Hi guys.
I hope I put this in the right forum.
First, here's the video, and then afterwards I'll elaborate. The point in question is by David Robinson at 11:46, but this is a continuation of an answer to a question that was asked at 5:23 and then elaborated on by Robinson at 8:47
https://www.youtube.com/watch?v=FEr5UfBjDSc (https://www.youtube.com/watch?v=FEr5UfBjDSc)
So, my question is:
Is what David Robinson claiming true or false, and if what he is claiming is true, is it also audible?
hopefully you can tell me something along the lines of:
"yes, what the person says in the video is completely correct, and that phase shift is always audible on all CDs", or "no, his claim is completely false and it's never audible under any legitimate circumstances, unless a CD is intentionally made to be like that, but nobody would do that when releasing music", or "his claim of audibility only applies when something has gone wrong when making a CD, so under legitimate circumstances where nothing has gone wrong, there is neither any 'drift' in phase above 1 kHz, and nor can there be any audible effect, since there is no issue", or "in most cases, phase on CDs doesn't deviate to an extent that is worth worrying about" or another more fitting description.
It's also worth noting that David Robinson actually says "in the beginning", implying that it might be different now, but he still seems to be a CD hater.
I am aware that this topic may very easily go off-topic, so hopefully we can leave aside all the other claims made in the video, about 192 kHz sampling rates being audibly superior to 44.1 kHz ones, as I believe that's just another case of them not understanding the Nyquist theorem, and they haven't done any blind tests, and hopefully we can also leave out the claim that PCM and DSD sounds notably different, and that a recording of a vinyl disc will sound different than the vinyl disc itself, as the converter will leave it's sound imprint on the recording. While strictly speaking the latter might be true, then I like to think that most properly constructed converters nowadays are extremely close to 100 % transparancy and certainly transparent within audibility.
That panel would likely not accept the results of a double-blind/ABX test anyway. The fact they claim audible differences that couldn't be proven in an ABX test is also expected. The explanation of phase shift caused by PCM sampling is complete nonsense. They presume vinyl/analog sound "better" because it doesn't suffer from that nasty digital phase shift problem. Except...they clearly have never bothered to look at the phase response, or square wave response of an analog tape recorder, or that tape cut to vinyl. Spoiler: Yes, it's different than PCM, No, it's not better than PCM.
The explanation of phase shift caused by PCM sampling is complete nonsense.
I think this type of misinformation happens because a lot of people think PCM audio works in discrete 'ticks', and that if the ratio between sampling rate and frequency becomes too low, you don't have enough "temporal resolution" to fully capture the sound wave.
It's all based on a gross misunderstanding of the Nyquist-Shannon sampling theorem.
I would
love to play back a format that
actually worked like that to these people, just to see their absolutely horrified expressions.
I would love to play back a format that actually worked like that to these people, just to see their absolutely horrified expressions.
Be the most entertaining thing in the world for me to witness as well.
they haven't done any blind tests
There goes your fishing expedition. Next.
But I don't really understand this, so hopefully you can tell me something along the lines of:
"yes, what the person says in the video is completely correct, and that phase shift is always audible on all CDs", or "no, his claim is completely false and it's never audible under any legitimate circumstances, unless a CD is intentionally made to be like that, but nobody would do that when releasing music", or "his claim of audibility only applies when something has gone wrong when making a CD, so under legitimate circumstances where nothing has gone wrong, there is neither any 'drift' in phase above 1 kHz, and nor can there be any audible effect, since there is no issue", or "in most cases, phase on CDs doesn't deviate to an extent that is worth worrying about" or another more fitting description.
And the right answer is:
...there is neither any 'drift' in phase above 1 kHz, and nor can there be any audible effect, since there is no issue...
For example, consider a modern converter that is designed to be Linear Phase, which is to say it has a phase versus frequency response that is practically the same as that of an ideal delay over the entire audible range.
These guys were as aware of the phase response of analog tape or vinyl, or microphones, or speakers or the general run of high quality studio gear that is in general use, they'd "Have a cow". Then there is the slight matter of the fact that the human ear does not perceive phase changes applied to both channels equally above about 1 KHz. There's no process in place to do so - the ear functions like a spectrum analyzer without any way to determine or pass on phase response. Finally, their claim that DSD is somehow free of these problems is yet another fantasy.
I would love to play back a format that actually worked like that to these people, just to see their absolutely horrified expressions.
Be the most entertaining thing in the world for me to witness as well.
Actually, analog media works much that way.
I would love to play back a format that actually worked like that to these people, just to see their absolutely horrified expressions.
Be the most entertaining thing in the world for me to witness as well.
Actually, analog media works much that way.
In discrete 'ticks'? I'm... not reeeaaaally 100% onboard with that.
Sure, once you get down to the level of the individual PVC molecules or magnetic particles, I guess you could use that description. But it definitely doesn't sit quite right with me.
But I don't really understand this, so hopefully you can tell me something along the lines of:
"yes, what the person says in the video is completely correct, and that phase shift is always audible on all CDs", or "no, his claim is completely false and it's never audible under any legitimate circumstances, unless a CD is intentionally made to be like that, but nobody would do that when releasing music", or "his claim of audibility only applies when something has gone wrong when making a CD, so under legitimate circumstances where nothing has gone wrong, there is neither any 'drift' in phase above 1 kHz, and nor can there be any audible effect, since there is no issue", or "in most cases, phase on CDs doesn't deviate to an extent that is worth worrying about" or another more fitting description.
And the right answer is:
...there is neither any 'drift' in phase above 1 kHz, and nor can there be any audible effect, since there is no issue...
For example, consider a modern converter that is designed to be Linear Phase, which is to say it has a phase versus frequency response that is practically the same as that of an ideal delay over the entire audible range.
These guys were as aware of the phase response of analog tape or vinyl, or microphones, or speakers or the general run of high quality studio gear that is in general use, they'd "Have a cow". Then there is the slight matter of the fact that the human ear does not perceive phase changes applied to both channels equally above about 1 KHz. There's no process in place to do so - the ear functions like a spectrum analyzer without any way to determine or pass on phase response. Finally, their claim that DSD is somehow free of these problems is yet another fantasy.
Thanks for your help :-).
So just to make sure I understand you correctly, although you were pretty clear about it:
What David Robinson is saying is basically just made up? There's no phase drift/skewing?
I've seen people make up these things before, but as I didn't know much about this issue it sounded "scientifically enough" to make me doubt whether it was true or false. I've also looked at Ethan Winer's "Audio Myths" video, and in there he plays clips with phase shift at certain frequencies with no audible effect.
Robinson did also say "in the beginning", so I figured maybe this was before linear phase filters became common, which I think was in the early 90s (but I could be mistaken).
There goes your fishing expedition. Next.
Rich B couldn't do better.
The statement that, with 44.1kHz sampling, a 20 Hz sine is well sampled, but above 10k the phase is "totally random", sounds to me like a "grid snapping" mental model, where you imagine that sine wave peaks must be snapped to the nearest sample point. e.g. at the Nyquist frequency of 22.050kHz, you would imagine that there is only one possible sine wave phase.
This is incorrect, though. I suggest watching Monty's "Digital Show and Tell" video, but in particular the part at 20:55 addresses this idea: https://xiph.org/video/vid2.shtml
I would be happy to hear more comments about the phase issue in particular (instead of sample rates etc.) :-).
JJ pointed me to some of his papers available publicly, and I think I might finally understand those points in his papers filtutv1 and src1_9_16:
“Linear Phase” (constant delay)
- If a filter has a constant delay, the phase shift of the filter will be t*w, where t is the time delay, and w the natural frequency (2 pi f).
* This means that a delay can exhibit enormous phase shift.
* This phase shift, however, is ONLY delay.
* In most cases, a symmetric FIR (convolutional) filter is used. This kind of filter has a fixed (constant) delay over all frequencies, which means
* It has a phase shift, relative to the input, of 2*pi*f*t, where f is the frequency of interest, and t is the time delay. This “linear phase” means that the signal is purely delayed, all frequencies arrive at the same instant.
* It has a substantial amount of energy before the middle (main lobe) of the filter, being symmetric.
* If it’s poorly designed, or is too short, you can get pre-echo.
*Oddly, that doesn’t happen if it’s not too short.
*That’s another story
* This kind of filter design accounts for most filters in use
So, if I understand this correctly (and I might not), then the phase skewing that David Robinson talks about can only happen in other types of filters than linear phase filters.
This sort of phase issue "can" happen with any sampling frequency not perfectly factoring against the sample rate of components of the signal. The thing is, due to how signal reconstruction works, this does not matter.
This sort of phase issue "can" happen with any sampling frequency not perfectly factoring against the sample rate of components of the signal. The thing is, due to how signal reconstruction works, this does not matter.
I don't think so. I've never seen this happen, and I've looked for it. Please provide an independent reliable reference or a test sample.
There are some odd effects such as those where quantization distortion can be zero with some precondidtions including sample rate being a strict multiple of the signal frequency, but that is amplitude distortion, not phase distoriton. Also, if your signal frequency is within a few Hz of the Nyquist frequency, things can get strange, but again that is a very special case.
Please educate me! ;-)
I would be happy to hear more comments about the phase issue in particular (instead of sample rates etc.) :-).
JJ pointed me to some of his papers available publicly, and I think I might finally understand those points in his papers filtutv1 and src1_9_16:
“Linear Phase” (constant delay)
- If a filter has a constant delay, the phase shift of the filter will be t*w, where t is the time delay, and w the natural frequency (2 pi f).
* This means that a delay can exhibit enormous phase shift.
* This phase shift, however, is ONLY delay.
* In most cases, a symmetric FIR (convolutional) filter is used. This kind of filter has a fixed (constant) delay over all frequencies, which means
* It has a phase shift, relative to the input, of 2*pi*f*t, where f is the frequency of interest, and t is the time delay. This “linear phase” means that the signal is purely delayed, all frequencies arrive at the same instant.
* It has a substantial amount of energy before the middle (main lobe) of the filter, being symmetric.
* If it’s poorly designed, or is too short, you can get pre-echo.
*Oddly, that doesn’t happen if it’s not too short.
*That’s another story
* This kind of filter design accounts for most filters in use
So, if I understand this correctly (and I might not), then the phase skewing that David Robinson talks about can only happen in other types of filters than linear phase filters.
True.
If the phase distortion due to time delay were an audible problem duing music playback, then all forms of media, not just digital would be greatly affected because we listen to recordings minutes to decades after the live performance, and the phase shift due to delay ranges from astronomical to even a whole lot bigger.
Furthermore I got the impression that the speaker was referring to a phase shift that wanders around, while the phase shift due to the usual run of digital filters is fixed in place, rock solid. This contrasts with analog media, where the phase shift (and amplitude response) wanders around, often fairly audibly.
May he's objecting to digital's inherent constant phase shift as opposed to his preferred analog media phase shift that wanders around. And wander around, analog media does. It is not unusual for there to be 1 dB and greater variations within the audible range with both vinyl and analog tape playback.
Probably quantization effects, then. Not lining up with the frequency of the waveform, it can result in sample points appearing to be some sort of gibberish, until you do something like intelligently plot a curve between the points, like a reconstruction filter would.
Don't mind me, I only have a rudimentary understanding of this stuff, and have zero college education to boot. I pick up bits and bobs as I crawl the web and read people's code, and maybe even learn something!
I did manage to sort of guess an algorithm that turned out to be really close to what Paulstretch does, though. I guess I'm not totally out of touch.
I would be happy to hear more comments about the phase issue in particular (instead of sample rates etc.) :-).
JJ pointed me to some of his papers available publicly, and I think I might finally understand those points in his papers filtutv1 and src1_9_16:
“Linear Phase” (constant delay)
- If a filter has a constant delay, the phase shift of the filter will be t*w, where t is the time delay, and w the natural frequency (2 pi f).
* This means that a delay can exhibit enormous phase shift.
* This phase shift, however, is ONLY delay.
* In most cases, a symmetric FIR (convolutional) filter is used. This kind of filter has a fixed (constant) delay over all frequencies, which means
* It has a phase shift, relative to the input, of 2*pi*f*t, where f is the frequency of interest, and t is the time delay. This “linear phase” means that the signal is purely delayed, all frequencies arrive at the same instant.
* It has a substantial amount of energy before the middle (main lobe) of the filter, being symmetric.
* If it’s poorly designed, or is too short, you can get pre-echo.
*Oddly, that doesn’t happen if it’s not too short.
*That’s another story
* This kind of filter design accounts for most filters in use
So, if I understand this correctly (and I might not), then the phase skewing that David Robinson talks about can only happen in other types of filters than linear phase filters.
True.
If the phase distortion due to time delay were an audible problem duing music playback, then all forms of media, not just digital would be greatly affected because we listen to recordings minutes to decades after the live performance, and the phase shift due to delay ranges from astronomical to even a whole lot bigger.
Furthermore I got the impression that the speaker was referring to a phase shift that wanders around, while the phase shift due to the usual run of digital filters is fixed in place, rock solid. This contrasts with analog media, where the phase shift (and amplitude response) wanders around, often fairly audibly.
May he's objecting to digital's inherent constant phase shift as opposed to his preferred analog media phase shift that wanders around. And wander around, analog media does. It is not unusual for there to be 1 dB and greater variations within the audible range with both vinyl and analog tape playback.
So, I just want to be competely sure I understand what both you, Arny, and the rest of you guys are saying:
On digital medias, there IS a phase shift, but it's never audible. And the phase shift is constant, so therefore there is no audible effect, as the phase shift manifests purely as a delay. Is that correct?
On analogue medias, there is even more phase shift, and it's not constant and doesn't manifest itself as a delay, so it can be audible. Is that also correct?
Are you also saying that on analogue medias, the phase shift varies from left to right channel, which is when phase shift actually becomes audible, and to some, like David Robinson, this phasing effect actually sounds pleasing, whereas the lack of audible phase shift, sounds "flat" to his ears.
So, David Robinson's claim was partly true, partly false...?
I'm also partly asking about this, as Michael Fremer bascially SCREAMED at me a few months ago on Youtube that vinyl has better phase resonse than CDs (as well as screaming that CDs have never been more transparent than, or as transparent as, vinyl + vinyl has wider bandwidth, as it can go to 30 kHz, but of course I do know that the claim of transparency is bullshit, and the claim of bandwidth is mosltly irrelevant, and vinyl rarely goes above 20 kHz except for distortion in the form of harmonics).
So, David Robinson's claim
Michael Fremer bascially SCREAMED
Both indispensable tools in the tackle box.
So, I just want to be competely sure I understand what both you, Arny, and the rest of you guys are saying:
On digital medias, there IS a phase shift, but it's never audible.
Given the really small phase shifts that we can now measure reliably and accurately, there is always some measurable phase shift.
However, in common kinds of digital audio gear, which is generally very good, well designed, and well made, there are no reliably audible artifacts while playing normal audio such as music, dialog, and dramatic sound effects.
And the phase shift is constant,
No, it takes certain combinations of phase shift versus frequency to obtain linear phase. zero phase shift at all frequencies is one of them, but it is impossible in the real world except for trivial cases.
... so therefore, there is no audible effect, as the phase shift manifests purely as a delay. Is that correct?
Yes, designing a linear phase filter with no audible effects is very doable. It could be an undergraduate class project in EE.
Designing linear phase 44.1 KHz and above sample rate reconstruction filters is a generally solved problem. For example there are at least two software products, one pretty expensive and one a freebie that will grind out the parameters for building an appropriate filter if you tell it what you want; This is known as a Cookbook design problem.
On analogue medias, there is even more phase shift, and it's not constant and doesn't manifest itself as a delay, so it can be audible. Is that also correct?
Yes. One of the big systematic lies in high-end audio is publicly obsessing over certain performance attributes of digital, ignoring the fact that they are so well solved in the digital domain as to not be audible problems, and simply not mentioning that they were never that well solved in the analog domain, and are often very audible there. Two that come to mind quickly is jitter and phase shift.
Are you also saying that on analogue medias, the phase shift varies from left to right channel, which is when phase shift actually becomes audible,
Having differing phase shifts in the two stereo channels vastly ups the odds of hearing it reliably as compared to having the identical phase shift in both channels. Having the phase shift vary pretty quickly with time makes it more audible, too. Combining it with concurrent variations in amplitude response increases the audibility as well. All of this is inherent and was never solved with analog media. They are inherent, and there is not much that can be done about it, except sometimes at great expense (FM recording) or by cutting to the chase and going to digital. By inherent I mean they are due to things like the geometry of LP grooves, cartridge armatures, analog tape heads, and turntable arms. You make them out of Unobtainium using those of Santa's Elves who have PhDs and who are also journeyman watchmakers, in a secret lab in Never-Never land, and the problems will still be there.
and to some, like David Robinson, this phasing effect actually sounds pleasing, whereas the lack of audible phase shift, sounds "flat" to his ears.
[/qite
So, David Robinson's claim was partly true, partly false...?
I'm also partly asking about this, as Michael Fremer bascially SCREAMED at me a few months ago on Youtube that vinyl has better phase resonse than CDs (as well as screaming that CDs have never been more transparent than, or as transparent as, vinyl + vinyl has wider bandwidth, as it can go to 30 kHz, but of course I do know that the claim of transparency is bullshit, and the claim of bandwidth is mosltly irrelevant, and vinyl rarely goes above 20 kHz except for distortion in the form of harmonics).
Human nature being what it is, adding the sound of a Troll gargling with pi$$ mouthwash to every recording may be someone's preference, but it isn't mine. Probably not yours, either.
Michael Fremer bascially SCREAMED at me a few months ago on Youtube...
Lucky me, he did it to me live and in person just before the 2005 Stereophile debate between John Atkinson and I. Somehow I recovered, I hope you do to, if you have not already done so... ;-)
Michael Fremer bascially SCREAMED at me a few months ago on Youtube...
Lucky me, he did it to me live and in person just before the 2005 Stereophile debate between John Atkinson and I. Somehow I recovered, I hope you do to, if you have not already done so... ;-)
Fremer also added on Youtube that his friends call him friendly and agreeable, but that's not how he comes across online at all, where he often has CAPS LOCK on AUTO and calls everybody morons if they simply present facts or voices their opinions, if any of those two go against his "all analogue medias are technically superior to digital".
Two more things:
Arny, do you know if other types of digital filters, for instance minimum phase filters or the filterless non-oversampling DACs, have audible phase shift?
And lastly, here's a quote from a different website. I tried what this person suggested with several CDs. With many CDs I couldn't hear any difference, but with some I actually heard a big difference (and no, not like a $2000 power cord difference, but a real, immediately noticeable difference). If any of you happen to have the "About a boy" soundtrack by Badly Drawn Boy (or you could download and burn a CD-R, or get it from the library) that was the one where I heard the biggest difference, but I heard differences with other albums as well. I connected the speaker wires as red to red and black to black on my speakers, and black to red and red to black on my amp when doing this. On a different website, someone claimed that 92 % of CDs and CD players inverted the polarity, and he listed many albums where this was the case.
If you're able to try this out, what do you think of this? I've tried hitting the "switch polarity" in Wavelab, and there was no audible difference.
The answer to this age-old (well, the age of the first CD player anyway) argument among audiophiles is far more simple than the above explanation.
I’ve worked on (repair and restoration of) high-end audio equipment for years now. The reason vinyl is perceived as “better” than CD does indeed lie in the DAC, but the reasons listed above are minor compared to the real issue: The inversion of the polarity of the signal.
Some people mistakenly call this “phase inversion”, but that is an argument for another time. Unless someone makes a mistake installing the cartridge, turntables do not invert the polarity of the signal. As long as the rest of the amplifier/speaker combination is set up or built correctly, the signal is presented in the correct, natural polarity.
I have yet to run across ANY CD player/DAC that DIDN’T invert the polarity of the signal, even the $12,000 Sonic Frontiers CD player I changed the laser in once. I don’t know why this is, unless the red book specifications did it to limit the dynamic range of CD playback. You see, back then it was a common mistake for people to turn the volume up too far, because once they hit “play”, they were expecting to hear a little bit of the “hiss” of a tape or (cheap) high-gain phono preamp.
Anyhow, in defense of that Sonic Frontiers player, and a few nicer other machines as well, there is a “polarity invert” button, which in most cases corrects the polarity of the signal. Once the sound is presented to you in the correct polarity from the CD player, there is basically no perceivable difference. We’ve tested it many times, with turntables and CD players ranging from $50 to $22,000. Always the same results.
Before you call me crazy (something I will not deny actually), try this simple test: Listen to a recording using anything you like for a source. Then, immediately after, switch your speaker connectors to be “backwards”, i.e. + to – and – to +. You will most likely find that if you are using a CD player (or an iPod or computer, they all tend to invert polarity), the sound will be MUCH improved, and if you are listening to vinyl, the sound will get MUCH worse. If you find the opposite to be true, your system may have been inverting from the very beginning.
The whole polarity inversion thing is bullshit as well, unless you're using speakers with severe excursion nonlinerarity.
The whole polarity inversion thing is bullshit as well, unless you're using speakers with severe excursion nonlinerarity.
I made a typo in my post, although not related to what you're saying. In Wavelab it was "invert phase", not "switch polarity".
Nevertheless, the whole polarity thing I consider complete bullshit as well - which is why I was so surprised to hear those big differences with certain CDs.
On my phono preamp I have a polarity inversion switch, which doesn't have any audible effect.
But if possible, please try out that experiment I mentioned :-).
As far as I know my speakers don't have problems.
I've tried it myself, not a lick of difference.
Funny thing, JBL used to label their speaker terminals backwards for decades, ie. a positive impulse at the black terminal would make the speaker cone move forwards. This was opposite to just about every other speaker maker in the world. And no one noticed, unless they were setting up a system with multiple speakers and needed to match the polarity and phase. I guarantee you that no home users actually noticed anything out of the ordinary.
The Chromecast Audio also inverts the signal, as do a lot of other devices. It seems to be almost random, and no one cares. Well, no one except obsessive audiophiles ;-)
I've tried it myself, not a lick of difference.
Funny thing, JBL used to label their speaker terminals backwards for decades, ie. a positive impulse at the black terminal would make the speaker cone move forwards. This was opposite to just about every other speaker maker in the world. And no one noticed, unless they were setting up a system with multiple speakers and needed to match the polarity and phase. I guarantee you that no home users actually noticed anything out of the ordinary.
The Chromecast Audio also inverts the signal, as do a lot of other devices. It seems to be almost random, and no one cares. Well, no one except obsessive audiophiles ;-)
I agree with this - which is also why I was surprised to actually hear a difference in that test with SOME CDs. Many CDs showed no difference at all. So if you keep trying this, maybe you will hear a difference eventually, although I am aware that you might not do this as you clearly find it pointless :-).
I agree with this - which is also why I was surprised to actually hear a difference in that test with SOME CDs. Many CDs showed no difference at all. So if you keep trying this, maybe you will hear a difference eventually, although I am aware that you might not do this as you clearly find it pointless :-).
You should definitely ABX test this.
The whole polarity inversion thing is bullshit as well, unless you're using speakers with severe excursion nonlinerarity.
When AM was a commonly used medium polarity really matters because minus going modulation is limited by the laws of physics to -100%, but positive going modulation is limited only by available power. FM' limits are symmetrical. Many natural sounds including many forms of the human voice are consistently asymmetrical, again with negative going peaks being limited to 100%, but only practical limits to postive peaks.
Polarity is pretty audible with certain steady test signals that are wholly in the audible range. For example 1F + 2F over a wide range. However, natural music and voice are not steady, and this seems to make a huge difference in audibility.
Of course anybody who bases judgement in this area with sighted evaluations is fooling themselves, and this is pretty easy to demonstrate as well.
Historical note: Vanderkooy and Lipshitz were introduced to ABX testing by our group here in Detroit as part of some polarity tests they wanted to run. They had developed a randomized AB switcher which was pretty good all by itself. We jointly ddi ABX tests related to polarity, some of which may have shown up in their subsequenty JAES paper. A few months later they showed up with a modified tester, that now could do ABX tests... ;-)
I agree with this - which is also why I was surprised to actually hear a difference in that test with SOME CDs. Many CDs showed no difference at all. So if you keep trying this, maybe you will hear a difference eventually, although I am aware that you might not do this as you clearly find it pointless :-).
You should definitely ABX test this.
Sure, but I can't at the moment, as my belongings are in storage. In this case it would have to be a "human" ABX as well, as someone would have to change the speaker wires.
But believe me, it wasn't a small difference. I've done plenty of ABX tests, and I've passed several with 16 out of 16 correct, where I hadn't even thought I would be able to hear a difference, and I have failed several - for instance 320 kbps mp3 vs. wave, hi-res vs. CD quality and others. So I'm not an obsessive audiophile who pays silly money for placebo effects. I just want to find out what was going on :-).
So if you can get a hold of that Badly Drawn Boy soundtrack to "About a boy" you will probably also hear a drastic difference - unless of course my speakers are seriously broken, ha ha ha (but I doubt that) :-).
I feel like adding my $0.02 to this, hopefully to increase understanding and not confuse you.
So, I just want to be competely sure I understand what both you, Arny, and the rest of you guys are saying:
On digital medias, there IS a phase shift, but it's never audible. And the phase shift is constant, so therefore there is no audible effect, as the phase shift manifests purely as a delay. Is that correct?
You first need to realize that the term "phase shift" implicitly assumes a reference, against which you could determine a shift. There is no "freestanding" phase shift, it is always relative to some reference. It is important to understand what the reference is, or else you can't understand the point being made. Since this reference is often tacitly assumed, there is ample chance for misunderstandings, and hence misinformation.
Another important point is that phase is a concept that relates to a periodic signal, which will always have a certain frequency. There is no sense in speaking about phase when the signal at hand isn't periodic. The only exception to this is a sloppy usage of the term in conjunction with a polarity reversal of the signal, which gets associated with a phase shift of 180 degrees. People usually know what is meant, but strictly speaking a phase shift by 180 degrees is not the same thing as a polarity reversal, a fact that becomes quite obvious once you deal with signals other than pure sine waves.
The term "phase distortion" usually implies that the reference is a different frequency component
within the same signal. For example, if you take the phase of a 1 kHz frequency component as the reference, the phase of a 2 kHz component may be shifted against it. Since signals that are not pure sine waves contain multiple frequency components, such phase distortion may alter the looks of the signal waveform on a display quite clearly, even though this kind of waveshape distortion is usually inaudible.
Many kinds of filters produce this kind of distortion as a side effect of their frequency-dependent behavior, particularly those in the analog domain. Digital filters of the linear phase type don't, but they have no equivalent in the analog domain, so digital technology is actually better than analog in this respect.
A pure delay causes a frequency-dependent phase shift with respect to the undelayed signal. It doesn't matter whether the delay was caused by digital circuitry or in another way. This is purely due to the fact that higher frequencies need less delay to accomplish a certain phase angle. The relaitionship is a simple mathematical one. It only becomes an issue when delayed and undelayed versions of the same signal get combined in a single signal, and the effect of this is known as comb filtering. Both analog and digital systems behave the same in this respect. The only difference is that it is more common to encounter some delay in a digital system than in an analog system. If you don't combine delayed and undelayed signal versions, you don't have a problem. In a pure playback situation, this shouldn't bother you.
On analogue medias, there is even more phase shift, and it's not constant and doesn't manifest itself as a delay, so it can be audible. Is that also correct?
If the phase shift is caused by a delay, it can't be constant across the frequencies. If the phase shift is constant across the frequencies, it can't be caused by a delay. That's due to mathematics.
Analog circuitry or media, as soon as it has a bandwidth limitation (i.e. always) will distort phase to some extent. But whether that's audible is a very different question. Phase distortions are by themselves unlikely to be audible.
Are you also saying that on analogue medias, the phase shift varies from left to right channel, which is when phase shift actually becomes audible, and to some, like David Robinson, this phasing effect actually sounds pleasing, whereas the lack of audible phase shift, sounds "flat" to his ears.
You probably mean slightly different delays between left and right. They can easily arise in analog media because of slight misalignments of tape heads or pickups. If you are fancy enough, try to calculate the relative delay between left and right channel in a stereo tape head depending on the angle between the head gap and tape motion. You will develop a feeling how sensitive to mechanical tolerances such a system is. Digital technology is close to ideal in this regard, beating practical analog systems by orders of magnitude.
So, David Robinson's claim was partly true, partly false...?
It was mainly bullshit. It is hard to say whether he was right or wrong when it isn't very clear what he actually tried to say. My impression is that he referred to neither delay nor phase distortion, but to the effects of sampling frequency and signal frequency being asynchronous to each other, leading to time-varying sampling positions on the signal's waveform. It looks like a shifting phase of the sampling clock with respect to the signal. This kind of phase shift is completely bogus as it doesn't affect the signal at all in a correctly implemented sampling system. The exact point on the waveform where the sampling happens is of no consequence as long as the bandwidth limitation is adhered to, on which digital sampling depends. So if my suspicion is correct, he just doesn't understand sampling.
I'm also partly asking about this, as Michael Fremer bascially SCREAMED at me a few months ago on Youtube that vinyl has better phase resonse than CDs (as well as screaming that CDs have never been more transparent than, or as transparent as, vinyl + vinyl has wider bandwidth, as it can go to 30 kHz, but of course I do know that the claim of transparency is bullshit, and the claim of bandwidth is mosltly irrelevant, and vinyl rarely goes above 20 kHz except for distortion in the form of harmonics).
If you compare CD with vinyl, you can do that fairly by recording a vinyl playback on a CD and then compare the vinyl playback with the CD recording. Or you can do the reverse and press the material found on CD on a vinyl disc, and then compare the two. It has been done several times. The result is as clear as the day: In the second case you can hear a clear difference, in the first case you can't. That has got to mean that the CD is the transparent medium and the vinyl disc isn't. That should be the end of the story, and in fact it was the end of the story 30 years ago. Nothing has changed since then.
If Fremer says anything that differs from that, it is bullshit. And if he's screaming it, it is screaming bullshit. He's full of it anyway, I wouldn't listen to it for a second.
You should definitely ABX test this.
Sure, but I can't at the moment
That would sink the boat
You should definitely ABX test this.
Sure, but I can't at the moment
That would sink the boat
???
If Fremer says anything that differs from that, it is bullshit. And if he's screaming it, it is screaming bullshit. He's full of it anyway, I wouldn't listen to it for a second.
Yup! Those that shout the loudest, usually have the least to say...
Thanks for your lengthy response, Pelmazo. It did indeed help me understand the issue better :-).
It's too bad that for panel debates like that there are no people present who actually understand the technologies. When people like those on the panel are allowed to speak freely, they often spread bullshit just to further their own agenda, and people who don't know enough about the technical issues (like myself) just assume it must be true, because these people have authority (although I'm a lot more suspicious than the average guy in the crowd).
Unfortunately, up until recently I've taken Framer and certain other audiophiles seriously. I agree with some of the things he says (not in the panel debate), but I could probably also say that of the most dispicable person I could think of. But I suppose dealing with a guy who thinks a $5000 power cord makes an audible difference, and then refuses to take any kind of blindtest to back up that claim as well as all his other claims, should probably have tipped me off a long time ago to ignore anything coming from him. Alas! What a life, eh?
It's too bad that for panel debates like that there are no people present who actually understand the technologies.
You have to understand that most of these events are sales events. Anything that make people think twice about their impulse buys has to be kept away.
When people like those on the panel are allowed to speak freely, they often spread bullshit just to further their own agenda, and people who don't know enough about the technical issues (like myself) just assume it must be true, because these people have authority (although I'm a lot more suspicious than the average guy in the crowd).
Exactly right, and by the way I totally agree with your positive comments about P's great post.
The more you know about real technology, the more high end sales events can irritate you. As a rule, people who actually know what's happening in audio avoid them like the plague. Its like over a decade later and I still remember the two days I spend perusing HE2005. Executive Summary: I came away very sad contemplating how many $millions are wasted on overpriced stuff that might even be pretty good, but is still vastly overpriced, and then there was the immense pile of outright snake oil and garbage.
Unfortunately, up until recently I've taken Framer and certain other audiophiles seriously. I agree with some of the things he says (not in the panel debate), but I could probably also say that of the most despicable person I could think of.
He's definitely close to The Edge, emotionally and way over The Edge intellectually.
He's a very powerful man in High End Audio it would seem, and as they say: Great Power corrupts greatly.
But I suppose dealing with a guy who thinks a $5000 power cord makes an audible difference, and then refuses to take any kind of blindtest to back up that claim as well as all his other claims, should probably have tipped me off a long time ago to ignore anything coming from him. Alas! What a life, eh?
I should point out that Real Working Audio Technology is both vastly different and also can be a lot of fun. It is more frustrating for most because in Real Reality things don't always work out the first time you try them. Some times they don't work out at all!
However, if you can enter a state of Zen where the value of the occasional reward is enhanced by the hard work and frustration it always seems to involve, it is very good.
The more you know about real technology, the more high end sales events can irritate you. As a rule, people who actually know what's happening in audio avoid them like the plague. Its like over a decade later and I still remember the two days I spend perusing HE2005. Executive Summary: I came away very sad contemplating how many $millions are wasted on overpriced stuff that might even be pretty good, but is still vastly overpriced, and then there was the immense pile of outright snake oil and garbage.
What's also a bit funny/sad is that when I found the 2005 debate on Youtube (haven't watched it though), I saw an equal split in the comments between "That Atkinson guy should really shut up! He's ruining it for everybody" and "That Krueger guy should really shut up! He's ruining it for everybody" :-D :-(.
On a different note, and to return to topic, I was told that ringing and phase are basically the same thing. Can you (or anybody) else say how "important" or "audible" ringing is in properly made digital audio? I was thinking maybe you could say it was -100 dB or -60 dB down, or perhaps you could even link to some files specifically made for the purpose that has increased the level of ringing so it's clearly audible (you once posted files like that with and without jitter so we could hear the difference).
The more you know about real technology, the more high end sales events can irritate you. As a rule, people who actually know what's happening in audio avoid them like the plague. Its like over a decade later and I still remember the two days I spend perusing HE2005. Executive Summary: I came away very sad contemplating how many $millions are wasted on overpriced stuff that might even be pretty good, but is still vastly overpriced, and then there was the immense pile of outright snake oil and garbage.
What's also a bit funny/sad is that when I found the 2005 debate on Youtube (haven't watched it though), I saw an equal split in the comments between "That Atkinson guy should really shut up! He's ruining it for everybody" and "That Krueger guy should really shut up! He's ruining it for everybody" :-D :-(.
Scientific facts are not established by means of public opinion polls. I'm surprized the split was 50-50, since my experience is that the pro-science, anti-audio-snake oil view is more generally unpopular. The pro-snake oil forums seem to outdraw the scientific ones by large margins.
On a different note, and to return to topic, I was told that ringing and phase are basically the same thing.
I'd say that one is a cause and the other is an effect in real-world situations, but which is the cause and which is the effect flip=-flops a lot. ;-)
Can you (or anybody) else say how "important" or "audible" ringing is in properly made digital audio?
In real world, mainstream, well-engineered gear, such ringing and phase shift as exists is not audible.
I was thinking maybe you could say it was -100 dB or -60 dB down, or perhaps you could even link to some files specifically made for the purpose that has increased the level of ringing so it's clearly audible (you once posted files like that with and without jitter so we could hear the difference).
Any ringing that is -100 or -60 dB down is wayyy into inaudible. You can have just about any artifact that exists or is imaginable at -100 dB and it is highly unlikely to be audible. At -60 dB down the story changes from "Highly unlikely" to "Pretty Unlikely".
Audibility depends on things like the regularity of the effect and percentage of time affected. For example you can have a ringing-induced but damped peak that is +3 dB for one cycle right below the 22.05 KHz Nyquist for 44/16 and its not audible. If it rings all of the time @ +3 dB at 13 KHz, then it might be audible. Ringing much of the time at + 3 dB at 3 KHz is going to be audible unless the range of frequencies that stimulates it is very narrow so it rarely happens.
Board,
This is worth a watch - https://www.youtube.com/watch?v=BYTlN6wjcvQ
It covers many of the topics raised in this thread.
Stephen, thanks for the suggestion. I actually watched that video more than once. I'm a fan of Ethan's :-). But it might have been too long since I watched it, although I was in touch with Ethan before making my post here, and he did direct me to this video and the part about phase shift. Still, I was in doubt about the issue at hand (hence this topic).
Arny, again, thanks for your help. My 50-50 comment was just a shot from the hip, so the real number might be very different. But yes, the scientific approach seems less popular overall.
Just to make sure I understand you correctly: The way I understand ringing in digital audio is that it's pretty much present constantly, but that it's always around the Nyquist frequency, so if there's a +3 dB spike around 22 kHz (or even at 20 kHz) it won't be audible due to it being at such high frequencies. But is your example of ringing at lower frequencies something you would actually find in real life?
As for the level, on a general level would you say the level of ringing in properly made digital audio is closer to -100 dB or -60 dB (if that's possible for you to say)?
In one of JJ's online powerpoint presentations he mentioned that he had looked at thousands of CD tracks and the anti-imaging filtering were questionable at best. I don't know if that's related to ringing.