just like a 1080p mobile phone vs a 1080p TV.


That's how I thought it should work. Any input is split into whatever resolution you want (16/24/32).
The folks propagating "extra bits = greater dynamic range" as the ONLY reason for greater bit depth are wrong.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: Wombat on 2017-04-23 15:51:35
I once (years ago) brought up the subject of CD burning, and that in an experiment I had done the newly burned CD sounded better than the stamped original. I didn't know why, it just did as far as I was concerned. I was pounded for my efforts, as "science says it isn't so", quoted Reed-Solomon etc, etc.
Recently I found out that Reed-Solomon is only effective if the disc is tracking properly, and mistracking could easily have been the reason for the difference in sound. I have seen many discs where the punched centre hole is completely misaligned with the data layer on the disc.
I fear you imagine things like you will imagine other things relating to the sound of DSD if you only find a written argument you feel that convinces you. Waste of time.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: shakeshuck on 2017-04-23 15:55:26
I once (years ago) brought up the subject of CD burning, and that in an experiment I had done the newly burned CD sounded better than the stamped original. I didn't know why, it just did as far as I was concerned. I was pounded for my efforts, as "science says it isn't so", quoted Reed-Solomon etc, etc.
Recently I found out that Reed-Solomon is only effective if the disc is tracking properly, and mistracking could easily have been the reason for the difference in sound. I have seen many discs where the punched centre hole is completely misaligned with the data layer on the disc.
I fear you imagine things like you will imagine other things relating to the sound of DSD if you only find a written argument you feel that convinces you. Waste of time.

My point exactly.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: greynol on 2017-04-23 16:05:09
Recently I found out that Reed-Solomon is only effective if the disc is tracking properly, and mistracking could easily have been the reason for the difference in sound.
Hogwash.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: shakeshuck on 2017-04-23 16:11:01
All by yourself, as relatively easy as it really is, most people resist doing proper listening tests.

This is not necessarily as easy as you make out. My first post on HAudio was in the ABX section for Foobar, which was making comparisons impossible as ReplayGain wouldn't set the levels between PCM and DSD correctly (possibly due to the noise, but I won't speculate as I get told off for doing that).

So I HAVE tried a proper listening test, and the technology failed me.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: shakeshuck on 2017-04-23 16:13:52
Recently I found out that Reed-Solomon is only effective if the disc is tracking properly, and mistracking could easily have been the reason for the difference in sound.
Hogwash.

Why do CD ripping programs have authentication databases if they get it right every time?
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: Arnold B. Krueger on 2017-04-23 16:19:41
All by yourself, as relatively easy as it really is, most people resist doing proper listening tests. I just got called a bunch of dirty names for politely pointing that out on another thread. But know it or not, you signed up to agree to that when you registered here. Please see TOS 8.

I didn't realise that I'd taken sides. I thought I'd deliberately been libertarian about the whole thing; not offending anyone or poking any cages by sitting on the fence. Now I'm being chastised for it?

Are you being chastised?  It was my intent to not chastise you just as surely as it was your intent to not take sides. ;-)

Here's a challenge - quote me chastising you. Bear in mind that I consider myself to be a member of the group I sometimes refer to as  "Most people". So, when I talked about "most people" I wasn't singling out you, I was talking about us.  Even though I've done a ton of DBTs, I resist them, probably as much as anybody. Difference is, I get over it some of the time.

Quote
I once (years ago) brought up the subject of CD burning, and that in an experiment I had done the newly burned CD sounded better than the stamped original. I didn't know why, it just did as far as I was concerned. I was pounded for my efforts, as "science says it isn't so", quoted Reed-Solomon etc, etc.

I would put that under the category "a little knowledge is a dangerous thing". You didn't deserve to get to be pounded because what you described is possible, if one understands more practical and theoretical details about CDs. Your correspondents didn't know the whole story.  Let me start it out like this.

What happens when there is a detected error reading a CD on a typical CD player? From a user standpoint there has been an organized, well-funded effort that has been going on for decades to keep you from knowing that there was an error. So as a user you often don't know there was an error  and have no practical way to find out.  Fact is that many CD players have an electrical line that changes state when there is an error reading a CD, and it is designed to be probed with an oscilloscope. I haven't looked for one or probed it lately, so this may not be true of the latest-greatest optical players. Back in the day I spent some time watching this line with my oscilloscopes., while I was playing various CDs. I even simulated damaged CDs by cutting narrow strips of black electrical tape to simulate damage to the disc.

There is a difference between ripping CDs and playing CDs on a regular optical disc player which is that the CD plays every track once and tries to do the best it can on the fly. When you rip the CD on a computer, most good ripping software watches for errors and tries to recover them by any number of means, most of which involve rereading the track until there is no error. Failing that it will reread the track many times and take advantage of the fact that the errors are inconsistent and  try to get a best average of the reads  and put that into the output file.

So knowing that, on several occasions I have taken  CDs that were damaged and played poorly on some CD player, ripped them, burned a CD and it played very nicely on that same player,  thank you!

Quote
Recently I found out that Reed-Solomon is only effective if the disc is tracking properly, and mistracking could easily have been the reason for the difference in sound.

I'll take that to be a concise version of what I just wrote.

Quote
I have seen many discs where the punched centre hole is completely misaligned with the data layer on the disc.

CD players have built-in means to compensate for that precisely and obtain error-free results provided the off-center condition is not too large.  They don't even need to look at the Reed-Solomon, they get a precise read on the eccentricity from the laser head tracking system (part of it is like a speaker voice coil)  and the error  gets compensated for right there.

Quote
Because of my own experiences, I don't try to tell anyone which side of the fence to sit on, that's up to them. I am happy for anyone to believe what they like until science changes its mind and proves otherwise.

The science of digital audio as applied to CDs is pretty cut-and-dried and has been so  for a decade or more.  I can and have backed up everything I write with real world experiments, peer reviewed papers and textbooks, but I've already done so, so many times that getting me to do so today may not be easy. ;-)

Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: Arnold B. Krueger on 2017-04-23 16:26:16
All by yourself, as relatively easy as it really is, most people resist doing proper listening tests.

This is not necessarily as easy as you make out. My first post on HAudio was in the ABX section for Foobar, which was making comparisons impossible as ReplayGain wouldn't set the levels between PCM and DSD correctly (possibly due to the noise, but I won't speculate as I get told off for doing that).

So I HAVE tried a proper listening test, and the technology failed me.

You may have given up far easier than I would have. I no doubt have many resources, both hardware, software, training and experience-wise, that most people don't have. For example, I  distributed free Windows ABX software that I wrote from scratch towards the end of the last Millenium. It worked in its way but it was bad enough that it had exactly the results I hoped for. Many others wrote better ones.

 I consider the current version of FOOBAR2000 to be good enough to recommend all the time.

Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: greynol on 2017-04-23 16:26:45
Recently I found out that Reed-Solomon is only effective if the disc is tracking properly, and mistracking could easily have been the reason for the difference in sound.
Hogwash.

Why do CD ripping programs have authentication databases if they get it right every time?
How does that defend the position that a CD-R will provide better quality audio than the original disc from which it was sourced, catastrophic failures, not withstanding?

Or are you only talking about completely uninteresting trivial cases?
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: Arnold B. Krueger on 2017-04-23 16:29:02
Recently I found out that Reed-Solomon is only effective if the disc is tracking properly, and mistracking could easily have been the reason for the difference in sound.
Hogwash.

Why do CD ripping programs have authentication databases if they get it right every time?

Obviously, they don't get it right all the time, and nobody with a brain expects them to. ;-)

However, if you get fairly consistent AccuRIP errors, that's nature's way of telling you that you may need to clean up your act. This is often as simple as physically cleaning the CD.

Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: shakeshuck on 2017-04-23 16:45:08
Recently I found out that Reed-Solomon is only effective if the disc is tracking properly, and mistracking could easily have been the reason for the difference in sound.
Hogwash.

Why do CD ripping programs have authentication databases if they get it right every time?
How does that defend the position that a CD-R will provide better quality audio than the original disc from which it was sourced, catastrophic failures, not withstanding?

Or are you only talking about completely uninteresting trivial cases?

I think Mr Krueger's post gives an explanation better than I ever could.

Consider yourself countered, and a little bit rude. ;)
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: greynol on 2017-04-23 16:52:40
My cd-r sounds better because it can be played back without skips and dropouts doesn't exactly rise to a level of being worthy of discussion, not that Barry Diament's bits are not bits nonsense is worthy, either.

My bad assuming you were naively defending the latter.  Based on the direction this topic was heading (acceptance of arguments coming from the Camp of Audiophile Woo), I hope you can understand why I jumped to this conclusion.  I had no idea I was being countered in a game of tiddlywinks.  You win, yay!
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: drewfx on 2017-04-23 18:39:35
That's how I thought it should work. Any input is split into whatever resolution you want (16/24/32).
The folks propagating "extra bits = greater dynamic range" as the ONLY reason for greater bit depth are wrong.

A dB is a logarithmic representation of a ratio. If dB is used with a suffix (for instance dB SPL), it is telling you the ratio relative to a predefined value specified by the suffix.

Increasing the bit depth increases the ratio between the largest and smallest value that can be represented by a factor 2x per bit (= ~6dB/bit).

The important thing to note is that if the smallest values that can be represented in a digital audio format are buried under noise in the listening environment or other factors or below you threshold of hearing when played back at a given listening level, then in the real world increasing bit depth doesn't really do anything. IOW increasing bit depth is only useful if it's the thing that is actually limiting resolution in the real world.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: greynol on 2017-04-23 18:43:10
The important thing to note is that if the smallest values that can be represented in a digital audio format are buried under noise in the listening environment or other factors or below you threshold of hearing when played back at a given listening level, then in the real world increasing bit depth doesn't really do anything. IOW increasing bit depth is only useful if it's the thing that is actually limiting resolution in the real world.
Noise already present in the content is one of those "other factors" which can be just as significant as the noise present in the listening environment.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: shakeshuck on 2017-04-23 18:47:02
My cd-r sounds better because it can be played back without skips and dropouts doesn't exactly rise to a level of being worthy of discussion, not that Barry Diament's bits are not bits nonsense is worthy, either.

My bad assuming you were naively defending the latter.  Based on the direction this topic was heading (acceptance of arguments coming from the Camp of Audiophile Woo), I hope you can understand why I jumped to this conclusion.  I had no idea I was being countered in a game of tiddlywinks.  You win, yay!

Without starting afresh, I am not talking about skips and dropouts. Does the error correction not attempt to 'fill in' the gaps it can't read? Or am I thinking of another technology? It all becomes a blur after a while...
And I don't know why you keep thinking I'm pro-DSD when I'm trying to insist that I'm not taking sides!
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: greynol on 2017-04-23 19:15:07
It's quite a stretch to claim audible differences in a situation where tracking problems have made it too difficult to provide data without interpolation, yet not so difficult for there to be skips and dropouts.  If these aren't just speculative anecdotal musings on your part, I'd love to see some evidence which conforms to the standard you agreed to follow when registering here.

There are proponents of both sides.
I like to try to gather information so I can make an informed decision, instead of blindly following one herd or the other - and understanding how these things work hopefully goes a long way to cut through the misinformation.

From the few samples I've downloaded so far, I can understand where the two camps come from, but I'm not sure the difference is worth fighting over.
In the case of taking sides, again, I'll point to the standard and its logical conclusion: all things should be assumed to sound the same unless it can be demonstrated otherwise.  Lending credibility to a side that has failed to fulfill the necessary burden of proof required by a science-based community is often taken as an affront.  This has been exacerbated thanks to a society that is increasingly rejecting science and the scientific method so belligerently.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: shakeshuck on 2017-04-23 20:08:12
It's quite a stretch to claim audible differences in a situation where tracking problems have made it too difficult to provide data without interpolation, yet not so difficult for there to be skips and dropouts.  If these aren't just speculative anecdotal musings on your part, I'd love to see some evidence which conforms to the standard you agreed to follow when registering here.

There are proponents of both sides.
I like to try to gather information so I can make an informed decision, instead of blindly following one herd or the other - and understanding how these things work hopefully goes a long way to cut through the misinformation.

From the few samples I've downloaded so far, I can understand where the two camps come from, but I'm not sure the difference is worth fighting over.

In the case of taking sides, again, I'll point to the standard and its logical conclusion: all things should be assumed to sound the same unless it can be demonstrated otherwise.

As it was years ago, as I stipulated, we'll have to consider it an anecdote. Of course the intention of my mentioning it was supposed to be "in passing" and related to the strict black-and-white types that are always right, even though they haven't tested something for themselves. I agree with earlier sentiments that you can't argue against ABX if you're not willing to try it. It's the same principle.

Probably a moot point now anyway as manufacturing qualities seem to have improved no end since the old days...
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: greynol on 2017-04-23 20:12:07
As it was years ago, as I stipulated, we'll have to consider it an anecdote.
...and as such I'm not buying it. ;)

Quote
Of course the intention of my mentioning it was supposed to be "in passing" and related to the strict black-and-white types that are always right, even though they haven't tested something for themselves. I agree with earlier sentiments that you can't argue against ABX if you're not willing to try it. It's the same principle.
Agreed.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: tedsmith on 2017-04-23 20:33:12
Without starting afresh, I am not talking about skips and dropouts. Does the error correction not attempt to 'fill in' the gaps it can't read? Or am I thinking of another technology? It all becomes a blur after a while...
In ripping about 2000 CDs I only had a handful that got any non-correctly corrected errors on first read at high speed.  Error handling on music CDs isn't great, but it's pretty good.  Some older CD transports did extrapolate or zero fill on significant enough errors and that could cause audible artifacts, but these errors are quite rare for most undamaged CDs.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: shakeshuck on 2017-04-23 21:22:37
...and as such I'm not buying it. ;)

We are all who we are because of our life experiences. It wouldn't do to have us all the same now, would it?
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: greynol on 2017-04-23 21:36:48
Of course not, but as human beings we are all prone to the same imperfections that lead us to draw erroneous conclusions, including how we interpret sound.  There are methods to get around this and these methods are required when one wants to posit claims about differences in sound quality.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: saratoga on 2017-04-23 22:38:30
Dynamic range doubles when you add one more bit because you now have twice as many values you can produce.

Pcm spaces values uniformly in amplitude such that all levels are equally encoded. There are formats that space them non-uniformly, mostly as a means of lossy compression.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: andy o on 2017-04-24 02:47:15
All by yourself, as relatively easy as it really is, most people resist doing proper listening tests. I just got called a bunch of dirty names for politely pointing that out on another thread. But know it or not, you signed up to agree to that when you registered here. Please see TOS 8.

I didn't realise that I'd taken sides. I thought I'd deliberately been libertarian about the whole thing; not offending anyone or poking any cages by sitting on the fence. Now I'm being chastised for it?

[...]

Because of my own experiences, I don't try to tell anyone which side of the fence to sit on, that's up to them. I am happy for anyone to believe what they like until science changes its mind and proves otherwise.

It may not have been your intention, but proclaiming there are 2 sides and implying they're equivalent when the science is well established is a well known and tired cliche of science deniers. We see it regarding global warming, medicine, biology (evolution especially), and yes, even audio. It's the false equivalence fallacy.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: Palladium on 2017-04-24 04:57:14
Ah yes, the stealth woo-woo pusher masquerading as an victimized oh-so-innocent answer seeker. Definitely never happened around these parts.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: andy o on 2017-04-24 06:27:29
yeah, I mean, I wasn't accusing him of it, but sometimes it's just indistinguishable from past experiences.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: shakeshuck on 2017-04-24 09:20:05
All by yourself, as relatively easy as it really is, most people resist doing proper listening tests. I just got called a bunch of dirty names for politely pointing that out on another thread. But know it or not, you signed up to agree to that when you registered here. Please see TOS 8.

I didn't realise that I'd taken sides. I thought I'd deliberately been libertarian about the whole thing; not offending anyone or poking any cages by sitting on the fence. Now I'm being chastised for it?

[...]

Because of my own experiences, I don't try to tell anyone which side of the fence to sit on, that's up to them. I am happy for anyone to believe what they like until science changes its mind and proves otherwise.

It may not have been your intention, but proclaiming there are 2 sides and implying they're equivalent when the science is well established is a well known and tired cliche of science deniers. We see it regarding global warming, medicine, biology (evolution especially), and yes, even audio. It's the false equivalence fallacy.


Guys!
I've obviously got off on a bad start here.
It appears a glib remark has been allowed to run riot and get completely out of hand.

My apologies to Arnold K, who I now realise was trying to steer me in the direction required by the forum members, which I wrongly took as an attack at the time.
I am surprised at the vehemence on here though to anyone that doesn't hold the same 100% view as 'the gang' - how can that lead to interesting discussions? I haven't tried to force an opinion on anyone, in fact I deliberately tried not to.

I've had my appendix out. Oh wait! I can't prove it to you, therefore it didn't happen. Maybe the surgeon didn't really take it out, and I'm the victim of a conspiracy theory that I'm attempting to pass on.

No bad intended! Lighten up, guys!
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: shakeshuck on 2017-04-24 10:21:29
Ah yes, the stealth woo-woo pusher masquerading as an victimized oh-so-innocent answer seeker. Definitely never happened around these parts.

You guys are paranoid!
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: shakeshuck on 2017-04-24 12:10:02
At risk of getting my head bitten off again, here's another question that might appeal more to your reasoning:

Is there any Windows software (preferably free, or a trial version) where I can load a PCM and a DSD track simultaneously and subtract one from the other to display any difference (or NO difference!) between the two?
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: Arnold B. Krueger on 2017-04-24 12:40:44

Without starting afresh, I am not talking about skips and dropouts.

You could be talking about skips and dropouts  without knowing it.  Between the decades of development of methods for filling in gaps that can't be read and the ability of the human ear to accept them without notice, what you hear is not necessarily what you got.

Furthermore, what else could it be?  You hear this over and over again, and it, is true: Digital data either is perfectly correct or it isn't data at all. Basically the only fault that can intrude on the playback of optical disks is some kind of data loss. The audio part of the player can't and won't change timbre, pace or timing audibly  due to any probable fault.

Quote
Does the error correction not attempt to 'fill in' the gaps it can't read?

Yes.  The fill comes from the fact that the data is recorded redundantly. If a little bit of it goes missing, there is another copy some place else to fill it in with. The results are perfect recovery. If the loss is too large to cover this way, then typically it is replaced with a carefully executed mute, which if short and infrequent enough will slip by your ears unnoticed.

Quote
Or am I thinking of another technology? It all becomes a blur after a while...

Quote
And I don't know why you keep thinking I'm pro-DSD when I'm trying to insist that I'm not taking sides!

Not taking sides implies that DSD has equal merits with PCM which is false on practical grounds. Yes, you can get very  good sound either way, but the DSD route involves using a ton more data to do the same basic thing.

There's a reason why there was no such thing as workable SACD or DSD until there was a DVD disk base, with its very much larger data capacity and data rate features to waste time and money with.

Think of PCM as round earth, and DSD as flat earth. If there was a discussion of those issues, would you have a hard time taking sides?  If one is sufficiently informed about these things, that is how it would look.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: tedsmith on 2017-04-24 13:10:41
At risk of getting my head bitten off again, here's another question that might appeal more to your reasoning:

Is there any Windows software (preferably free, or a trial version) where I can load a PCM and a DSD track simultaneously and subtract one from the other to display any difference (or NO difference!) between the two?
You have to convert the DSD to PCM with a low pass filter first.  You will have the inevitable problem of finding the best alignment before subtracting.  Also if you download two files you don't know the provenance of the two files.

The short version is that the reconstruction filter for the PCM and the low pass output filter for the DSD drive most of the differences you'd see.  If the PCM is a high enough sample rate the differences will be negligible.

More details: if you take a high rate test tone PCM file (say >= 2.8224MHz), band limit it to less than 1/2 of the PCM output frequency for the PCM test tone, convert the original high rate PCM to DSD, run the DSD output filter on that to get back to high rate PCM, align the outputs with the input and subtract each output from the input (at the original high rate), then:  for PCM outputs that are lower sample rate than twice the DSD output filer cutoff you'll see more difference between them and the original test tone than between the DSD and the original test tone.  Conversely if your output sample rate is more than twice the DSD output filter cutoff you'll see a bigger difference between the DSD output and the test tone.  All of these differences are more affected by the implementation of the reconstruction filter for the PCM and the DSD output low pass filter than by whether they were ever DSD or high rate PCM.

There will also be some low level noise (below -120dBFS) in the DSD output over the audio band (say up to 30 or 40kHz) and then some rising ultrasonic noise up to the DSD output filter cutoff.  That noise will never exceed -40dBFS for SACDs and for higher rate DSD can be more negligible.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: shakeshuck on 2017-04-24 13:30:07
Think of PCM as round earth, and DSD as flat earth. If there was a discussion of those issues, would you have a hard time taking sides?  If one is sufficiently informed about these things, that is how it would look.

That's exactly what I'm trying to do: inform myself about the technology so I can work it out for myself!
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: Arnold B. Krueger on 2017-04-24 13:41:04
At risk of getting my head bitten off again, here's another question that might appeal more to your reasoning:

Is there any Windows software (preferably free, or a trial version) where I can load a PCM and a DSD track simultaneously and subtract one from the other to display any difference (or NO difference!) between the two?

I can save you some work. 

If one held everything else constant, that is made them the same except for the format,  the PCM and DSD audio would be identically the same. Subtract them and you get a vanishing result.

But, in just about  every commercial case everything was not held the same, so there will almost  always be some difference that actually has nothing to do with comparing DSD to PCM. For example, commercial recordings  of music that was previously released in PCM are typically remastered. So they are different, but not because one is DSD and the other is PCM.  They are not quite the same recording.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: tedsmith on 2017-04-24 13:52:10
That's exactly what I'm trying to do: inform myself about the technology so I can work it out for myself!
I'm not trying to be flippant here, but if you want to work it out for yourself from fundamentals you'll need to truly understand sampling theory, noise shaping and sigma delta modulation.  Most people who get thru an undergraduate degree in the sciences or math still apparently don't know enough (or can't apply the theory to practice enough) to understand the technical nuances.

As other's have said, the details of practical implementations, mastering, etc. swamp the technical differences.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: ajinfla on 2017-04-24 13:58:02
You guys are paranoid!
Or a lot more clever and experienced than "clever" trolls think
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: nastea on 2017-04-24 14:36:18
At risk of getting my head bitten off again, here's another question that might appeal more to your reasoning:

Is there any Windows software (preferably free, or a trial version) where I can load a PCM and a DSD track simultaneously and subtract one from the other to display any difference (or NO difference!) between the two?
teac hi-res editor (http://www.teac-audio.eu/en/products/teac-hi-res-editor-135099.html)
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: Arnold B. Krueger on 2017-04-24 14:42:46
Ah yes, the stealth woo-woo pusher masquerading as an victimized oh-so-innocent answer seeker. Definitely never happened around these parts.

You guys are paranoid!

The apparent paranoia comes from about 17 years experience with DSD advocates trying to pull the wool over our eyes.  That it was a scam to foist off highly restrictive IP protection and open up a new market for redundant, overpriced hardware was apparent to many of us immediately. 

The means by which the scam would be falsely granted improved sound quality by means of remastering was also obvious to many of us.

But, nevertheless better men than I (example: Meyers and Moran of AES Journal fame) have succumbed at least temporarily (for several years - almost a decade). 

The extent the scam was that as of about 2006, about half of all SACD (DSD) recordings were based on analog recordings whose dynamic range and bandwidth was actually inferior to plain vanilla PCM (e.g. CD), yet no golden eared reviewer that I know of has ever blown the whistle on this to this day. 

Techies did start blowing the whistle on this part of the scam  a few years later, but only a few noticed it. The whistle blowing was squirreled away in some dusty old AES Conference papers.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: greynol on 2017-04-24 15:14:01
The results are perfect recovery. If the loss is too large to cover this way, then typically it is replaced with a carefully executed mute, which if short and infrequent enough will slip by your ears unnoticed.
Almost, except that individual samples with errors that cannot be corrected are interpolated. This is what usually happens with light damage which does not cover the case when the laser can no longer track.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: Arnold B. Krueger on 2017-04-24 15:32:59
The results are perfect recovery. If the loss is too large to cover this way, then typically it is replaced with a carefully executed mute, which if short and infrequent enough will slip by your ears unnoticed.
Almost, except that individual samples with errors that cannot be corrected are interpolated. This is what usually happens with light damage which does not cover the case when the laser can no longer track.

Yup, that, too. Slipped my mind.  The  modern times real world circumstances by which this happens escapes me.

While audiophiles seem to have focused their attention on sound quality and more particularly DAC and output buffer quality,  the technical battle that absorbed the most resources and provided the most audible benefits for say, the first decade of the CD was tracking performance.  Not being able to play the disc at all or not being able to play it without audible breaks resulted in a lot of product returns, and retailers take that pretty seriously.  Sound quality can be a judgement call, but not playing is pretty cut and dried.

Surprisingly, good examples of the original CDP 101 tracked pretty well, and even handled CD-Rs well enough. However many early samples had defective chips related to the tracking servos, that failed pretty early in life. The shared mono DAC with 1/2 sample delay between the channels was audible if you had a center channel, and response roll-offs above 13 KHz due to the analog filters could be heard, given the right program material.

The first generation DACs generally had barely audible failings, and the better quality second generation DACs were essentially audibly perfect. That left things like cost and size as the areas of actual real progress. The perfection of delta-sigma technology which was pretty well done by the mid-90s took care of that.

An exception was high end CD players which were separated into two boxes, one with the transport and one with the DAC. This was a lot easier to sell to audiophiles than to justify technically and it did lead to audible problems with interfacing the two boxes.  There is a well known right way to do this, but a lot of the high end products whose circuits I reviewed cheaped-off and did it wrong.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: Arnold B. Krueger on 2017-04-24 15:37:15
At risk of getting my head bitten off again, here's another question that might appeal more to your reasoning:

Is there any Windows software (preferably free, or a trial version) where I can load a PCM and a DSD track simultaneously and subtract one from the other to display any difference (or NO difference!) between the two?
teac hi-res editor (http://www.teac-audio.eu/en/products/teac-hi-res-editor-135099.html)

I downloaded it and tested it and it ran without serious problems in Windows 10/64.  It does seem to have some subtle issues. I round tripped the RMAA6 24/96 test file through DSD 2.8 land and it came back changed a little. There were  some subtle 0.2 dB frequcny response roll offs, and visible transient distortion of the raw test signal.  Probably worth listening to.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: greynol on 2017-04-24 17:35:29
I've had my appendix out. Oh wait! I can't prove it to you, therefore it didn't happen. Maybe the surgeon didn't really take it out, and I'm the victim of a conspiracy theory that I'm attempting to pass on.
AJ might not agree, but I do believe you honestly thought (https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=8&cad=rja&uact=8&ved=0ahUKEwjzloT0w73TAhXnq1QKHRnCDYsQtwIITDAH&url=https%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DG-lN8vWm3m0&usg=AFQjCNGahI3hEJrZVphpTdIJ-cNOkJvNjQ&sig2=Fy2j6OryNZxHr45kmJkDGQ) you saw a UFO.  But since you can't provide objective proof then it isn't worthy of discussion (https://hydrogenaud.io/index.php/topic,3974.html#post_tos8).
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: ajinfla on 2017-04-24 17:54:22
Well, to be fair, I believe he (Rich?) honestly thought he heard this also:
I once (years ago) brought up the subject of CD burning, and that in an experiment I had done the newly burned CD sounded better than the stamped original. I didn't know why, it just did as far as I was concerned.
Much like my 6yr old nephew heard Santa last Christmas.
I just believe there may be alternate explanations for what was "heard". Honestly. Of course, one of those could be ABX'd per forum standards.
And now back to fishing...
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: greynol on 2017-04-24 18:02:56
The latest "Rich" sock was banned a few days ago.

This would be a new record from him for rising from the dead, but yeah, "both sides have valid points" is an effective dog whistle around here, ignoring the implied claim that a difference was heard, even if he wasn't sure it was "worth fighting over."

All by yourself, as relatively easy as it really is, most people resist doing proper listening tests. I just got called a bunch of dirty names for politely pointing that out on another thread.
;)
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: shakeshuck on 2017-04-24 18:20:27
I've had my appendix out. Oh wait! I can't prove it to you, therefore it didn't happen. Maybe the surgeon didn't really take it out, and I'm the victim of a conspiracy theory that I'm attempting to pass on.
AJ might not agree, but I do believe you honestly thought (https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=8&cad=rja&uact=8&ved=0ahUKEwjzloT0w73TAhXnq1QKHRnCDYsQtwIITDAH&url=https%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DG-lN8vWm3m0&usg=AFQjCNGahI3hEJrZVphpTdIJ-cNOkJvNjQ&sig2=Fy2j6OryNZxHr45kmJkDGQ) you saw a UFO.  But since you can't provide objective proof then it isn't worthy of discussion (https://hydrogenaud.io/index.php/topic,3974.html#post_tos8).

Fair cop.

It also says: "by being polite and encouraging him to join in with how we do things here."
and "the manner in which you coax these people into doing things the right way is very important"

Mmmmh.
Or maybe I was just being a little OCD sensitive?
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: shakeshuck on 2017-04-24 18:23:02
The latest "Rich" sock was banned a few days ago.

This would be a new record from him for rising from the dead, but yeah, "both sides have valid points" is an effective dog whistle around here, ignoring the implied claim that a difference was heard, even if he wasn't sure it was "worth fighting over."

All by yourself, as relatively easy as it really is, most people resist doing proper listening tests. I just got called a bunch of dirty names for politely pointing that out on another thread.
;)

And I told you I HAD tried ABX testing, the software just made it too easy for a valid comparison.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: greynol on 2017-04-24 18:25:28
Test samples to allow others to reproduce your findings?

https://hydrogenaud.io/index.php/topic,3974.html
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: andy o on 2017-04-24 18:45:57

It may not have been your intention, but proclaiming there are 2 sides and implying they're equivalent when the science is well established is a well known and tired cliche of science deniers. We see it regarding global warming, medicine, biology (evolution especially), and yes, even audio. It's the false equivalence fallacy.


Guys!
I've obviously got off on a bad start here.
It appears a glib remark has been allowed to run riot and get completely out of hand.

My apologies to Arnold K, who I now realise was trying to steer me in the direction required by the forum members, which I wrongly took as an attack at the time.
I am surprised at the vehemence on here though to anyone that doesn't hold the same 100% view as 'the gang' - how can that lead to interesting discussions? I haven't tried to force an opinion on anyone, in fact I deliberately tried not to.

I've had my appendix out. Oh wait! I can't prove it to you, therefore it didn't happen. Maybe the surgeon didn't really take it out, and I'm the victim of a conspiracy theory that I'm attempting to pass on.

No bad intended! Lighten up, guys!
You are not getting it. You think you're just being jokey, but what you're saying (again) is not different to what trolls and science deniers say for real.

Saying that you're taking no position or that you're somehow in the middle implies there are 2 positions that are equivalent in the first place. You may not be taking any of your 2 imagined positions, but you are taking a position. A better way to approach this if you are a newb to the subject is to just ask what does the science say up to now.

Getting out an appendix is not an extraordinary thing that goes against previous established science. If you said you were alien-abducted or were seeing ghosts, or that you hear DSD and PCM difference, then that's something that would require evidence from you.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: greynol on 2017-04-24 19:00:01
It also says: "by being polite and encouraging him to join in with how we do things here."
and "the manner in which you coax these people into doing things the right way is very important"
That's what David said.  Neither the terms nor their descriptions say either.  Not that we shouldn't operate in this manner.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: shakeshuck on 2017-04-24 19:18:26
Test samples to allow others to reproduce your findings?

https://hydrogenaud.io/index.php/topic,3974.html

You're taking me wrong again.
I said the software wouldn't adjust the volume of the two tracks correctly to make it a fair comparison. The difference in volumes made it too easy to tell which was which.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: greynol on 2017-04-24 19:23:41
What software?
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: Arnold B. Krueger on 2017-04-24 19:30:15
[And I told you I HAD tried ABX testing, the software just made it too easy for a valid comparison.


The way to solve this problem was to do as  much processing as you could outside of the listening test comparison.

AFAIK the volume leveling feature you used involved steady state gain changes, so it really was just a standard question of level matching.

The way I level match is to do the following in preparation for the listening test:

(1) Mark out comparable timed points on the music. This is very much facilitaed if time-synching is done first. Time synching is done by identifying a unique feature and adding or subtracting silence at the beginning of the files until the feature has the same timing in both files, within a millisecond or less.

(2) Measure the RMS or if no RMS calculation is available, the average value associated with the level of music between the two points in each piece of music.

(3) Adjust the level in one file or the other until the RMS values are the same +/- 0.05 dB or better.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: shakeshuck on 2017-04-24 19:37:50
Saying that you're taking no position or that you're somehow in the middle implies there are 2 positions that are equivalent in the first place. You may not be taking any of your 2 imagined positions, but you are taking a position. A better way to approach this if you are a newb to the subject is to just ask what does the science say up to now.

I obviously misread the tone of the forum when I first started posting.
I assumed there would be proponents of both PCM and DSD on here, happily comparing notes - so I made a comment aimed at not offending one way or the other.
Boy, was that a misguided thing to do. In trying NOT to offend, I seem to have achieved exactly the opposite.

As for joking, I'm trying to make light of a situation that is approaching the absurd.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: bennetng on 2017-04-24 19:39:47
https://hydrogenaud.io/index.php/topic,107354.msg937970.html#msg937970
Just look at shakeshuck's posting history...
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: shakeshuck on 2017-04-24 19:44:39
[And I told you I HAD tried ABX testing, the software just made it too easy for a valid comparison.


The way to solve this problem was to do as  much processing as you could outside of the listening test comparison.

AFAIK the volume leveling feature you used involved steady state gain changes, so it really was just a standard question of level matching.

The way I level match is to do the following in preparation for the listening test:

(1) Mark out comparable timed points on the music. This is very much facilitaed if time-synching is done first. Time synching is done by identifying a unique feature and adding or subtracting silence at the beginning of the files until the feature has the same timing in both files, within a millisecond or less.

(2) Measure the RMS or if no RMS calculation is available, the average value associated with the level of music between the two points in each piece of music.

(3) Adjust the level in one file or the other until the RMS values are the same +/- 0.05 dB or better.

I tried manually adjusting the ReplayGain tags on the files to get them somewhere close, but it turns out that Foobar/ABX Comparator ignores the tags when doing a comparison. You can use auto ReplayGain, but that was what was not working for me.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: greynol on 2017-04-24 19:52:54
(2) Measure the RMS or if no RMS calculation is available, the average value associated with the level of music between the two points in each piece of music.
He should probably lowpass to no greater than the nyquist of the PCM samplerate first.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: shakeshuck on 2017-04-24 21:19:16
(3) Adjust the level in one file or the other until the RMS values are the same +/- 0.05 dB or better.

If I'm still trying to use Foobar for the comparison, what software would you suggest I use to alter the file levels? Obviously it would have to be PCM based, as - if I read it correctly - I can't alter a DSD file easily.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: shakeshuck on 2017-04-24 21:23:50
At risk of getting my head bitten off again, here's another question that might appeal more to your reasoning:

Is there any Windows software (preferably free, or a trial version) where I can load a PCM and a DSD track simultaneously and subtract one from the other to display any difference (or NO difference!) between the two?
teac hi-res editor (http://www.teac-audio.eu/en/products/teac-hi-res-editor-135099.html)

I have used this for converting between file types for testing, but I am not aware that it will allow me to subtract samples from each other in order to easily see any differences.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: shakeshuck on 2017-04-24 21:33:24
Test samples to allow others to reproduce your findings?

https://hydrogenaud.io/index.php/topic,3974.html

I appreciate you weren't on the ball when you made this statement, but I'm curious as to what this would achieve?
Let's say as an example that I could do multiple ABX tests and do relatively well consistently.
If others try the same test with the same samples and fail, does that then invalidate my findings?
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: saratoga on 2017-04-24 22:04:17
Test samples to allow others to reproduce your findings?

https://hydrogenaud.io/index.php/topic,3974.html

I appreciate you weren't on the ball when you made this statement, but I'm curious as to what this would achieve?
Let's say as an example that I could do multiple ABX tests and do relatively well consistently.
If others try the same test with the same samples and fail, does that then invalidate my findings?

If no one ever could? Yes, probably people would assume you got lucky or made a mistake.

Usually though the purpose of this is to sanity check someone else's results for obvious errors and then to let other people try and see how they do. Replication and review are the basis of all science and engineering.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: saratoga on 2017-04-24 22:06:02
At risk of getting my head bitten off again, here's another question that might appeal more to your reasoning:

Is there any Windows software (preferably free, or a trial version) where I can load a PCM and a DSD track simultaneously and subtract one from the other to display any difference (or NO difference!) between the two?
teac hi-res editor (http://www.teac-audio.eu/en/products/teac-hi-res-editor-135099.html)

I have used this for converting between file types for testing, but I am not aware that it will allow me to subtract samples from each other in order to easily see any differences.

You can't subtract audio in two different sampling rates because there are not the same number of samples. You must first convert them to a common sampling rate, time align them, and then subtract. Note that this process is often very complex for formats that are substantially different.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: saratoga on 2017-04-24 22:09:53
Saying that you're taking no position or that you're somehow in the middle implies there are 2 positions that are equivalent in the first place. You may not be taking any of your 2 imagined positions, but you are taking a position. A better way to approach this if you are a newb to the subject is to just ask what does the science say up to now.

I obviously misread the tone of the forum when I first started posting.
I assumed there would be proponents of both PCM and DSD on here, happily comparing notes - so I made a comment aimed at not offending one way or the other.



DSD is a dead format, in part because it was a technically inferior option, in part because it was used as a vehicle for pushing DRM, and in part due to questionable marketing and product positioning.

Outside of hipsters, collectors and the ignorant, you probably won't find many proponents of the format.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: shakeshuck on 2017-04-24 22:49:14
Saying that you're taking no position or that you're somehow in the middle implies there are 2 positions that are equivalent in the first place. You may not be taking any of your 2 imagined positions, but you are taking a position. A better way to approach this if you are a newb to the subject is to just ask what does the science say up to now.

I obviously misread the tone of the forum when I first started posting.
I assumed there would be proponents of both PCM and DSD on here, happily comparing notes - so I made a comment aimed at not offending one way or the other.



DSD is a dead format, in part because it was a technically inferior option, in part because it was used as a vehicle for pushing DRM, and in part due to questionable marketing and product positioning.

Outside of hipsters, collectors and the ignorant, you probably won't find many proponents of the format.

If a particularly masterful performance of a favourite piece of yours came out only on DSD, would you be happy to miss out on that simply because of your anti-DSD stance?
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: Wombat on 2017-04-24 23:31:05
If a particularly masterful performance of a favourite piece of yours came out only on DSD, would you be happy to miss out on that simply because of your anti-DSD stance?
If that was the case i'd join an online forum to ask how to make it PCM.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: shakeshuck on 2017-04-24 23:49:08
If a particularly masterful performance of a favourite piece of yours came out only on DSD, would you be happy to miss out on that simply because of your anti-DSD stance?
If that was the case i'd join an online forum to ask how to make it PCM.

LOL
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: greynol on 2017-04-25 00:08:51
...or 256kbit Quick Time AAC.

Bet the OP won't be able to tell the difference either way. ;)
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: shakeshuck on 2017-04-25 00:31:22
...or 256kbit Quick Time AAC.

Bet the OP won't be able to tell the difference either way. ;)

Even I won't bet against you on that one.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: ajinfla on 2017-04-25 02:17:10
If a particularly masterful performance of a favourite piece of yours came out only on DSD, would you be happy to miss out on that simply because of your anti-DSD stance?
If that was the case i'd join an online forum to ask how to make it PCM.
Or make a better sounding copy of it on a black sacd
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: Arnold B. Krueger on 2017-04-25 02:29:45
At risk of getting my head bitten off again, here's another question that might appeal more to your reasoning:

Is there any Windows software (preferably free, or a trial version) where I can load a PCM and a DSD track simultaneously and subtract one from the other to display any difference (or NO difference!) between the two?
teac hi-res editor (http://www.teac-audio.eu/en/products/teac-hi-res-editor-135099.html)

I have used this for converting between file types for testing, but I am not aware that it will allow me to subtract samples from each other in order to easily see any differences.

The Teac editor seems pretty limited but it adds the functions needed to do production with DSD files, including file conversions. . In general many common editing functions have never been implemented in the DSD domain, so DSD files are often converted to PCM, edited, and then converted back to DSD. Most DSD recordings also have a PCM stage of processing of sorts as part of their workflow to avoid problems that are inherent in DSD-only ADCs.

Freebie PCM editors like Audacity Audacity Team Site (http://www.audacityteam.org/) are far more full-function, and have built-in functions for cutting, pasting, copying, changing levels and equalization. Audacity can exploit plug-ins of several different kinds including the very numerous VST plug-ins to obtain a wide variety of  additional editing functions.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: Arnold B. Krueger on 2017-04-25 02:31:57
If a particularly masterful performance of a favourite piece of yours came out only on DSD, would you be happy to miss out on that simply because of your anti-DSD stance?
If that was the case i'd join an online forum to ask how to make it PCM.

LOL

I have a *universal" optical player that plays SACD, DVD-A and regular CDs.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: saratoga on 2017-04-25 02:41:16
Saying that you're taking no position or that you're somehow in the middle implies there are 2 positions that are equivalent in the first place. You may not be taking any of your 2 imagined positions, but you are taking a position. A better way to approach this if you are a newb to the subject is to just ask what does the science say up to now.

I obviously misread the tone of the forum when I first started posting.
I assumed there would be proponents of both PCM and DSD on here, happily comparing notes - so I made a comment aimed at not offending one way or the other.



DSD is a dead format, in part because it was a technically inferior option, in part because it was used as a vehicle for pushing DRM, and in part due to questionable marketing and product positioning.

Outside of hipsters, collectors and the ignorant, you probably won't find many proponents of the format.

If a particularly masterful performance of a favourite piece of yours came out only on DSD, would you be happy to miss out on that simply because of your anti-DSD stance?

What anti-DSD stance? 
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: bennetng on 2017-04-25 07:05:54
In general many common editing functions have never been implemented in the DSD domain, so DSD files are often converted to PCM, edited, and then converted back to DSD. Most DSD recordings also have a PCM stage of processing of sorts as part of their workflow to avoid problems that are inherent in DSD-only ADCs.
Yeah, for example DXD
https://en.wikipedia.org/wiki/Digital_eXtreme_Definition

At the end the OP could just be comparing PCM with PCM.

I fiddled with Sony's DSD recorder in the past and used some tools to convert DSD to PCM.
https://hydrogenaud.io/index.php/topic,107529.0.html

Apart from the shaped noise in the DSF file, RMAA did not show meaningful differences in the recorded PCM and DSD file.

https://hydrogenaud.io/index.php/topic,107528.0.html

Maybe the tools I used to convert DSF to PCM are more reliable than TEAC's.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: judd on 2017-04-25 07:59:50
Quote
Is there any Windows software (preferably free, or a trial version) where I can load a PCM and a DSD track simultaneously and subtract one from the other to display any difference (or NO difference!) between the two?

FYI, my converter ReSampler (https://github.com/jniemann66/ReSampler) is a command-line tool which can convert DSD .dsf and .dff files to various PCM formats. (Or if you prefer a graphical interface, use ferocious (https://github.com/jniemann66/ferocious))

You could use an audio editor (eg Audacity or Cool Edit / Audition) to do the subtracting.
However, in terms of subtracting / nulling a DSD and PCM version of the same track, as pointed-out by others, they would need to be converted to a common sample rate.

Also, the group delay caused by the FIR filter used in the conversion may result in the two files not being exactly time-aligned, so care would need to be taken to ensure that they are time-aligned properly,
Additionally, SACDs are supposed to be mastered so that they peak at -6dB FS or 50% (although, in practice, many are a bit hotter than that), so care would also need to be taken to ensure the levels are matched.

Finally, if you were to compare a PCM release of a track, with it's DSD-to-PCM counterpart, then you would need to be absolutely sure that the two versions are from the same master, which is not at all easy to be certain about.

In my experience, many DSD titles (with a few notable exceptions, such as the files from the 2L Hires Test-Bench (http://www.2l.no/hires/index.html?), which do seem to have genuine supersonic audio content) don't appear to have meaningful musical content that the PCM version doesn't have, with the main difference being the presence of DSD dither in the DSD version (which is noise-shaped into the supersonic region, and easily recognizable in a spectrum analyzer).  
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: shakeshuck on 2017-04-25 09:58:41
Thanks all, I think I'm getting the message.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: castleofargh on 2017-04-25 12:14:29
If a particularly masterful performance of a favourite piece of yours came out only on DSD, would you be happy to miss out on that simply because of your anti-DSD stance?
that's capitalism, you get to vote all day long with your wallet. if I was against DSD but bought 3 SACDs every month, it wouldn't make much sense.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: Arnold B. Krueger on 2017-04-25 12:22:10
If a particularly masterful performance of a favourite piece of yours came out only on DSD, would you be happy to miss out on that simply because of your anti-DSD stance?
If that was the case i'd join an online forum to ask how to make it PCM.

LOL

The joke is on anybody who thinks that they can actually listen to DSD music without PCM being part of the signal flow.  For example, try to find any audio  product that implements bass management or room optimization in the digital domain without first converting it to PCM.

Even Sony admits that their studio recording process for DSD involves a stage of what for all practical purposes is PCM.

The latter fact is another one of those things that is hidden away in dusty academic papers, and not exactly headlined in the high end audio publications or enthusiast's web site.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: Arnold B. Krueger on 2017-04-25 12:56:22
(2) Measure the RMS or if no RMS calculation is available, the average value associated with the level of music between the two points in each piece of music.
He should probably lowpass to no greater than the nyquist of the PCM samplerate first.

Yes, that's just common sense, but I need to remember that which is common sense for old hands is not necessarily common sense for newbies.

BTW I analyzed a randomly selected file from  the stereo 24/192 files ("2L-086") from the "High Rez Test Bench" Hi Rez Test bench Files (http://www.2l.no/hires/index.html?)

The "Total RMS Power" of the file as received was -22.23 dB (L) and -21.66 (R). High pass filtered with a filter @ 20 KHz with a 1 KHz transition band, the residual > 20 KHz was -71.68 (L) and -65.56 (R).  Rough estimate is that the residual, being > 40 dB down would at worst cause an error to the level matching process on the order of < 1% or  < 0.1 dB and would therefore not be significantly  detrimental to the level matching process.

This is consistent with other similar tests I have performed over the past 15 or so years.

That these kind of results are common is just one more reason why removing all music > 20 KHz is generally inaudible. There is hardly any energy > 20 KHz in natural music. Compound that by the ear's lack of sensitivity > 20 KHz and frost the cake with masking and you've basically got Mission Impossible.

A possible exception would be a file where there was a continuous high amplitude tone, such as might be added by a switchmode power supply and/or  a poor equipment design or test setup.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: greynol on 2017-04-25 13:06:44
Quote
A possible exception would be a file where there was a continuous high amplitude tone
...or DSD.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: bennetng on 2017-04-25 13:14:09
FYI, my converter ReSampler (https://github.com/jniemann66/ReSampler) is a command-line tool which can convert DSD .dsf and .dff files to various PCM formats. (Or if you prefer a graphical interface, use ferocious (https://github.com/jniemann66/ferocious))

Maybe a bit off-topic but it seems that your software's UI is not DPI-aware. While hi-res audio does no harm to my ears a hi-res monitor does hurt my eyes. The recent Windows 10 Creators Update advocating improvement in UI scaling is a pure joke, I used it for no more than 3 hours and reverted to Windows 7.

files from the 2L Hires Test-Bench (http://www.2l.no/hires/index.html?)

Back to the topic. Just in case shakeshuck wants to ABX 2L's files.
"Some 2L Hi-Res music samples are either botched or cooked"
https://hydrogenaud.io/index.php/topic,111416.0.html
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: Arnold B. Krueger on 2017-04-25 13:25:21

Back to the topic. Just in case shakeshuck wants to ABX 2L's files.
"Some 2L Hi-Res music samples are either botched or cooked"
https://hydrogenaud.io/index.php/topic,111416.0.html

Amen brother. A good example of a botched high-profile test would be the AIX files that were circulated via AVS a few years back.

To be sure, the best practice is to examine the high rez file with a good audio editor for glitches, including looking at its spectral content for spurious tones ands noises,  and then do the downsampling to CD quality yourself using well known or at least well-tested tools, followed by  examination of your newly created file as above and additional checks for level matching and  time synchronization.

 When you downsample, beware that some downsamplers have incredibly narrow transition bands and might  cause exceptional  ringing that in turn might even create an audible tell in the form of clipping or IM.

In addition, it can be wise to make some downsampled files to lower sample rates and word lengths then your target test for listener training.  Trying to detect something that isn't already familiar sets the stage for frustration and failure.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: shakeshuck on 2017-04-25 17:43:55

Back to the topic. Just in case shakeshuck wants to ABX 2L's files.
"Some 2L Hi-Res music samples are either botched or cooked"
https://hydrogenaud.io/index.php/topic,111416.0.html

I wouldn't have expected that. Thanks for pointing it out.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: Arnold B. Krueger on 2017-04-25 17:48:04
Quote
A possible exception would be a file where there was a continuous high amplitude tone
...or DSD.

Close, but not quite. I did the same test on 24/96  silence processed by the Teac DSD editor, and found that the noise > 20 KHz had an average  power of -67 dB, which is also significantly more than 40 dB below  the average power of my reference music. So whether you filter it out or not, its impact is exceedingly unlikely to be audible (< 0.1 dB) .
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: greynol on 2017-04-25 18:19:49
Perhaps relevant enough to cause a noticeable difference in calculated playback gain when sent through a broken volume leveler?
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: Wombat on 2017-04-25 18:31:06
What about a simple start with a dsd->PCM->dsd 2x converted file created with something like SoX? Pimped SoX for dsd support by Mans Rullgard (https://github.com/mansr/sox)
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: drewfx on 2017-04-25 18:53:25
I think these DSD discussions often get confused because IMO there are two different questions and sometimes we end up with a moving target, deliberate or not:

1. CD quality vs. higher res, be it PCM or DSD.
2. DSD vs. higher res. PCM.

#1 is easier to test and if one can't ABX a difference, then #2 is pretty much academic.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: Wombat on 2017-04-25 19:08:26
Sure but this is the problem here. PCM behaves different because of the playbackchain makes it hard to do an abx. You can do dsd->44.1->dsd to have some easy start for your playbackchain in comparing the original dsd to. If there is nothing heard you are done with the dsd or PCM discussion.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: Case on 2017-04-25 20:04:24
Perhaps relevant enough to cause a noticeable difference in calculated playback gain when sent through a broken volume leveler?
I don't know how Arnold made his test but I took Pink Floyd - Wish You Were Here DSD file and converted it to 352800 Hz PCM. This file showed RMS level at -18.39 dB. Filtering away frequencies below 22 kHz showed RMS level for the ultrasonics at -22.28 dB. And after resampling to 44.1 kHz to remove the inaudible sounds RMS level was at -20.68 dB.
At least with this this file and with this much ultrasonic content RMS is not a valid method for loudness matching. Computing ReplayGain for a 44.1 kHz or 48 kHz resampled version gives better results.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: StephenPG on 2017-04-25 21:18:03
I've never understood the point of using old analogue recordings which are never going to use more than 13 bits or need a sampling rate greater than 48kHz.

Wasn't this used to try and discredit the Meyer-Moran SACD test?
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: Arnold B. Krueger on 2017-04-25 21:36:19
Perhaps relevant enough to cause a noticeable difference in calculated playback gain when sent through a broken volume leveler?
I don't know how Arnold made his test but I took Pink Floyd - Wish You Were Here DSD file and converted it to 352800 Hz PCM. This file showed RMS level at -18.39 dB. Filtering away frequencies below 22 kHz showed RMS level for the ultrasonics at -22.28 dB. And after resampling to 44.1 kHz to remove the inaudible sounds RMS level was at -20.68 dB.
At least with this this file and with this much ultrasonic content RMS is not a valid method for loudness matching. Computing ReplayGain for a 44.1 kHz or 48 kHz resampled version gives better results.
Where might I find that DSD file to try my procedure on?
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: Arnold B. Krueger on 2017-04-25 21:51:42
I've never understood the point of using old analogue recordings which are never going to use more than 13 bits or need a sampling rate greater than 48kHz.

Wasn't this used to try and discredit the Meyer-Moran SACD test?

More than try... It was a relevant critical influence on the outcome of the tests.

Meyer and Moran were blindsided as was much as the rest of the audio world. Greisinger had pointed this situation out in a paper he gave at an AES conference in Baniff in 2003, but it just seems to have slipped everybody by. The  PPT of the slides are here:

David Griesinger intermod.ppt from Baniff, 2003 (http://www.davidgriesinger.com/intermod.ppt)
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: greynol on 2017-04-26 00:05:41
@Arnod:

In case you didn't notice these two posts of mine are related:

g=938570 date=1493059974]
(2) Measure the RMS or if no RMS calculation is available, the average value associated with the level of music between the two points in each piece of music.
He should probably lowpass to no greater than the nyquist of the PCM samplerate first.

They were an attempt to address the very real problem faced by the OP in his botched attempt to perform a controlled comparison:
This is not necessarily as easy as you make out. My first post on HAudio was in the ABX section for Foobar, which was making comparisons impossible as ReplayGain wouldn't set the levels between PCM and DSD correctly (possibly due to the noise, but I won't speculate as I get told off for doing that).

So I HAVE tried a proper listening test, and the technology failed me.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: judd on 2017-04-26 01:25:09
FYI, my converter ReSampler (https://github.com/jniemann66/ReSampler) is a command-line tool which can convert DSD .dsf and .dff files to various PCM formats. (Or if you prefer a graphical interface, use ferocious (https://github.com/jniemann66/ferocious))



Maybe a bit off-topic but it seems that your software's UI is not DPI-aware. While hi-res audio does no harm to my ears a hi-res monitor does hurt my eyes. The recent Windows 10 Creators Update advocating improvement in UI scaling is a pure joke, I used it for no more than 3 hours and reverted to Windows 7.

Yes, thanks for this. I addressed it in my other thread (https://hydrogenaud.io/index.php/topic,111777.new.html#new) (so as not to derail this one)


files from the 2L Hires Test-Bench (http://www.2l.no/hires/index.html?)

Back to the topic. Just in case shakeshuck wants to ABX 2L's files.
"Some 2L Hi-Res music samples are either botched or cooked"
https://hydrogenaud.io/index.php/topic,111416.0.html

Ok, Thanks. I wasn't aware of that, but I haven't checked out all of the files. The ones I looked at seemed ok.
I'm away from my audio setup atm, but will review when i get back.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: Case on 2017-04-26 06:40:00
Where might I find that DSD file to try my procedure on?
I happened to find it with SACD and DSD Google searches when I was looking for DSD test files to play with. Its legality might be an issue.

I used poor wording in my original post, I did see how you performed your test but I used a different method that matched OP's ABX trial.

Converting the previous Pink Floyd DSD file to 96 kHz PCM leaves majority of the noise shaping noise out. RMS power for this version of the file is -22.68 dB. High passed to 22 kHz shows -61.99 dB for the ultrasonics. 22 kHz lowpass gives the same RMS power as the full file, -22.68 dB. And for Greynol's information, ITU 1770 loudness scanner gives the same loudness value for the full 96 kHz track and the lowpassed one.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: greynol on 2017-04-26 07:11:47
Is in error when the PCM samplerate is 88.2k?

What happens with the old and new algorithms when you take the samplerate up so that a good portion of the noise shaping is still present?

Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: Case on 2017-04-26 08:07:29
Is in error when the PCM samplerate is 88.2k?
What do you mean?

What happens with the old and new algorithms when you take the samplerate up so that a good portion of the noise shaping is still present?
I ran the scan with the original 2Bdecided algorithm and the new ITU one on a few different sample rates. The results are rather strange:

Sample rateITUOriginal
352.8 kHz-3.99 dB-1.98 dB
192 kHz-0.96 dB-3.74 dB
48 kHz-1.47 dB-1.98 dB
44.1 kHz-1.47 dB-2.07 dB
Especially curious how the 192 kHz version gets treated so differently.

Edit: wrong sample rate had slipped in the table.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: tedsmith on 2017-04-26 11:42:22
FWIW Here's a little more info:
From the SACD and foobar2000 v1.3.15 - Converted .iso to 32 bit floating point PCM @ 2.8224MHz and used Adobe Audition 3.0 to down sample from that to the below rates (with prefilter quality 999) and then raise the level by 6dB (for SACD's 50% modulation level)

44.1k   -1.47dB
48k      -1.47dB
88.2k   -1.46dB
96k      -1.45dB
176.4k -0.79dB
192k    -0.29dB
352.8k -5.12dB
Raising the 352.8k by 6dB cause a few samples to clip but doing replay gain on the original 50% version got exactly +0.90dB so that didn't cause any problems.

The weirdness at 352.8k is probably explained by the HF noise that a normal SACD player would filter out (with say a 4th or 5th order filter at 50k to 80k)

Here is the FFT of 2:20 to 2:30 from 88.2k, 176.4k and 352.8k:
(http://oi64.tinypic.com/2ah5paq.jpg)
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: judd on 2017-04-26 12:45:50
Back to the topic. Just in case shakeshuck wants to ABX 2L's files.
"Some 2L Hi-Res music samples are either botched or cooked"
https://hydrogenaud.io/index.php/topic,111416.0.html

Just getting back to the topic of "Some 2L Hi-Res music samples are either botched or cooked"  - from the few examples I can see in that thread, it seems to be a case of the exact phenomenon I was referring to earlier in this thread:

Quote
..., the group delay caused by the FIR filter used in the conversion may result in the two files not being exactly time-aligned, so care would need to be taken to ensure that they are time-aligned properly,

Using a linear-phase FIR for the Lowpass Filter in the conversion process (which is what is usually used) will inevitably cause a delay, and the delay time will vary depending on the length of the FIR filter used.
The length of the FIR (and therefore the delay time) will vary from one converter to another, and even within the same converter, when the LPF parameters are varied (eg cutoff frequency / transition band / steepness etc).

Therefore, unless I'm mistaken (and always happy to be wrong),  I don't consider slight timing variations in various conversions of the same original material to be evidence of "botching" or "cooking". It is just an inevitable consequence of having a linear-phase FIR filter for anti-aliasing.

Of course, you can avoid the delay by using a minimum-phase FIR, but then you will be introducing phase distortion, and since you will have changed the phase of individual frequency components, your subtraction trick will not work properly.
   


Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: Case on 2017-04-26 12:53:15
That's some extensive testing, tedsmith.
I'm happy to report that Peter seems to finally be adding ultrasound filtering to the foobar2000's ReplayGain scanner. Should improve ABX situation and perhaps even help any real RG user.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: tedsmith on 2017-04-26 13:12:46
I'm glad it came up.  I had wondered why all of the sudden my replay gains for DSD started looking different some releases of foobar2000 ago.  I didn't care enough to run it down then (and running the new scanner on all of the SACDs I'd ripped took days and days.)  Also I just wanted to get a known good DSD copy of Pink Floyd's "Wish You Were Here" since you'd mentioned that the provenance wasn't known.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: Arnold B. Krueger on 2017-04-26 13:58:49
Back to the topic. Just in case shakeshuck wants to ABX 2L's files.
"Some 2L Hi-Res music samples are either botched or cooked"
https://hydrogenaud.io/index.php/topic,111416.0.html

Just getting back to the topic of "Some 2L Hi-Res music samples are either botched or cooked"  - from the few examples I can see in that thread, it seems to be a case of the exact phenomenon I was referring to earlier in this thread:

Quote
..., the group delay caused by the FIR filter used in the conversion may result in the two files not being exactly time-aligned, so care would need to be taken to ensure that they are time-aligned properly,

Using a linear-phase FIR for the Lowpass Filter in the conversion process (which is what is usually used) will inevitably cause a delay, and the delay time will vary depending on the length of the FIR filter used.
The length of the FIR (and therefore the delay time) will vary from one converter to another, and even within the same converter, when the LPF parameters are varied (eg cutoff frequency / transition band / steepness etc).

Therefore, unless I'm mistaken (and always happy to be wrong),  I don't consider slight timing variations in various conversions of the same original material to be evidence of "botching" or "cooking". It is just an inevitable consequence of having a linear-phase FIR filter for anti-aliasing.

Of course, you can avoid the delay by using a minimum-phase FIR, but then you will be introducing phase distortion, and since you will have changed the phase of individual frequency components, your subtraction trick will not work properly.

Another common source of time-shifting  involves the use of a hardware resampler. This was the probable cause of the problem with the AIX samples that I have been discussing.  I'm sure you know how this happens - the resampler has its own free-running output sampling clock, and the rest is history. Resampling on the computer is different because the clock is inherent in the process, not the operational environment.

Generally a millisecond or less of time delay that affects all audible channels equally is not a problem. But more than that and an audible echo may be heard at switching points. The tell may not be perceived as a delay, the possibility of being perceived as a difference is greater. 

When it comes to interchannel time delays on the order of 10s of microseconds can be audible, especially when channels are electrically mixed. This showed up in the old days when good DACs cost serious money and were often time-shared among the channels.

Common  things that one hears uttered by people who actually do reliable sensitive properly blinded tests is that the smallest differences are only heard as some uncharacterizable difference that is  nevertheless reliably detectable. As long as people talk about some specific kind of change,  the odds are high that either the difference is relatively large or that  they are imagining, not actually hearing. Of course imaging as opposed to hearing becomes obvious in the statistical analysis.   Once one starts hearing a difference, the reliability of detection usually starts increasing, and high levels of reliability such as 99% confidence generally become doable. 

Listening fatigue usually comes from not really hearing anything, which is why placebophiles complain about it so  much.

Listener training where the technical difference being investigated is gradually reduced from truly obvious to well below levels where we expect to hear something based on standard psychoacoustic guides is probably the best form of listener training.

On occasion I've seen ABX results that are almost an order of magnitude better than what the psychoacoustics text lead me to believe is possible.  I attribute that to the fact that so many psychoacoustics texts are based on the other kind of ABX tests, the ones that date back to the 50's. Not that they are invalid, but that they  seem to be designed to minimize listener training. Minimizing the effects of listener training is reasonable too, when audibility is related to the real world where we mostly only get to hear any particular thing once, and in a complex environment. Listening to recordings is different, because we can and often do listen repetitively and in a very focused and isolated way.

The most common form of cooked samples from 2L is that a lot of their higher sample rate files show evidence of being slavishly resampled from lower sample rate files, and not actually having any signficiant  performance source-related content above say, 70 KHz.  The fraud that hobbled Meyer and Moran for example is still being practiced it seems, just moved up a few octaves.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: Arnold B. Krueger on 2017-04-26 14:16:57
That's some extensive testing, tedsmith.
I'm happy to report that Peter seems to finally be adding ultrasound filtering to the foobar2000's ReplayGain scanner. Should improve ABX situation and perhaps even help any real RG user.

If the files being level-matched are sufficiently similar, then the need for filtering goes away. Differences that total out to be 40 or more dB down would seem to represent a conservative limit.

It seems to me that the level-matched files are very different, even different pieces of music then the level matching would seem to need to be much more  sophisticated.  By that I mean that the side chain processing would need to have spectral shaping that is based on audibility along the lines of Fletcher Munson, perhaps even including the different shapes for different levels.

BTW in my DSD tests the format used for all PCM processing was 24/192.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: Arnold B. Krueger on 2017-04-26 15:02:26
I think these DSD discussions often get confused because IMO there are two different questions and sometimes we end up with a moving target, deliberate or not:

1. CD quality vs. higher res, be it PCM or DSD.
2. DSD vs. higher res. PCM.

#1 is easier to test and if one can't ABX a difference, then #2 is pretty much academic.

I think both have been tested to death, and the outcome can be obtained by searching archives.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: greynol on 2017-04-26 15:17:09
Bugs aside (in the ABX 2.0 thread, 44.1 vs 48 is mentioned, hence my inquiry about 88.2 vs 96), I suspect the true issue is really quite simple: David's equal loudness filter adequately removes the noise shaping, whereas the hastily adopted (yes I did in fact just say that once again) shiny new little darling pet 1770 doesn't. However, I'm happy Peter can fix this with an ultrasonic filter bandaid. I'll reserve my future grumblings over 1770 to where it matters to me: real issues with real music, rather than esoteric, as a delivery format DSD can be different than PCM for commercially available content, nonsense.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: bennetng on 2017-04-26 15:38:41
Using a linear-phase FIR for the Lowpass Filter in the conversion process (which is what is usually used) will inevitably cause a delay, and the delay time will vary depending on the length of the FIR filter used.
The length of the FIR (and therefore the delay time) will vary from one converter to another, and even within the same converter, when the LPF parameters are varied (eg cutoff frequency / transition band / steepness etc).

Therefore, unless I'm mistaken (and always happy to be wrong),  I don't consider slight timing variations in various conversions of the same original material to be evidence of "botching" or "cooking". It is just an inevitable consequence of having a linear-phase FIR filter for anti-aliasing.

Of course, you can avoid the delay by using a minimum-phase FIR, but then you will be introducing phase distortion, and since you will have changed the phase of individual frequency components, your subtraction trick will not work properly.

I am not familiar with the relationship between filtering and delay, but from what I interpret from your reply, does it mean using a linear phase filter, in general, will introduce such a large amount of time delay?

Since the ABX log from mzil contains SHA checksums and 2L is still providing those files, I downloaded the 2496 file and use SoX to to convert them to 44k. I used the default settings as shown in the screenshot.

I inspected the converted file. As you can see in the screenshots, timing differences are basically nonexistent. (self-convert.png vs 96k.png)

Now, look at 2L's 44k file, the timing differences are so huge that I need to use a different zoom scale to show the differences. (2L 44k zoomout.png vs 2L 96k zoomout.png)

Since you also write your own converter so it should be fair to use your converter as an example. Indeed, your converter does produced some offsets, but the differences is much much smaller than 2L's and I would say negligible in ABX tests. (ferocious 44k zoomout.png)

I made SoundFonts in the past and used resamplers frequently, I did know some sample loops gone bad after resampled and I needed to re-align the loop point, but I never see a resampler can produce such a large offset.

Don't know if mzil is still visiting this forum or not...
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: saratoga on 2017-04-26 16:12:21
The delay depends on the length of the filter.  For a steep linear phase filter, the length will be very long, and therefore so will the delay.   For a filter with gradual roll off, the delay can be very short, but the rejection will not be good.  If you don't need good rejection, you can therefore have a relatively low delay (or else you must be willing to give up being linear phase).

This is why I said before that if you want to compare DSD and PCM you have to look carefully at time alignment.  Filters that are likely to be transparent (good image rejection, linear phase) are likely to have long delays 

Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: judd on 2017-04-26 16:16:01
Using a linear-phase FIR for the Lowpass Filter in the conversion process (which is what is usually used) will inevitably cause a delay, and the delay time will vary depending on the length of the FIR filter used.
The length of the FIR (and therefore the delay time) will vary from one converter to another, and even within the same converter, when the LPF parameters are varied (eg cutoff frequency / transition band / steepness etc).

Therefore, unless I'm mistaken (and always happy to be wrong),  I don't consider slight timing variations in various conversions of the same original material to be evidence of "botching" or "cooking". It is just an inevitable consequence of having a linear-phase FIR filter for anti-aliasing.

Of course, you can avoid the delay by using a minimum-phase FIR, but then you will be introducing phase distortion, and since you will have changed the phase of individual frequency components, your subtraction trick will not work properly.

I am not familiar with the relationship between filtering and delay, but from what I interpret from your reply, does it mean using a linear phase filter, in general, will introduce such a large amount of time delay?

Since the ABX log from mzil contains SHA checksums and 2L is still providing those files, I downloaded the 2496 file and use SoX to to convert them to 44k. I used the default settings as shown in the screenshot.

I inspected the converted file. As you can see in the screenshots, timing differences are basically nonexistent. (self-convert.png vs 96k.png)

Now, look at 2L's 44k file, the timing differences are so huge that I need to use a different zoom scale to show the differences. (2L 44k zoomout.png vs 2L 96k zoomout.png)

Since you also write your own converter so it should be fair to use your converter as an example. Indeed, your converter does produced some offsets, but the differences is much much smaller than 2L's and I would say negligible in ABX tests. (ferocious 44k zoomout.png)

I made SoundFonts in the past and used resamplers frequently, I did know some sample loops gone bad after resampled and I needed to re-align the loop point, but I never see a resampler can produce such a large offset.

Don't know if mzil is still visiting this forum or not...


Hey - nice work ! Love it ...

Yes, fair enough - the timing offsets should be relatively small.

I just did an experiment using my converter.
I put a single impulse at 1.0s in a 96kHz file, and downsampled it to 44kHz twice, using two different filter settings
a) Std (shorter FIR)
b) Steep (longer FIR)
... and it seemed to make a 1ms difference !

So, yeah - not sure why the timing difference in 2L's is so large.
Maybe it's more to do with their workflow - they are just busy and in a hurry, so they just select a block of audio in their DAW and trim it - not taking care to make it consistent.

So, if this is the case, I would concede that they probably did botch it  :-[ 
... particularly when you consider that the timing discrepancies are enough to produce tells in ABX tests. I am in agreement there.

Damn nice recordings, though ...
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: judd on 2017-04-26 16:17:44
The delay depends on the length of the filter.  For a steep linear phase filter, the length will be very long, and therefore so will the delay.   For a filter with gradual roll off, the delay can be very short, but the rejection will not be good.  If you don't need good rejection, you can therefore have a relatively low delay (or else you must be willing to give up being linear phase).

This is why I said before that if you want to compare DSD and PCM you have to look carefully at time alignment.  Filters that are likely to be transparent (good image rejection, linear phase) are likely to have long delays 



Amen
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: bennetng on 2017-04-26 16:18:40
The point I don't understand is the SoX plugin can be easily downloaded and checked and by default (see the screenshot setting) it can already give a super beautiful sweep, why there is no delay?
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: Wombat on 2017-04-26 16:30:22
It was taken care of in SoX to avoid that delay.
For dsd to PCM there is also the problem for the beginning of files often having a click because the filter has no real data to handle the amplitude of the very first sample when it only has 1 bit of info. (this is my theory based on limited understanding)
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: bennetng on 2017-04-26 16:40:08
Before I am getting more replies, let me make a guess...

Since file conversion is a non-realtime process, so adjustment can always be made afterwards, and SoX made such a correction?
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: Case on 2017-04-26 16:41:21
David's equal loudness filter adequately removes the noise shaping, whereas the hastily adopted (yes I did in fact just say that once again) shiny new little darling pet 1770 doesn't.
If you look at my short test you see that David's version has values varying all over the place too depending on the sample rate. The 192 kHz file gives entirely different results than the others. 1770-thingie has very stable results until 96 kHz is passed, as shown by tedsmith's extended testing.
It's weird how you are in such a disagreement with the new scanner. I have files from all genres except rap and jazz and I'm pleased with the volume leveling.

Edit: forgot to mention that you can still use the old ReplayGain scanner before libebur128 implementation by manually replacing the foo_rgscan.dll file. You can find the last version in foobar2000 v1.1.5 (http://filehippo.com/download_foobar2000/9356/)

The point I don't understand is the SoX plugin can be easily downloaded and checked and by default (see the screenshot setting) it can already give a super beautiful sweep, why there is no delay?
Any resampler could compensate for the delay. It's just a matter of dropping the extra silent samples.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: Wombat on 2017-04-26 16:43:47
Before I am getting more replies, let me make a guess...

Since file conversion is a non-realtime process, so adjustment can always be made afterwards, and SoX made such a correction?
Seems so. izotope lately fixed the sample allignment also.
I wonder after all the posts about SoX at HA people still fiddle around with other half-baked resampler options!?
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: Wombat on 2017-04-26 16:46:37
The point I don't understand is the SoX plugin can be easily downloaded and checked and by default (see the screenshot setting) it can already give a super beautiful sweep, why there is no delay?
Any resampler could compensate for the delay. It's just a matter of dropping the extra silent samples.
For plugins i guess the offset relates in big part to the host. Not sure the SoX plugin can work as proper as SoX standalone.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: judd on 2017-04-26 16:57:04
Quote
Any resampler could compensate for the delay. It's just a matter of dropping the extra silent samples.

They are not necessarily silent, though - if the source doesn't start with silence, they are "pre-ringing" oscillations (albeit with small amplitude)
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: bennetng on 2017-04-26 17:02:09
I wonder after all the posts about SoX at HA people still fiddle around with other half-baked resampler options!?
Because SoX is free and popular, it must be mediocre.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: greynol on 2017-04-26 17:05:05
If you look at my short test you see that David's version has values varying all over the place too depending on the sample rate.
Nonsense:
Sample rateITUOriginal
352.8 kHz-3.99 dB-1.98 dB
192 kHz-0.96 dB-3.74 dB
48 kHz-1.47 dB-1.98 dB
44.1 kHz-1.47 dB-2.07 dB
The 192 kHz file gives entirely different results than the others.
True, but as you were so quick to point the finger at an implementation bug (44.1 vs 48), let's not rule this out as a possibility for 192.

It's weird how you are in such a disagreement with the new scanner. I have files from all genres except rap and jazz and I'm pleased with the volume leveling.
It's not weird at all.  I'm not at all pleased with the leveling for metal.  I suspect it is because metal has more prevalent HF content causing 1770 to over-compensate because it doesn't implement an effective equal loudness filter.  If you're trying to model the human perception of loudness I would think one should use a proper model of human frequency response and I suspect 1770 is lacking in this regard, if not inferior to the original algorithm at the very least.  That Peter needs to add an ultrasonic bandaid in order to get the new algorithm to spit out accurate results only stands to confirm my suspicions.

Edit: forgot to mention that you can still use the old ReplayGain scanner before libebur128 implementation by manually replacing the foo_rgscan.dll file. You can find the last version in foobar2000 v1.1.5 (http://filehippo.com/download_foobar2000/9356/)
Thanks! YAY!!!!
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: greynol on 2017-04-26 17:06:28
They are not necessarily silent, though - if the source doesn't start with silence, they are "pre-ringing" oscillations (albeit with small amplitude)
Pffft.

What's the frequency of this pre-ringing again?!?

Please don't interpret this as my being critical of you, judd, rather I see pre-ringing as the new jitter; striking fear in the hearts of placebophiles.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: judd on 2017-04-26 17:13:11
Pffft.
What's the frequency of this pre-ringing again?!?

why is that relevant ?
My point was about truncating non-zero samples.
It's probably not a big concern, but then neither is a sub-1ms delay.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: greynol on 2017-04-26 17:15:33
You posted while I was in the midst of an edit, as I suspected you would take it the wrong way and get defensive.
Please don't interpret this as my being critical of you, judd, rather I see pre-ringing as the new jitter; striking fear in the hearts of placebophiles.
My apologies for not having the foresight to initially include it prior to posting.

(EDIT: I decided to split the infinitive (http://www.quickanddirtytips.com/education/grammar/split-infinitives) rather than put initially at the end.)
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: judd on 2017-04-26 17:21:36
You posted while I was in the midst of an edit, as I suspected you would take it the wrong way and get defensive.
Please don't interpret this as my being critical of you, judd, rather I see pre-ringing as the new jitter; striking fear in the hearts of placebophiles.

Ah, it's all good :-) 

I was looking at it more from a mathematical PoV.

Just so you know, I'm not a "true believer" about ringing.
Personally, I've tried to ABX various filter "steepnesses" and I can't hear it.
But that's just me ...



 
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: greynol on 2017-04-26 17:28:10
Just so you know, I'm not a "true believer" about ringing.
No worries, I could tell.
But that's just me
...and just about everyone else who knows something about the nuts and bolts of audio including an understanding of the perception of sound which is something many (most?) computer scientists and engineers (both degreed and ceremoniously titled audio engineers, alike) woefully lack.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: judd on 2017-04-26 18:41:09
I thought this was a nice example from the 2L site of Cello harmonics popping out of the mix, extending up into the 30-40kHz range.
(It's a bit hard to see in a static screenshot; easier to spot on live display)

This is a nice change from the familiar 20kHz roll-off you see all-too-often in many other hi-re$ files.
image link (https://ibb.co/kVv2T5)
(https://preview.ibb.co/n2Wbo5/Actual_Harmonics.jpg)
from the 96kHz version of "North Country II" at around 3:54.3 (unfortunately, DSD version is not available for comparison)

Of course, I don't claim to be able to hear any of the ultrasonic stuff, but it is interesting to see some actual (music-related) ultrasonic material in there - even though the high-frequency components are at pretty low amplitudes.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: greynol on 2017-04-26 18:55:35
Just to play devil's advocate, are we sure the harmonics actually emanated from the cello or could they have been the result of non-linear distortion somewhere in the signal chain?
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: judd on 2017-04-26 19:02:33
Just to play devil's advocate, are we sure the harmonics actually emanated from the cello or could they have been the result of non-linear distortion somewhere in the signal chain?

yeah, it could be distortion. Who knows ? 
I think they are using some pretty nice mics, though, so i suspect it could be the real deal.
Some mics, (like an Earthworks QTC-50) can capture those frequencies.

It definitely correlates nicely with that particular Cello note, if you watch/listen at the same time.
I was looking for a solo example of one of the brighter-sounding instruments to isolate it, but so far I haven't come across a solo Cello passage yet.
 
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: bennetng on 2017-04-26 19:07:22
Real or not, I can't hear that. But of course I can set the playback speed to 48k and listen to them in an octave lower and in half speed. ;)
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: judd on 2017-04-26 19:12:47
Real or not, I can't hear that. But of course I can set the playback speed to 48k and listen to them in an octave lower and in half speed. ;)
Haha - I've done that too.
Sometimes I completely filter out the audible spectrum first, and slow down what's left :-)
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: greynol on 2017-04-26 19:14:36
Real or not, I can't hear that. But of course I can set the playback speed to 48k and listen to them in an octave lower and in half speed. ;)
So the stuff at 40k would then be at 20k but still well below your threshold of hearing at those frequencies, assuming you still have the hair cells to detect them. ;)
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: judd on 2017-04-26 19:17:34
 Quarter-speed  ? :D
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: greynol on 2017-04-26 19:18:33
Heavily amplified ? :D
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: Wombat on 2017-04-26 19:19:01
Only thing missing in this thread now is why to wurry about dsd and alike when the best ears in this world know only de-blurred MQA sounds right...
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: judd on 2017-04-26 19:30:36
Only thing missing in this thread now is why to wurry about dsd and alike when the best ears in this world know only de-blurred MQA sounds right...

Well after all. everyone knows that MQA "delivers fuller, more tonally satisfying bass notes and a better sense of the space surrounding [Ella] Fitzgerald’s voice" :D
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: bennetng on 2017-04-26 19:41:13
...and a fuller spectrogram
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: greynol on 2017-04-26 20:48:25
...but it's lossy and lossy is evil.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: .halverhahn on 2017-04-26 20:57:05
Lossy is fine as long as it's MQA certified ;)
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: Arnold B. Krueger on 2017-05-03 13:28:12

Any resampler could compensate for the delay. It's just a matter of dropping the extra silent samples.

In my book, careful examination of the files being ABX'd is a natural part of the work flow for preparing the experiment.

The basic influences that most often need to be controlled are:

(1) Level   goal: +/- 0.1 dB

(2) Frequency response: +/- 0.1 dB 20-20K

(3) Time offsets +/- 100 uSecc

Level and time offsets are easy enough to check and correct with any of a large number of audio editors.

Measuring and correcting frequency responsemay be non trivial. One shortcut is to run the Audio Rightmark test file and then evaluate it.  IOW. append the Rightmark test file to the beginning or end of the test file, run it through your test process, and then edit it out and test it with the Rightmark program.

These are what I would call "Best practices" numbers - goals that ensure very unambiguous results.

You may  need to compromise them, but then the compromise should be tested all by itself to ensure that it is not an audible influence in context.

Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: shakeshuck on 2017-05-04 13:17:49
Jumping back to my pressed/burned CD "anecdote", now that I know a (little) bit more about PCM, I am willing to accept that a digital data misread could not alter a soundwave (yay!).
Any difference could also have been in my imagination (shock horror!).

Out of curiosity, though, for those folk who know about such things, is there any way (perhaps a badly designed power supply?) that a laser head fighting to read a mistracked disc, or power fluctuations due to spin speed alteration for the same reason may have any effect on an outgoing analogue signal? (more so with 30-year old CD players?).

Yes, I'm clutching at straws in the hope that my youth wasn't quite as misguided as it looks like it might have been...  :-[
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: Case on 2017-05-04 14:02:06
But it can alter the soundwave. If a sector failed to read and the missing content was interpolated it will almost certainly be different from the original. Depending on how often this happens and what was replaced it can be audible.

But there's no need to speculate. You could extract the data and compare if they are different. Or if the difference was only audible on some specific player record that player's output. This would not only reveal if the difference can be audible but also show what has changed.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: Wombat on 2017-05-04 16:13:48
Jumping back to my pressed/burned CD "anecdote"...
You can do dsd->44.1->dsd to have some easy start for your playbackchain in comparing the original dsd to. If there is nothing heard you are done with the dsd or PCM discussion.
Are you still interested in the main question of the thread? Did you try this?
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: shakeshuck on 2017-05-04 16:24:25
Jumping back to my pressed/burned CD "anecdote"...
You can do dsd->44.1->dsd to have some easy start for your playbackchain in comparing the original dsd to. If there is nothing heard you are done with the dsd or PCM discussion.
Are you still interested in the main question of the thread? Did you try this?

I've been fighting with a snap/crackle/pop problem with general file playback on Win 7 and USB, and have not yet quite bottomed the issue; I've gone as far as converting the Windows image from a raid setup to a single disk (which has helped, but I'm not there yet).
It's a bit distracting when trying to listen critically!
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: Arnold B. Krueger on 2017-05-04 21:07:14
Jumping back to my pressed/burned CD "anecdote", now that I know a (little) bit more about PCM, I am willing to accept that a digital data misread could not alter a soundwave (yay!).

Not so. In general misread digital data can be audible as clicks  or pops. If misread digital data is expected to be more frequent than a rare occurrence, sometimes circuits will be added to conceal the misread data without making a click or a pop. One such situation exists with standard CD player designs.

Quote
Any difference could also have been in my imagination (shock horror!).

Always possible, which is why we do scientific listening tests and technical tests to get us closer to the actual truth.

Quote
Out of curiosity, though, for those folk who know about such things, is there any way (perhaps a badly designed power supply?) that a laser head fighting to read a mistracked disc, or power fluctuations due to spin speed alteration for the same reason may have any effect on an outgoing analogue signal? (more so with 30-year old CD players?).

Yes, I'm clutching at straws in the hope that my youth wasn't quite as misguided as it looks like it might have been...  :-[

I'll give it to you straight. All things are possible if only a tiny amount, and while the probability of a given symptom can be guessed at or estimated from a distance, in reality one  has to be there to find the real truth.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: shakeshuck on 2017-05-05 13:59:44
In ripping about 2000 CDs I only had a handful that got any non-correctly corrected errors on first read at high speed.  Error handling on music CDs isn't great, but it's pretty good.

I'm just going through a similar exercise myself, and I can only say that I either I'm particularly unlucky, or you got off lightly!
Some disks will only rip from a certain PC, while others will only rip from another but not the first. I have about four machines I'm swapping between to try to find one that will work for that disk. It's crazy.
All the disks look good to the eye, too.

I read somewhere that CDRs are only 'lifed' (data wise) to about 25 years (although I've had some lose data much quicker than that), but I wasn't expecting to have any hassle with stamped ones.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: krabapple on 2017-05-05 17:42:42
In ripping about 2000 CDs I only had a handful that got any non-correctly corrected errors on first read at high speed.  Error handling on music CDs isn't great, but it's pretty good.

I'm just going through a similar exercise myself, and I can only say that I either I'm particularly unlucky, or you got off lightly!
Some disks will only rip from a certain PC, while others will only rip from another but not the first. I have about four machines I'm swapping between to try to find one that will work for that disk. It's crazy.
All the disks look good to the eye, too.

I read somewhere that CDRs are only 'lifed' (data wise) to about 25 years (although I've had some lose data much quicker than that), but I wasn't expecting to have any hassle with stamped ones.


I ripped over a thousand , mostly over ten years ago, using EAC and a ~2003-vintage Plextor drive and a standard Gateway or Dell PC, and had no such problems...
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: bennetng on 2017-05-05 18:18:29
Some disks will only rip from a certain PC, while others will only rip from another but not the first. I have about four machines I'm swapping between to try to find one that will work for that disk. It's crazy.
All the disks look good to the eye, too.

I read somewhere that CDRs are only 'lifed' (data wise) to about 25 years (although I've had some lose data much quicker than that), but I wasn't expecting to have any hassle with stamped ones.
CDR deterioration under 10 years is not news. How long will they die depends on the quality of disc and storage condition.

For pressed CD, it is possible that some drives + discs combination may cause trouble, but in my experience, it is not caused by deterioration. That means a newly bought disc can have such problems as well.

You can refer to spoon's post.
https://hydrogenaud.io/index.php/topic,111284.0.html

However, it seems that the thread is not related to PCM vs DSD anymore?
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: Arnold B. Krueger on 2017-05-05 19:29:47
In ripping about 2000 CDs I only had a handful that got any non-correctly corrected errors on first read at high speed.  Error handling on music CDs isn't great, but it's pretty good.

I'm just going through a similar exercise myself, and I can only say that I either I'm particularly unlucky, or you got off lightly!
Some disks will only rip from a certain PC, while others will only rip from another but not the first. I have about four machines I'm swapping between to try to find one that will work for that disk. It's crazy.
All the disks look good to the eye, too.

I read somewhere that CDRs are only 'lifed' (data wise) to about 25 years (although I've had some lose data much quicker than that), but I wasn't expecting to have any hassle with stamped ones.


I ripped over a thousand , mostly over ten years ago, using EAC and a ~2003-vintage Plextor drive and a standard Gateway or Dell PC, and had no such problems...

I've ripped about 2,000 pressed CDs in the past year or so, and had similar results. I ripped them as received which left me with about 20 bad ones, washed those with mild soap in the sink, and ripped the rest. They mostly looked like they were used at home and had a few light scratches or less, as compared to the rough patina that they sometimes pick up when used extensively in cars.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: saratoga on 2017-05-05 20:03:56

I've been fighting with a snap/crackle/pop problem with general file playback on Win 7 and USB, and have not yet quite bottomed the issue; I've gone as far as converting the Windows image from a raid setup to a single disk (which has helped, but I'm not there yet).

Changing the windows disk is very unlikely to help.  Usually problems like that are caused by bad audio drivers (no surprise) or bad chipset drivers.  Less often they can be caused by an unrelated device sharing the same bus.  I would start with drivers first though. 
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: shakeshuck on 2017-05-05 21:03:57

I've been fighting with a snap/crackle/pop problem with general file playback on Win 7 and USB, and have not yet quite bottomed the issue; I've gone as far as converting the Windows image from a raid setup to a single disk (which has helped, but I'm not there yet).

Changing the windows disk is very unlikely to help.  Usually problems like that are caused by bad audio drivers (no surprise) or bad chipset drivers.  Less often they can be caused by an unrelated device sharing the same bus.  I would start with drivers first though. 

After digging up posts from elsewhere, a suggestion was to try some software called LatencyMon. I ran this, and the main item flagged was a SCSI driver (which turned out to be a raid SATA chip), so I 'unraided' the OS.
The problem persists, but is better than it was.
Title: Re: PCM, DSD - Trying to get my head round some basics
Post by: 2tec on 2017-06-21 03:54:15
Edit: forgot to mention that you can still use the old ReplayGain scanner before libebur128 implementation by manually replacing the foo_rgscan.dll file. You can find the last version in foobar2000 v1.1.5 (http://filehippo.com/download_foobar2000/9356/)
it's here ~ http://filehippo.com/download_foobar2000/history/13/