• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

High Resolution Audio: Does It Matter?

Soniclife

Major Contributor
Forum Donor
Joined
Apr 13, 2017
Messages
4,499
Likes
5,417
Location
UK
What you say may be quite true about the aforementioned Barry Diament, judging by his many, many old comments at CA Forum. Not a fan of his, by the way. It seems also true about Mark Waldrep, again not a particular favorite of mine..
Mark has done the blind test now, and failed it, credit to him for saying so.

I don’t pay much attention to rock/pop engineers or labels, myself. I do listen to classical engineers, a few of whom I know personally, and honestly I do not see them claiming to hear better than we do or engaging in flagrant self promotion.
The ones you talk to, who you respect, who say they can hear a difference, has that been double blind tested?
 

Soniclife

Major Contributor
Forum Donor
Joined
Apr 13, 2017
Messages
4,499
Likes
5,417
Location
UK
It may not make sense to all, especially the skeptics and nay sayers here, but the concept of also not limiting yourself in recording for archiving purposes makes a great deal of sense to me. The “RBCD or 96k is good enough” argument is not without considerable controversy and it also lacks concrete proof beyond question, both now and into the future. That the benefits of 352.8k are totally insignificant to nonexistent vs. say, 96k, might be viewed differently years from now, at least in customer belief or perception from a marketing perspective.
I don't see anyone arguing the recording should not be made in higher resolution formats. Personally I would encourage an engineer to use whatever they think is the most accurate ADC they have access to.
 

Fitzcaraldo215

Major Contributor
Joined
Mar 4, 2016
Messages
1,440
Likes
632
And none are particularly good at designing experiments. So I don't think -- I'm quite sure really, in Diament's case, he being a veritable fount of utter nonsense about digital -- any of them have their own DBT results to back up their claims. Bias control being an *elementary* step in any such investigation.
Not saying Diament’s or anybody else’s tests are valid from the perspective of standards used for scientific inquiry. I have no idea how any recording engineers may have reached their conclusions, via controlled DBT testing or otherwise. They did not publish details of their decision making process, and they are not audio scientists in the business of designing and formally testing audio perception.

My only point earlier is that by whatever processes, scientifically valid or not, different engineers believe different things about hirez, about different sampling rate thresholds, about PCM vs. DSD, etc. @Kees de Visser confims that. Some use ultra, ultra hirez only because the customer believes in it and requests it. Others use it because they themselves are convinced it improves the quality of the recording.

Of course, audio amateurs and consumers have also provided many listening anecdotes about their own firm conclusions on the matter, but based on unreliable testing.

So, there is no simple consensus and no rigorous scientific testing on determining a simple, magic sampling rate number that is objectively sufficient in all cases. There is uncertainty, and people are likely to disagree forever on what the sampling rate ought to be, both recording engineers and consumers. Yes, audio science advocates hate that uncertainty. But, some things are just that way.

Consumers are free to make their own judgements, and recording labels and engineers are free to invest in the technology for higher sampling rates. Some feel that it may all just be marketing hype, but that lacks conclusive evidence, as well. And, choosing hirez as a distribution format involves the penalty of selling into a much smaller, niche market, as not everyone has the playback capabilities. So, usually, the bets are hedged and releases done in multiple formats, your choice as a consumer and listener.

I know some would probably like to see a royal edict forbidding use of sampling rates above 44k, or 96k, etc. under penalty of death. For better or worse, we don’t live in a world where that is possible. And, what is the harm, anyway? If you don’t believe in the improvement, just don’t buy it.
 

Fitzcaraldo215

Major Contributor
Joined
Mar 4, 2016
Messages
1,440
Likes
632
The ones you talk to, who you respect, who say they can hear a difference, has that been double blind tested?

Blind, yes, double blind, I don’t think so. But, I do not know the details. And, remember, they test for their own purposes, not for scientific publication or boasting rights in online forums. Well, except maybe for boasting by the aforementioned Barry Diament. I own none of his recordings myself.
 

Blumlein 88

Grand Contributor
Forum Donor
Joined
Feb 23, 2016
Messages
20,521
Likes
37,050
Not saying Diament’s or anybody else’s tests are valid from the perspective of standards used for scientific inquiry. I have no idea how any recording engineers may have reached their conclusions, via controlled DBT testing or otherwise. They did not publish details of their decision making process, and they are not audio scientists in the business of designing and formally testing audio perception.

My only point earlier is that by whatever processes, scientifically valid or not, different engineers believe different things about hirez, about different sampling rate thresholds, about PCM vs. DSD, etc. @Kees de Visser confims that. Some use ultra, ultra hirez only because the customer believes in it and requests it. Others use it because they themselves are convinced it improves the quality of the recording.

Of course, audio amateurs and consumers have also provided many listening anecdotes about their own firm conclusions on the matter, but based on unreliable testing.

So, there is no simple consensus and no rigorous scientific testing on determining a simple, magic sampling rate number that is objectively sufficient in all cases. There is uncertainty, and people are likely to disagree forever on what the sampling rate ought to be, both recording engineers and consumers. Yes, audio science advocates hate that uncertainty. But, some things are just that way.

Consumers are free to make their own judgements, and recording labels and engineers are free to invest in the technology for higher sampling rates. Some feel that it may all just be marketing hype, but that lacks conclusive evidence, as well. And, choosing hirez as a distribution format involves the penalty of selling into a much smaller, niche market, as not everyone has the playback capabilities. So, usually, the bets are hedged and releases done in multiple formats, your choice as a consumer and listener.

I know some would probably like to see a royal edict forbidding use of sampling rates above 44k, or 96k, etc. under penalty of death. For better or worse, we don’t live in a world where that is possible. And, what is the harm, anyway? If you don’t believe in the improvement, just don’t buy it.

The harm is in proliferating formats without end. The harm is there being some cost in time, effort and even physical resources to satisfy these multiple formats. The harm is increasing the cost of music because of so many formats.

As for consensus is there anyone scientifically knowledgeable in the field who thinks there is any benefit beyond 96 khz? I think one reason you see so little rigorous testing of so called hi-res is prior scientific knowledge of hearing makes the idea of 352 khz ludicrous. Knowledge of noise sources makes 32 bit ridiculous.

I suppose my distaste for some of these ideas is similar to the distaste for alternative medicine. The alternative medicine is more dangerous, but there are plenty of similarities with high end audio beliefs. That won't prevent the billions of dollars being made in the non-prescription boner pill market for just one example. But that also seems a weak defense of such practices all things considered.
 

Fitzcaraldo215

Major Contributor
Joined
Mar 4, 2016
Messages
1,440
Likes
632
The harm is in proliferating formats without end. The harm is there being some cost in time, effort and even physical resources to satisfy these multiple formats. The harm is increasing the cost of music because of so many formats.

As for consensus is there anyone scientifically knowledgeable in the field who thinks there is any benefit beyond 96 khz? I think one reason you see so little rigorous testing of so called hi-res is prior scientific knowledge of hearing makes the idea of 352 khz ludicrous. Knowledge of noise sources makes 32 bit ridiculous.

I suppose my distaste for some of these ideas is similar to the distaste for alternative medicine. The alternative medicine is more dangerous, but there are plenty of similarities with high end audio beliefs. That won't prevent the billions of dollars being made in the non-prescription boner pill market for just one example. But that also seems a weak defense of such practices all things considered.
Economics and competitive markets take care of the proliferation without end and the cost issues. The guys with actual monetary skin in the game - labels, engineers and buying consumers - ought to be the ones making those choices, not you unilaterally on behalf of all of them. Also, as @Kees de Visser pointed out, the incremental engineering cost is insignifant. Manufacturing cost may be significantly higher for an SACD or BD-A vs. a CD. So, the consumer pays a few extra bucks for it. Big deal. His choice. And, most studios that do hirez, don’t do CDs. So, they can’t use price inflated CD revenues to subsidize losses in hirez sales. Any, why would they even if they could? If hirez ain’t profitable on its own, shut it down. Business 101.

But, in reality, with physical discs declining and streaming or downloads increasing, it may be a new ball game, possibly one that provides even better support than now for format proliferation. Welcome to the 21st Century. Choices are proliferating in many, many consumer goods, like it or not.

96k? I know of no killer, slam dunk science that gives you the simple, universal answer for all time you want. Apparently, some labels feel the same. So, many are still messing with higher rez and/or DSD. Agreed, 32-bit depth is ridiculous, but I don’t see any mad rush to go there on the recording/distribution side. It’s there in DACs essentially only for bit-shifting digital volume controls. I also do not seek out downloads of PCM352.8 or DSD256. I agree with the notion that enough is enough. RBCD is not high enough, though, in my estimation as a recording or distribution format. But, my line in the sand of where that limit should be drawn probably differs from yours and from many other labels, engineers and consumers. Maybe, just maybe, they hear things you or I don’t or vice versa.

Alternative medicine or snake oil are really poor analogies. I see you have you own mind made up, but can’t you do better than that?
 

Blumlein 88

Grand Contributor
Forum Donor
Joined
Feb 23, 2016
Messages
20,521
Likes
37,050
Economics and competitive markets take care of the proliferation without end and the cost issues. The guys with actual monetary skin in the game - labels, engineers and buying consumers - ought to be the ones making those choices, not you unilaterally on behalf of all of them. Also, as @Kees de Visser pointed out, the incremental engineering cost is insignifant. Manufacturing cost may be significantly higher for an SACD or BD-A vs. a CD. So, the consumer pays a few extra bucks for it. Big deal. His choice. And, most studios that do hirez, don’t do CDs. So, they can’t use price inflated CD revenues to subsidize losses in hirez sales. Any, why would they even if they could? If hirez ain’t profitable on its own, shut it down. Business 101.

But, in reality, with physical discs declining and streaming or downloads increasing, it may be a new ball game, possibly one that provides even better support than now for format proliferation. Welcome to the 21st Century. Choices are proliferating in many, many consumer goods, like it or not.

96k? I know of no killer, slam dunk science that gives you the simple, universal answer for all time you want. Apparently, some labels feel the same. So, many are still messing with higher rez and/or DSD. Agreed, 32-bit depth is ridiculous, but I don’t see any mad rush to go there on the recording/distribution side. It’s there in DACs essentially only for bit-shifting digital volume controls. I also do not seek out downloads of PCM352.8 or DSD256. I agree with the notion that enough is enough. RBCD is not high enough, though, in my estimation as a recording or distribution format. But, my line in the sand of where that limit should be drawn probably differs from yours and from many other labels, engineers and consumers. Maybe, just maybe, they hear things you or I don’t or vice versa.

Alternative medicine or snake oil are really poor analogies. I see you have you own mind made up, but can’t you do better than that?

An open mind is good as long as it isn't so open your brain falls out.

If I simply go by what most recording people use, its likely above 95% that record at 48 or 44 khz. You seem to be willing to give the benefit of the doubt without question until a negative is proven otherwise. I've tried everything up to 192 recording and playing back. I don't hear any improvement over 48 khz. There is some small areas where maybe a little higher could be better. Beyond 96 or even 88 until someone comes up with some new discovery so far unseen, unknown and unheard there is nothing behind the idea. I think the alternative medicine analogies are spot on.

I agree that current streaming and downloading encourage format proliferation even more so. Which is why it is such a shame that it doesn't help. Other consumers, engineers and labels maybe, and really more than just maybe are hearing something that isn't really there.
 

danadam

Addicted to Fun and Learning
Joined
Jan 20, 2017
Messages
956
Likes
1,496
It may be just my cynicism, but:
Agreed, 32-bit depth is ridiculous, but I don’t see any mad rush to go there on the recording/distribution side.
Give it time :) Maybe demand is not big enough... yet.
 
Last edited:

Blumlein 88

Grand Contributor
Forum Donor
Joined
Feb 23, 2016
Messages
20,521
Likes
37,050
I may be just my cynicism, but:

Give it time :) Maybe demand is not big enough... yet.

Yes. I believe the latest RME boxes do the super high 768 khz sample rate on the ADC, and I believe they also claim 32 bit recording. AKM offers several versions of ADC chips that are 768 khz sample rate with 32 bit.

I'm sure the smaller steps sound more analog and smoother with less glare. Don't they always?

RME ADI-Pro AD/DA
Analog Inputs
XLR
• Input: XLR, servo-balanced • Input sensitivity switchable +24 dBu, +19 dBu, +13 dBu, +4 dBu @ 0 dBFS
• Signal to Noise ratio (SNR) @ +13/19/24 dBu: 120.x dB RMS unweighted, 124.x dBA
• Signal to Noise ratio (SNR) @ +4 dBu: 119 dB RMS unweighted, 123 dBA
• Frequency response @ 44.1 kHz, -0.1 dB: 5 Hz – 20.5 kHz
• Frequency response @ 96 kHz, -0.5 dB: 3 Hz – 45.5 kHz
• Frequency response @ 192 kHz, -1 dB: 2 Hz – 92.7 kHz
• Frequency response @ 384 kHz, -1 dB: < 1 Hz – 124 kHz
• Frequency response @ 768 kHz, -3 dB: < 1 Hz – 180 kHz
• THD @ -1 dBFS: -116 dB, 0.00016 %
• THD+N @ -1 dBFS: -112 dB, 0.00025 %
• Channel separation: > 110 dB
 
Last edited:

Sal1950

Grand Contributor
The Chicago Crusher
Forum Donor
Joined
Mar 1, 2016
Messages
14,068
Likes
16,598
Location
Central Fl
I'm sure the smaller steps sound more analog and smoother with less glare. Don't they always?
They get closer and closer to the glorious sound of vinyl every day.. One day even Fremer may approve. LOL
 

Sal1950

Grand Contributor
The Chicago Crusher
Forum Donor
Joined
Mar 1, 2016
Messages
14,068
Likes
16,598
Location
Central Fl
^ LMAOROTF
 

Soniclife

Major Contributor
Forum Donor
Joined
Apr 13, 2017
Messages
4,499
Likes
5,417
Location
UK
If I simply go by what most recording people use, its likely above 95% that record at 48 or 44 khz.
I was under the impression that the majority use more than 48khz these days, but I cannot find anything online to say one way or another.
 

svart-hvitt

Major Contributor
Joined
Aug 31, 2017
Messages
2,375
Likes
1,253
I was under the impression that the majority use more than 48khz these days, but I cannot find anything online to say one way or another.

Apple as of 2012:

«Even so, many experts feel that using higher resolution PCM files during production provides better-quality audio and a superior listening experience in the end product. For this reason, 96/24 resolution is quickly becoming a standard format in the industry, and it’s also common to see higher resolution files, such as 192/24».
Source: https://www.apple.com/ca/itunes/mastered-for-itunes/docs/mastered_for_itunes.pdf

However, we see repeatedly that old geezers question the correctness of Apple’s official statement ;)
 

andreasmaaan

Master Contributor
Forum Donor
Joined
Jun 19, 2018
Messages
6,652
Likes
9,399
In my limited experience, mastering engineers tend to expect the recording to be delivered to them at 88.2 or 96KHz.

I presume this is to ensure that anti-aliasing filters don't cascade anywhere near the audible band. I imagine this is very genre/scene/scale dependent though, and as I say, my experience is limited.

EDIT: I'm talking here about masters that are destined for RBCD release.
 
Last edited:

Soniclife

Major Contributor
Forum Donor
Joined
Apr 13, 2017
Messages
4,499
Likes
5,417
Location
UK
Apple as of 2012:

«Even so, many experts feel that using higher resolution PCM files during production provides better-quality audio and a superior listening experience in the end product. For this reason, 96/24 resolution is quickly becoming a standard format in the industry, and it’s also common to see higher resolution files, such as 192/24».
Source: https://www.apple.com/ca/itunes/mastered-for-itunes/docs/mastered_for_itunes.pdf

However, we see repeatedly that old geezers question the correctness of Apple’s official statement ;)
That's about the best sort of quote I could find, it suggests it's becoming the norm, but there are no stats to back it up.

I don't see any problem with what apple is saying here, it seems sensible, they claim to be able to create better 16/44.1 lossy files from a 24/96 starting point, mathematically they are probably right. And it's the same logic for why I think using more than CD throughout the chain makes sense even when CD is the target. There is a lot of stuff out there about how DAW plugins sound better when fed 24/96 than lower, how much is BS I have no idea but again mathematically it makes sense to maintain the quality for as long as possible, and only drop it at the final stage. None of this answers the question of if that final 24/96 master sounds any different to the final CD to human ears.
 

svart-hvitt

Major Contributor
Joined
Aug 31, 2017
Messages
2,375
Likes
1,253
That's about the best sort of quote I could find, it suggests it's becoming the norm, but there are no stats to back it up.

I don't see any problem with what apple is saying here, it seems sensible, they claim to be able to create better 16/44.1 lossy files from a 24/96 starting point, mathematically they are probably right. And it's the same logic for why I think using more than CD throughout the chain makes sense even when CD is the target. There is a lot of stuff out there about how DAW plugins sound better when fed 24/96 than lower, how much is BS I have no idea but again mathematically it makes sense to maintain the quality for as long as possible, and only drop it at the final stage. None of this answers the question of if that final 24/96 master sounds any different to the final CD to human ears.

I guess Apple have access to pretty good in-house statistics themselves, but it seems like they (and others?) have a policy of not being transparent on their own big data.

So the question is: Do we think Apple are lying?
 
Top Bottom