Sampling Rates and Music Quality

Lee · May 8, 2012

This is an interesting read. It also explains to a degree why I believe the sampling rate is equal to or more important than the bit rate for true hirez listening.

http://www.mother-of-tone.com/cd.htm

My opinion goes like this:

We can have higher fidelity with 16 bits amplitude resolution and 10 or more sample-points per wave-period, as we can recover the original signal with a very small amount of beat products (less than1%).

As we cannot meet all those infinity requirements of exact sampling according to Shannon in the real world, and because of those issues that the sharp filters do not necessarily sound best to our ears, we can loosen all those sharp conditions by just sampling faster.

And then we can achieve better sound quality in reality, not just on the paper.

rbbert · May 8, 2012

A few years ago I thought sampling rate was a more important component of sound quality than bit-depth, and it may be once we get past 20 bits, but in the last couple of years I have listened to quite a bit of 24/44.1 and 24/48 music that sounds clearly better to me than the 16/44.1 versions made from them. Over in the 24/192 topic I think many of amir's posts have made clear why 16 bits is probably not quite adequate, although with good noise shaping it can come close.

Lee · May 8, 2012

rbbert said:
A few years ago I thought sampling rate was a more important component of sound quality than bit-depth, and it may be once we get past 20 bits, but in the last couple of years I have listened to quite a bit of 24/44.1 and 24/48 music that sounds clearly better to me than the 16/44.1 versions made from them. Over in the 24/192 topic I think many of amir's posts have made clear why 16 bits is probably not quite adequate, although with good noise shaping it can come close.

I agree that 24/44 beats 16/44 but in my experience, 24/96 or 24/192 or 24/176 is even better.

I think ProTools has unfortunately "trained" a lot of engineers to accept 24/44 and 24/48. Even a good sound-oriented band like Tom Petty has released 24/48 as a hirez file...but it is really is not. Historically 24/88.2 or higher has been considered hirez.

It's a shame that much modern music is being captured at 48khz and robbing the user of a much better listening experience.

Phelonious Ponk · May 8, 2012

To me it's all a moot point because, to roll out a rather shopworn cliche, it's about the music. SACD and DVDA had their run. Folks talked about Blue Ray audio, but it seems to be going nowhere fast. And I'm not interested in audiophile media because I have so little interest in audiophile music. There are hi-res digital masters of classic recordings out there, but they almost invariably come from analog masters, and every time I've peeled the veneer off and compared them to exactly the same thing down sampled to 16/44.1, it became obvious that it was the mastering, not the resolution that made the difference.

Would hi-res media of hi-res digital recordings sound better? Maybe. Probably. Is there ever going to be enough demand for it that the recordings I care about will be released in that format? I sincerely doubt it. Currently, the body of hi-res work available that I'm interested in is so small that it's not even worth running the copy of Pure Music I already own, to get them to play in their native resolution.

Hi-res is very close to hypothetical.

Tim

Lee · May 8, 2012

Phelonious Ponk said:
To me it's all a moot point because, to roll out a rather shopworn cliche, it's about the music. SACD and DVDA had their run. Folks talked about Blue Ray audio, but it seems to be going nowhere fast. And I'm not interested in audiophile media because I have so little interest in audiophile music. There are hi-res digital masters of classic recordings out there, but they almost invariably come from analog masters, and every time I've peeled the veneer off and compared them to exactly the same thing down sampled to 16/44.1, it became obvious that it was the mastering, not the resolution that made the difference.

Would hi-res media of hi-res digital recordings sound better? Maybe. Probably. Is there ever going to be enough demand for it that the recordings I care about will be released in that format? I sincerely doubt it. Currently, the body of hi-res work available that I'm interested in is so small that it's not even worth running the copy of Pure Music I already own, to get them to play in their native resolution.

Hi-res is very close to hypothetical.

Tim

Tim,

It is all about music but the reason that both engineers and audiophiles discuss hirez so much is because it is so much better! It's not subtle, it's not even incremental.

P.S. There is a lot of non-audiophile music that is best represented digitally by DVD-Audio and SACD. And both formats are far from done. DVD-Audio lives as hirez PCM downloads of great albums on HDTracks. SACD enjoys a thriving niche among classical and bop jazz lovers. And we even get an occasional landmark disc like Wish You Were Here.

Phelonious Ponk · May 8, 2012

Lee said:
Tim,

It is all about music but the reason that both engineers and audiophiles discuss hirez so much is because it is so much better! It's not subtle, it's not even incremental.

P.S. There is a lot of non-audiophile music that is best represented digitally by DVD-Audio and SACD. And both formats are far from done. DVD-Audio lives as hirez PCM downloads of great albums on HDTracks. SACD enjoys a thriving niche among classical and bop jazz lovers. And we even get an occasional landmark disc like Wish You Were Here.

We'll have to agree to disagree on the meaning of subtle and incremental. And most of the jazz I'm interested in that is available on hi-res was originally recorded on analog tape. I suspect we'd have to agree to disagree on the resolution of analog tape and whether or not it can be captured on 16/44.1 as well.

Tim

Lee · May 8, 2012

Phelonious Ponk said:
We'll have to agree to disagree on the meaning of subtle and incremental. And most of the jazz I'm interested in that is available on hi-res was originally recorded on analog tape. I suspect we'd have to agree to disagree on the resolution of analog tape and whether or not it can be captured on 16/44.1 as well.

Tim

No worries, I guess we have honest disagreements. I'm real impressed with the Blue Note SACDs that Chad has made from analog tape. Also check out the Duke Ellington Big Band 24/192 download on HDTracks. It sounds wonderful.

Let me know next time you get to Atlanta. We can meet at my friend Nick's and he can play some analog tapes that are 1:1 dubs of masters and they sound amazing. He has a J-Corder deck for playback now.

Bruce B · May 8, 2012

I don't think sampling rate has much to do with the quality of the recording. On the other hand, bit-depth has more of an impact.
I've been doing 16/44.1 CD's for FIM and no one has complained about the quality. But.... I sure wish I had the freedom to do all of these in 24bit though. Especially some of the Chinese drums! It's A LOT harder to make a great 16/44.1 CD than it is to make a good 24/44.1 file.

This is strictly speaking of PCM.... DSD is another issue.

Phelonious Ponk · May 8, 2012

Bruce B said:
I don't think sampling rate has much to do with the quality of the recording. On the other hand, bit-depth has more of an impact.
I've been doing 16/44.1 CD's for FIM and no one has complained about the quality. But.... I sure wish I had the freedom to do all of these in 24bit though. Especially some of the Chinese drums! It's A LOT harder to make a great 16/44.1 CD than it is to make a good 24/44.1 file.

This is strictly speaking of PCM.... DSD is another issue.

What makes it harder, Bruce? What are the challenges?

Tim

rbbert · May 8, 2012

Bruce B said:
I don't think sampling rate has much to do with the quality of the recording. On the other hand, bit-depth has more of an impact.
I've been doing 16/44.1 CD's for FIM and no one has complained about the quality. But.... I sure wish I had the freedom to do all of these in 24bit though. Especially some of the Chinese drums! It's A LOT harder to make a great 16/44.1 CD than it is to make a good 24/44.1 file.

This is strictly speaking of PCM.... DSD is another issue.

As with Bruce, in my experience the biggest quality jump is from 16 to 24 bits. One can easily demonstrate this for oneself (on playback only) by starting with a 24/96 or 24/88.2 file and listening after SRC and then again after decimating with dither. Now, Barry says that once you get the sampling rate from 96 up to 192, there is an even bigger quality jump. I'm not sure my digital hardware is good enough for me to appreciate that.

rbbert · May 8, 2012

Phelonious Ponk said:
... Is there ever going to be enough demand for it that the recordings I care about will be released in that format? I sincerely doubt it. Currently, the body of hi-res work available that I'm interested in is so small that it's not even worth running the copy of Pure Music I already own, to get them to play in their native resolution.

Hi-res is very close to hypothetical.

Tim

For older music this may be true. For newer music, as manufactured physical media (unfortunately) fades away, why will higher resolution digital be significantly different in availability than 16/44.1? Both may be niche markets compared to VBR or 256k MP3.

Andre Marc · May 8, 2012

Side note.

I think that higher bit depths and sample rates pay very little dividends for typical computer based Pro Tools similar recordings.

I would go as far as to say that an analog tape sampled at 44.1/16 can sound better than a 96/24 pro tools recording.

I think what is way more important is the original recording methodology.

Rush's Moving Pictures was one of the first big digital rock recordings and it was 44.1/16. It sounds phenomenal with MFSL release and the recent remaster.

Same thing for U2's Achtung Baby which was mixed to 44.1/16 and sounds way more real than modern rock recordings.

I don't believe it is about numbers.

However, when archiving classic natural sounding analog masters, 96 Khz or 192 khz are great to use.

Just my opinion.

Lee said:
This is an interesting read. It also explains to a degree why I believe the sampling rate is equal to or more important than the bit rate for true hirez listening.

http://www.mother-of-tone.com/cd.htm

DonH50 · May 8, 2012

The article has a lot of interesting technical issues, starting with the fact he's talking about sampling in general and not PCM (most of the discussion bears no relation to actual PCM encoding/decoding).

Higher sampling rate allows slower filter roll-off before the ADC (anti-alias filter) and after the DAC (image-reject filter). The resulting reduction of in-band filter artifacts is probably the biggest argument in favor of higher rates.

Higher resolution, i.e. more bits, has much more benefit because of the additional headroom. You always worry about clipping peaks, and with peaks 20 - 30 dB (or more?) in a lot of music you need to run the average signal level pretty durn low into the ADCs. That means low-level signals are being sampled with much lower effective resolution, meaning they do not sound as good, and you begin to approach the noise floor of a 16-bit device. If you give up 30 dB in headroom to allow for transient peaks, your loudest average sounds are now only using about 11 bits of your 16-bit converter, and the quiet passages that are say another 30 dB below that are only using 5 bits. Throw in studio or live sound where you have to mix a bunch of channels at various levels, unknown in advance, and it's pretty easy to see why 16 bits is not enough.

And yeah I agree the music and mastering dominates over the bits, but that does not mean bits don't matter...

My 0.000001 cents - Don

Phelonious Ponk · May 8, 2012

rbbert said:
For older music this may be true. For newer music, as manufactured physical media (unfortunately) fades away, why will higher resolution digital be significantly different in availability than 16/44.1? Both may be niche markets compared to VBR or 256k MP3.

Because there is a huge library of 16/44.1 already in existence and because hi-res will not be adopted by anyone other than the tiny audiophile market.

Tim

rbbert · May 8, 2012

Phelonious Ponk said:
Because there is a huge library of 16/44.1 already in existence and because hi-res will not be adopted by anyone other than the tiny audiophile market.

Tim

But I'm talking more specifically about music to be released (and some of it also recorded) in the future, and the huge existing library has little or nothing to do with that. If a recording is mastered at 24/192, why should it be any easier to release at 16/44.1? Slightly cheaper, perhaps, but right now CD's are often cheaper than MP3's, so file size alone is only part of the retail cost (and I don't want to get into discussing the reasons for that); releasing the 24/192 will bypass several processing steps.

Andre Marc · May 8, 2012

Norah?

Hey Bruce:

what is the deal with the new Norah Jones Come Away With Me "High Res" download in 192 and 96 Khz?

I though it was discussed here it is a native 44.1 recording/

Bruce B said:
I don't think sampling rate has much to do with the quality of the recording. On the other hand, bit-depth has more of an impact.
I've been doing 16/44.1 CD's for FIM and no one has complained about the quality. But.... I sure wish I had the freedom to do all of these in 24bit though. Especially some of the Chinese drums! It's A LOT harder to make a great 16/44.1 CD than it is to make a good 24/44.1 file.

This is strictly speaking of PCM.... DSD is another issue.

Andre Marc · May 8, 2012

Sorry, Bruce, and addendum..did I misunderstand..is the original two track mixdown analog, and the digital master 44.1?
Just wondering how we have a 192 Khz download out of the blue here.

FrantzM · May 8, 2012

DonH50 said:
The article has a lot of interesting technical issues, starting with the fact he's talking about sampling in general and not PCM (most of the discussion bears no relation to actual PCM encoding/decoding).

Higher sampling rate allows slower filter roll-off before the ADC (anti-alias filter) and after the DAC (image-reject filter). The resulting reduction of in-band filter artifacts is probably the biggest argument in favor of higher rates.

Higher resolution, i.e. more bits, has much more benefit because of the additional headroom. You always worry about clipping peaks, and with peaks 20 - 30 dB (or more?) in a lot of music you need to run the average signal level pretty durn low into the ADCs. That means low-level signals are being sampled with much lower effective resolution, meaning they do not sound as good, and you begin to approach the noise floor of a 16-bit device. If you give up 30 dB in headroom to allow for transient peaks, your loudest average sounds are now only using about 11 bits of your 16-bit converter, and the quiet passages that are say another 30 dB below that are only using 5 bits. Throw in studio or live sound where you have to mix a bunch of channels at various levels, unknown in advance, and it's pretty easy to see why 16 bits is not enough.

And yeah I agree the music and mastering dominates over the bits, but that does not mean bits don't matter...

My 0.000001 cents - Don

As per his now usual, Don has made sense. This to me close the chapter.

Lee · May 8, 2012

Bruce B said:
I don't think sampling rate has much to do with the quality of the recording. On the other hand, bit-depth has more of an impact.
I've been doing 16/44.1 CD's for FIM and no one has complained about the quality. But.... I sure wish I had the freedom to do all of these in 24bit though. Especially some of the Chinese drums! It's A LOT harder to make a great 16/44.1 CD than it is to make a good 24/44.1 file.

This is strictly speaking of PCM.... DSD is another issue.

Bruce, this is different from my experience. How specifically did you come to the conclusion that bit depth trumps sampling rate from a sonic standpoint?

I do hear some improvement from the "medium rez" 24/44 Beatles recordings but it is slight. However, notwithstanding the different mastering, the Love DVD-Audio is a clear upgrade on Beatles digital.

Lee · May 8, 2012

Also, with respect to Don's comments, are there not other improvements from sampling rate increases as suggested by the Coding2 paper from Bob Stuart?

http://www.meridian.co.uk/ara/coding2.pdf

Increase the sample rate by a margin sufficient to move the phase, ripple and transition regions
further away from the human audibility cut-off. One could probably make a sensible argument for
PCM sampled at 66.15kHz (44.1kHz times 1.5). The potential response is shown in figure 26.

Given our current position, there are strong arguments for maintaining integer relationships with
existing sampling rates – which suggests that 88.2kHz or 96kHz should be adopted. This would not be
an efficient way of conveying the relatively small extra bandwidth thought to be needed, but the impact
of using these higher rates can be substantially reduced by using lossless compression (packing).

Sampling Rates and Music Quality

Well-Known Member

Well-Known Member

Well-Known Member

New Member

Well-Known Member

New Member

Well-Known Member

WBF Founding Member, Pro Audio Production Member

New Member

Well-Known Member

Well-Known Member

Member Sponsor

Member Sponsor & WBF Technical Expert

New Member

Well-Known Member

Member Sponsor

Member Sponsor

Member Sponsor & WBF Founding Member

Well-Known Member

Well-Known Member

Similar threads