The Optimal Sample Rate for Quality Audio

[Julf]

As for car analogies, I think the best car should have no cylinders at all... just a single big jet turbine will nicely fit the job. =P

 

Absolutely - if you own your own oil well

 

I have seen the Rover effort - and have one of the offsprings of that effort in my garage, but in hindsight, it was less than a resounding success.

[spdif-usb]

Absolutely - if you own your own oil well

 

I have seen the Rover effort - and have one of the offsprings of that effort in my garage, but in hindsight, it was less than a resounding success.

Hehe.

I know of no converter in the $1000 range that does 4x rates. Not talking what's on the spec sheet but what comes out of my speakers. Even at $2000, the best I've heard isn't even spec'd for 4x rates - but does a wonderful job at 96k.

I wasn't trying to suggest my DAC sounds fully transparent at 192 kHz, just that I think it sounds slightly better at 192 kHz when compared to 96 kHz. So, I have no other choice but to disagree with Dan Lavry's paper.

[bdiament]

Hi spdif-usb,

 

Hehe.

 

I wasn't trying to suggest my DAC sounds fully transparent at 192 kHz, just that I think it sounds slightly better at 192 kHz when compared to 96 kHz. So, I have no other choice but to disagree with Dan Lavry's paper.

 

The statement you quoted from me was a response to someone else's question. (Perhaps you already knew this.)

 

Still, I understand exactly where you are coming from. ;-}

 

Best regards,

Barry

Soundkeeper Recordings

Barry Diament Audio

[spdif-usb]

(Perhaps you already knew this.)

Yes, so I did.

Still, I understand exactly where you are coming from. ;-}

Would you please care to explain?

[esldude]

Hi elsdude,

 

 

 

To be clear, with regard to 4x rates, I said I find the threshold crossed*with the best converters I've heard*. I've also said many converters I've heard which have a "192" in their spec sheets actually perform *worse* (to my ears) than they do at 2x rates.

 

I know of no converter in the $1000 range that does 4x rates. Not talking what's on the spec sheet but what comes out of my speakers. Even at $2000, the best I've heard isn't even spec'd for 4x rates - but does a wonderful job at 96k.

 

Best regards,

Barry

Soundkeeper Recordings

Barry Diament Audio

 

Thanks Barry,

 

Just the kind of answer I was seeking as you have a perspective most of us don't get to have as someone who does high quality recording.

[bdiament]

Hi spdif-usb,

 

...Would you please care to explain?

 

It seems to me you prefer 192 over 96.

Did I misunderstand?

 

Best regards,

Barry

Soundkeeper Recordings

Barry Diament Audio

[spdif-usb]

It seems to me you prefer 192 over 96.

Did I misunderstand?

I would be exaggerating alot by saying I can't enjoy listening to music in 96 but yeah, on my system it seems like 192 adds that extra hint of convincing realism. It sounds somewhat more involving, with slightly better perceived dynamics and detail, as well as an improved three-dimensionality of the soundstage.

[sandyk]

"it seems like 192 adds that extra hint of convincing realism. It sounds somewhat more involving, with slightly better perceived dynamics and detail, as well as an improved three-dimensionality of the soundstage. "

 

That's exactly how my 73 year old ears hear it too, although according to most presently accepted theory, I shouldn't be able to do so. (grin)

I hope to hear the same with Barry's "Americas" as I did with "Equinox" in 24/192.

Alex

[Julf]

I wasn't trying to suggest my DAC sounds fully transparent at 192 kHz, just that I think it sounds slightly better at 192 kHz when compared to 96 kHz. So, I have no other choice but to disagree with Dan Lavry's paper.

 

Well, as the nit-picking anorak engineer, I have to point out that the fact that *your* DAC sounds slightly better at 192 K doesn't necessarily contradict Lavry's paper... A specific DAC might be better at handling one sample rate than another, but that might not apply as a general rule to all DACs.

[bdiament]

Hi Julf,

 

...A specific DAC might be better at handling one sample rate than another, but that might not apply as a general rule to all DACs.

 

I've seen statements like this from other folks and my question in response is always:

Have you ever heard a converter that did 4x rates (176.4k and 192k) well that did not do 2x rates well too? I haven't.

 

What I've heard are converters spec'd for 4x rates that don't deliver at those rates.

In my view, these are not "better at handling one sample rate than another" (though I do understand why some feels would describe them that way), they are simply marketed as being able to do something they can not.

 

Oh, perhaps they can fulfill a laboratory requirement to attain the number on the spec sheet but in my experience, operating at a certain frequency is not at all the same as being able to properly play Music sampled at that frequency.

 

As I see and hear it, those converters that show just how magical 4x rates are, that, for the first time in my experience, deliver recordings I cannot distinguish from the direct mic feed, do indeed contradict the paper (to put it mildly). Unless I misunderstood, the paper, I took it too say there is no benefit to rates beyond 24/96. I've never confused 24/96 with the mic feed. Never confused the finest analog with the mic feed either. So to me, either the paper is valid or having a recording that sounds like the mic feed is valid. They seem at odds with each other, so personally, I'm going with the latter.

 

Best regards,

Barry

Soundkeeper Recordings

Barry Diament Audio

[Julf]

Hi, Barry,

 

Have you ever heard a converter that did 4x rates (176.4k and 192k) well that did not do 2x rates well too? I haven't.

 

I definitely have - several DACs out there upsample everything to the highest rate, and the upsampling algorithm is of course always a compromise, so I can see why some of those DACs might work better with "native" 192K vs having to upsample 96K material.

 

What I've heard are converters spec'd for 4x rates that don't deliver at those rates.

 

Agree that just because a converter is 192K-capable on paper does not guarantee it does it well.

 

As I see and hear it, those converters that show just how magical 4x rates are, that, for the first time in my experience, deliver recordings I cannot distinguish from the direct mic feed, do indeed contradict the paper (to put it mildly). Unless I misunderstood, the paper, I took it too say there is no benefit to rates beyond 24/96. I've never confused 24/96 with the mic feed. Never confused the finest analog with the mic feed either. So to me, either the paper is valid or having a recording that sounds like the mic feed is valid. They seem at odds with each other, so personally, I'm going with the latter.

 

I am not doubting your observations. I specifically prefixed my comment with a disclaimer about being in "nit-picking anorak engineer" mode, and was just pointing out that there was an alternative explanation to the observation by spdif-usb on one DAC that wasn't necessarily contradicting the paper.

[bdiament]

Hi Julf,

 

...I definitely have - several DACs out there upsample everything to the highest rate, and the upsampling algorithm is of course always a compromise, so I can see why some of those DACs might work better with "native" 192K vs having to upsample 96K material...

 

We might be talking about two different things.

I was referring to the performance of a given converter at a given sample rate.

 

I tend to avoid converters that automatically convert everything to a single sample rate because I find their on-the-fly conversion to do more sonic damage than any theoretical benefit. Those always seem to brighten and harden the sound of the source material and while some may enjoy this type of sound, it isn't what I look for.

 

With that type of converter, I think it a *given* that they perform better when playing files at their native rate. I was referring to converters playing files at their native rate. I've never heard one that does 192 well that does not also do 96 well. And of those, I've never heard one that did 96 better than 192. This last only seems to occur in the converters that are *marketed* as 192 capable but which in fact, don't have the clocking that is up to the significantly increased demands of 192 and don't have the analog stage performance at the wider bandwidths.

 

Best regards,

Barry

Soundkeeper Recordings

Barry Diament Audio

[Julf]

Hi again, Barry,

 

We might be talking about two different things.

I was referring to the performance of a given converter at a given sample rate.

 

And I was referring to what spdif-usb was saying: "I wasn't trying to suggest my DAC sounds fully transparent at 192 kHz, just that I think it sounds slightly better at 192 kHz when compared to 96 kHz. So, I have no other choice but to disagree with Dan Lavry's paper.".

I don't know which DAC spdif-usb has, and if it upsamples 96 K to 192 K, or processes it natively. All I am saying is that *if* it does upsample, then that can cause (as you put it) "sonic damage". So that could explain why spdif-usb finds the DAC sounds better with 192K material than with 96K material. Without disagreeing with the Lavry paper.

[Miska]

Of course it isn't. That was actually my point, even my $1K DAC (which I consider relatively cheap) can do it.

 

And I do it for about tenth of the price...

 

Yeah but IMO that's overkill if your DAC is connected directly to your power amp, using no EQ nor preamp nor analog attenuation. The theoretical 144 dB of dynamic range you'll get with just 24-bit integer output already provides sufficient headroom due to thermal noise kicking in at around -120 dB, and 32-bit float internal data path ought to be just as good as 64-bit float internal data path for just upsampling 24-bit 192 kHz material. Or am I wrong?

 

Of course I'm also doing also room EQ etc at the same time. For 24-bit output it is possible to achieve same sample values with some of the upsamplers using 32-bit floats (not for all). But for the newer 32-bit DACs it makes a difference.

[spdif-usb]

I don't know which DAC spdif-usb has

It's the Eastern Electric MiniMax DAC Plus (with the vacuum tube removed from it and using its in-built M2Tech OEM async USB input). It's based on the ESS SABRE³² Reference ES9018 chip and with the tube removed IMO it sounds surprisingly closer to a Weiss DAC202 than the price difference would suggest. The Weiss is obviously more accurate and more analytical but I'd say the Weiss also can be more fatiguing in the long run exactly because of that. lol There is absolutely nothing wrong with the 96 kHz performance of the Plus. In fact, I think it's better than for example Wyred4Sound DAC2, by a mile or so.

[Julf]

It's the Eastern Electric MiniMax DAC Plus (with the vacuum tube removed from it and using its in-built M2Tech OEM async USB input). It's based on the ESS SABRE³² Reference ES9018 chip

 

I think the ES9018 internally upsamples everything to 192K.

[spdif-usb]

I think the ES9018 internally upsamples everything to 192K.

I think it upsamples everything to 1536 kHz but I also think it's extremely difficult, if not completely impossible, to find an affordable DAC that can make 96 kHz sound better than 192 kHz sounds on my DAC.

All affordable DACs that use the ES9018 sounded harsh to me, or both harsh and clinical, but not the Plus. None of the affordable DACs that don't use the ES9018 sounded to me like they could even begin to compete with the Plus, simply because the ES9018 chip typically always outputs such a maniacal amount of detail (true detail, not brightness) that it makes the dual mono Wolfsons sound completely muffled in comparison. For only $1K, that IMO is something really very special.

The Dagogo review Doug Schroeder wrote about the Plus was published after I had already purchased my Plus (I paid the introductory price for it). I still can't believe a single word of what he wrote in that review TBH. Increasing the value of a system by $50K? Go away, please... I keep telling my ears the cake is a lie, but my ears don't want to listen to me. Instead, they believe the review!

[Julf]

I think it upsamples everything to 1536 kHz

 

1536 kHz is the maximum rate the ES9018 is capable of (with a proviso in the data sheet about "oversample bypass mode only"), but I wasn't sure the Eastern Electric design used that frequency. In any case, the point is that it definitely upsamples the data - and for 96 kHz it has to upsample more than for 192 kHz.

 

I also think it's extremely difficult, if not completely impossible, to find an affordable DAC that can make 96 kHz sound better than 192 kHz sounds on my DAC.

 

I haven't actually heard that DAC, but based on other DACs with the ES9018, I am sure you are right.

[Sumflow]

See why many professional engineers still work at 96kHz years after 192kHz became available.

"An examination of converter IC data sheets will reveal that virtually all audio converter ICs deliver their peak performance near 96 kHz. The 4x (176.4 kHz and 192 kHz) mode delivers poorer performance in many respects. In most cases, noise, distortion, pass-band ripple, stop-band attenuation and other key performance measurements are significantly better in the 2X (88.2 kHz and 96 kHz) mode of operation.

 

Every A/D and D/A conversion IC that we have tested performs better at 96 kHz than at 192 kHz. In most cases THD+N, SNR, passband ripple, and stopband attenuation are all poorer at 192 kHz than at 96 kHz. Based upon these tests, I am not surprised that there is not yet any conclusive evidence that 192 kHz is better than 96 kHz. Given the current state of the art, 192 kHz should sound poorer than 96 kHz. 192 kHz provides additional bandwidth between 48 kHz and 96 kHz but there is no real evidence that this is useful given the limitations of our microphones, speakers, and hearing. 192 kHz adds useless bandwidth while decreasing performance."

 

From what Converter Manufacturers Don’t Want You to Know! http://www.benchmarkmedia.com/discuss/sites/default/files/Upsampling-to-110kHz.pdf

[bdiament]

I have come to think it is no simple task for a designer of digital converters to create one that does justice to the 4x sample rates (i.e., 176.4k and 192k). While there are plenty of decent sounding converters that do 2x rates (i.e. 88.2k and 96k), as I've said elsewhere, my experience has been that very few with "192" in their spec sheets seem to be able to *truly* deliver at these rates.

 

So, we get "truth"s and "white papers" from some quarters telling us "secrets" about why 4x rates are "a waste" or "useless" or "worse" than 2x rates. What I find interesting is that we don't hear this sort of nonsense from some other quarters, where instead, we get converters that *do* perform at the 4x rates and do so to an extent I wouldn't have imagined only a few years ago.

 

The best of these represent the first recording devices I've heard in my life that deliver results that truly sound like the mic feed prior to conversion. (I'm not talking about the ones that use phrases like that in their marketing literature, I'm referring to the ones that actually *do* it.) I am very glad these designers chose to apply their skills toward taking the recording and playback arts across the threshold into true transparency (again, something I've never heard before this, from any analog device and certainly any digital device - and of the latter, *none* operating at less than 4x rates).

 

Best regards,

Barry

Soundkeeper Recordings

Barry Diament Audio

 

P.S. I'll repeat what I recalled earlier of an experience all this nay-saying about 4x rates brings to mind. A manager at a large NYC recording studio once told me he was sure the female orgasm is a myth. His "proof"?: He'd been with dozens of women and not once did he ever witness one exhibiting signs of having an orgasm.

 

There you have it. Folks telling me 4x rates are "useless", etc. remind me of that manager.

[bdiament]

To the off-topic subject at hand, I was reporting what someone told me. It seems to have its parallels to folks telling me 192k is "useless".

 

Best regards,

Barry (not Bruce)

http://www.soundkeeperrecordings.com

http://www.barrydiamentaudio.com

[Jud]

 

"An examination of converter IC data sheets will reveal that virtually all audio converter ICs deliver their peak performance near 96 kHz. The 4x (176.4 kHz and 192 kHz) mode delivers poorer performance in many respects. In most cases, noise, distortion, pass-band ripple, stop-band attenuation and other key performance measurements are significantly better in the 2X (88.2 kHz and 96 kHz) mode of operation.

 

Every A/D and D/A conversion IC that we have tested performs better at 96 kHz than at 192 kHz. In most cases THD+N, SNR, passband ripple, and stopband attenuation are all poorer at 192 kHz than at 96 kHz. Based upon these tests, I am not surprised that there is not yet any conclusive evidence that 192 kHz is better than 96 kHz. Given the current state of the art, 192 kHz should sound poorer than 96 kHz. 192 kHz provides additional bandwidth between 48 kHz and 96 kHz but there is no real evidence that this is useful given the limitations of our microphones, speakers, and hearing. 192 kHz adds useless bandwidth while decreasing performance."

 

From what Converter Manufacturers Don’t Want You to Know! http://www.benchmarkmedia.com/discuss/sites/default/files/Upsampling-to-110kHz.pdf

 

Wow, what a credible paper (not). First we're "informed" that ASRC (asynchronous sample rate conversion) is better at removing jitter than synchronizing with the digital input (i.e., adaptive interface). There's no mention whatever that an asynchronous USB interface, taking the clock from within the DAC itself, is better than either at reducing jitter.

 

Then we are told in breathless tones about Benchmark determining that using sample rates well above Shannon-Nyquist minimum allows non-"brickwall" filtering. Glad that Benchmark figured this out, only 25 years or so after the first consumer DACs employing oversampling.

 

Then comes the stuff above, about DAC chips and performance at 96kHz. Of course nothing specific is said about the chips used by good designers in DACs that work well at 4x rates or with DSD, leading one to wonder whether this is a third instance in the same article of Benchmark leaving out important information. (My guess, based on the rest of the article, would be yes.)

[Julf]

Barry (not Bruce)

 

My sincerest apologies - my brain must still have been in the Monty Python universe as a result of the ear plug thread.

[Lavry tech]

While it is possible to get "good" results with converters optimized to operate at sample rates higher than 96kHz; this does not "prove" that higher sample rates do not come without a cost. The unfortunate fact that, by design, a multi-bit sigma delta converter optimized to operate at 192kHz (or higher) cannot also be optimized to operate at 96kHz makes it nearly impossible to make a meaningful comparison of the effects of changing ONLY the sample rate via listening tests. Other factors such as differences in analog circuitry, jitter in conversion clocking, or even PC board layout can have a significant effect on the perceived "sound" of the entire converter unit; which makes comparing different models of converters useless in this regard.

 

To help illustrate the difference between scientific facts and subjective test results I submit the following argument:

 

How about that joker Columbus? Can you believe that B.S. about the world being "round?" Anyone with eyes can just look around and see that the world is FLAT. Anyone.

 

As Dan Lavry points out in the following response; the Nyquist theorem is not intuitive. Trying to apply "common sense" analogies such as comparing samples to pixels only confuses the matter.

 

Dan Lavry 's response:

I have been making the case against higher sample rates for audio for a long time. I have encountered no credible arguments to my paper “Sampling Theory”. The same is true for my recent paper “The Optimal Sample Rate for Quality Audio”. I encounter some that want to counter the message by “shooting the messenger”. Meanwhile the facts I preset are correct and UN-challenged. I realize that reading the papers demands time and concentration. So here is a shorter description of many of the points I presented in the papers. Let’s refrain from diverting the conversation away from the topics.

 

1. Sampling is not intuitive. SAMPLING IS NOT ANALOGUS TO PIXELS! A more detailed picture may require more pixels, but more audio detail does NOT require more samples. There is an “electronic tool” (filter) that enables recovering ALL of the audio from a limited number of samples. It is not intuitive and requires much study. In fact it is counter-intuitive and goes against “everyday common sense.” This is the reason why the marketing of “more samples is better” is successful in convincing so many of the false notion.

2. Nyquist theorem (theorem is a PROVEN theory) tells us that recovering ALL the audio intact does require the sampling rate (frequency of sampling) to be at least twice as fast as the highest signal (audio) frequency. Theory demands a perfect “reconstruction tool” filter. In practice, real world filters require sampling a little faster than twice the audio bandwidth. For 20 KHz audio bandwidth, the theory requires at least 40 KHz sample rate. The 44.1 KHz standard provides 4.1 KHz margin. The margin for the filter (from the theoretical filter) is 100*(44.1KHz-2*20KHz)/(2*20KHz) = 10.25%

3. Some people argue that we need more than 20 KHz for audio. The decision as to how wide the audio range is should be left to the ears. Say we agree to accept a 25 KHz as the audio bandwidth. When using 88.2 KHz sampling, (and 25 KHz for the audio bandwidth) the margin is i100*(88.2KHZ-2*25KHz) /(2*25KHZ) = 76.4%.

4. At 96 KHz sampling and 25 KHz audio, the margin is 92%. At 96 KHz sampling and 30 KHz audio the margin is 60%. At 192KHz sampling and 30KHz the margin is 220%!. For anyone crazy enough to claim they hear or feel 40 KHz, when sampling at 192 KHz the margin is still 140%. At 384 KHz sampling the margin is 380%!

5. Some argue that at 44.1 KHz the margin of 10.25% is tight, and that theoretical filters fail to provide a near perfect reconstruction. Others argue that 20 KHz audio is too small to accommodate some ears. Such arguments support some reasonable increase in sampling rate. Many argue that 44.1 KHz rate is good enough. Others disagree. But few will argue with the statement that 44.1 KHz is at least pretty close to acceptable. In order to accommodate those that want improvements, let’s increase the margin by a factor of say 2. You want more, OK, by a factor of 4. You want more audio bandwidth? OK let’s raise it to a factor of 5… And all that is more than covered by the use of 96 KHz sample rate!

6. A few manufacturers are starting to advocate 384 KHz and even 768 KHz sample rates. When audio sampled at 44.1KHz is considered as being somewhere between “not perfect” and “near perfect”, the notion of sampling 870% faster (for 384KHz) or even 1741% (for 768KHz) faster than a CD makes no sense. I expect even the least competent of designers to be able to design a filter that does not require such huge margins. I would also expect any converter designer to have enough background to know that more samples are not analogous to more pixels! I would expect converter designers to insist that their marketing department knows that, instead of closing their eyes to the crock of steering audio in the wrong direction. I also understand it is not easy when one’s job is on the line.

7. It is not wise to keep increasing the sample rate unnecessarily. The files keep growing, and faster sampling yields less accuracy. Yet the marketing of higher sample rates has no basis, other than some spreading of misinformation. The latest I saw claims that faster sampling yields better stereo location (time resolution). The argument is false. Faster sampling offers the ability to process wider bandwidth, but has no impact what so ever on stereo location!

8. Faster sampling for capturing bandwidth that we do not hear (ultrasonic) is not wise. If we did not hear it (or feel it) we don’t need it. If we did hear it (or feel it) it is not ultrasonic, it is audible bandwidth (by definition). Ultrasonic energy may cause problems by spilling over to the audible range (intermodulation distortions). At best case, ultrasonic energy adds nothing to audio while requiring faster sampling, thus larger files and slower file transfers. In reality there is another price to pay; the faster one samples, the less accurate the result.

 

Dan Lavry

[The Computer Audiophile]

While it is possible to get "good" results with converters optimized to operate at sample rates higher than 96kHz; this does not "prove" that higher sample rates do not come without a cost. The unfortunate fact that, by design, a multi-bit sigma delta converter optimized to operate at 192kHz (or higher) cannot also be optimized to operate at 96kHz makes it nearly impossible to make a meaningful comparison of the effects of changing ONLY the sample rate via listening tests. Other factors such as differences in analog circuitry, jitter in conversion clocking, or even PC board layout can have a significant effect on the perceived "sound" of the entire converter unit; which makes comparing different models of converters useless in this regard.

Dan Lavry

 

I agree 100% with that paragraph.