• 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!

AES Paper Digest: Differences among Several High Sampling Digital Recording Formats

Blumlein 88

Grand Contributor
Forum Donor
Joined
Feb 23, 2016
Messages
20,524
Likes
37,057
Casting doubt is easy. Backing up doubts in a credible way is more difficult.
I don't see olvavoi's comments as casting doubt. More like asking questions from his level of understanding hoping to learn more. To illustrate where he is coming from he makes comments of how he's thinking about the topic. I don't see in forum context that such needs to be done as if peer review research level commentary.
 

Wombat

Master Contributor
Joined
Nov 5, 2017
Messages
6,722
Likes
6,459
Location
Australia
General comment: When someone cites a comment, and adds a comment of their own, it is easy for the first commenter to perceive the second comment as related to his own comment, even though it is prefixed by the words "general comment". Just a general comment, of course.

Fair enough but some are more sensitive than others and assuming sensitivity can be taken as condescension. I may be getting the hang of this. ;)
 

Wombat

Master Contributor
Joined
Nov 5, 2017
Messages
6,722
Likes
6,459
Location
Australia
I don't see olvavoi's comments as casting doubt. More like asking questions from his level of understanding hoping to learn more. To illustrate where he is coming from he makes comments of how he's thinking about the topic. I don't see in forum context that such needs to be done as if peer review research level commentary.

Discussion for discussion's sake on well established matters leads to lengthy discourse with usually little or no benefit. Philosophical pursuits can be undertaken as such in clearly entitled threads or in extremis in Fight Club. Continual engaging in 'what if's'(hypotheticals) is a distraction from the core..

My .o2c.
 

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
4,600
Likes
3,065
Location
Zg, Cro
Casting doubt is easy. Backing up doubts in a credible way is more difficult.

Definitely so. Maybe we can throw the light from this side to make it more clear to them: It seems to me @oivavoi and @Cosmik are much more doubtful toward this Japanese paper than toward the Swedish one - I'm really in doubt if it is becuase they like Swedes more than Japanese people! :D :D
 

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
4,600
Likes
3,065
Location
Zg, Cro
Fair enough but some are more sensitive than others and assuming sensitivity can be taken as condescension.

Not to mention that some also get angry when their arguments are questioned! ;)

P.S. Ok, switching off now as I don't want to be blamed for another pissing contest! :D
 

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
4,600
Likes
3,065
Location
Zg, Cro
Discussion for discussion's sake on well established matters leads to lengthy discourse with usually little or no benefit. Philosophical pursuits can be undertaken as such in clearly entitled threads or in extremis in Fight Club. Continual engaging in 'what if's'(hypotheticals) is a distraction from the core..

My .o2c.

Cannot agree more with this, hence switching off. ;)

P.S. Have a nice weekend down under! :)
 

oivavoi

Major Contributor
Forum Donor
Joined
Jan 12, 2017
Messages
1,721
Likes
1,934
Location
Oslo, Norway
Discussion for discussion's sake on well established matters leads to lengthy discourse with usually little or no benefit. Philosophical pursuits can be undertaken as such in clearly entitled threads or in extremis in Fight Club. Continual engaging in 'what if's'(hypotheticals) is a distraction from the core..

My .o2c.

But the discussion here is not on an "established matter". Is it an established matter whether hi-rez or different sampling rates either can or can't be detected? I don't think so. Some of the scientific articles posted in this thread clearly contradict each other.

Definitely so. Maybe we can throw the light from this side to make it more clear to them: It seems to me @oivavoi and @Cosmik are much more doubtful toward this Japanese paper than toward the Swedish one - I'm really in doubt if it is becuase they like Swedes more than Japanese people! :D :D

Which Swedish paper? Do you mean the discussion about dac testing from the other thread? Well, AFAIK @Cosmik thinks obsessing about normal well-designed dacs is a complete waste of time, and I tend to agree with him (any hypothetical audible difference that well-designed dacs could create pales completely when compared to the distortion and smeared time response one typically finds in normal passive speakers). For my part, that dac debate is/was mostly a question of trying to stay open to evidence which conflicts with my convictions. I have spent quite some time claiming on audio forums that all modern dacs sound the same, and I therefore find it interesting that there may be evidence that contradicts this conviction.

But in any case: Wishing you an excellent weekend!
 
Last edited:

Blumlein 88

Grand Contributor
Forum Donor
Joined
Feb 23, 2016
Messages
20,524
Likes
37,057
But the discussion here is not on an "established matter". Is it an established matter whether hi-rez or different sampling rates either can or can't be detected? I don't think so. Some of the scientific articles posted in this thread clearly contradict each other.



Which Swedish paper? Do you mean the discussion about dac testing from the other thread? Well, AFAIK @Cosmik thinks obsessing about normal well-designed dacs is a complete waste of time, and I tend to agree with him (any hypothetical audible difference that well-designed dacs could create pales completely when compared to the distortion and smeared time response one typically finds in normal passive speakers). For my part, that dac debate is/was mostly a question of trying to stay open to evidence which conflicts with my convictions. I have spent quite some time claiming on audio forums that all modern dacs sound the same, and I therefore find it interesting that there may be evidence that contradicts this conviction.

But in any case: Wishing you an excellent weekend!

Hopefully not too far further off topic. I too have said well designed modern DACs are a solved problem. OTOH, Amir's testing has turned up some units not so well designed. More than I would have expected. There seems little exotic about the differences. The differences come down to crossing T's and dotting I's during design and manufacture.
 

Blumlein 88

Grand Contributor
Forum Donor
Joined
Feb 23, 2016
Messages
20,524
Likes
37,057
@amirm This paper is done in an impressive way, thank you for sharing it! Comparing it's result to this paper, do you think they correlate or conflict with each other?

https://www.academia.edu/441305/Sampling_Rate_Discrimination_44.1_KHz_Vs._88.2_KHz

Edit: If I understood correctly the difference in the "Sampling Rate Discrimination.." paper was observed only with complex (orchestral) music so it would be interesting to know what kind of music material was used in the "Differences among.." paper.

I've read this paper, but it has been awhile. I don't have time to review it here so hopefully not adding noise to the discussion. There was discrimination between downsampled 88 vs native 88, and downsampled 88 vs native 44. Indicating the downsampling was audible. But native 88 vs 44 was not discriminated at better than the 5% level. Though it tended heavily toward a difference below the 5% confidence level.

All of which seems to indicate what has been the case for a 20+ years now. There might maybe could be for trained young listeners differences in 44 and more than 44 khz rates. The difference must be fairly small. Record at 88 or 96 to leave no doubt. Does 48 khz get us over the hump so it never sounds different than higher rates?

Ah, who cares? We've market forces for 192, 384 and the 768 khz DACs (are there any 768 khz ADCs for recording yet?).

Maybe we need the thread I've had in mind about whether upsampling can offer any measured benefits. I'm pretty sure such benefits are exhausted long before 768 khz.
 

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
4,600
Likes
3,065
Location
Zg, Cro
I've read this paper, but it has been awhile. I don't have time to review it here so hopefully not adding noise to the discussion. There was discrimination between downsampled 88 vs native 88, and downsampled 88 vs native 44. Indicating the downsampling was audible. But native 88 vs 44 was not discriminated at better than the 5% level. Though it tended heavily toward a difference below the 5% confidence level.

All of which seems to indicate what has been the case for a 20+ years now. There might maybe could be for trained young listeners differences in 44 and more than 44 khz rates. The difference must be fairly small. Record at 88 or 96 to leave no doubt. Does 48 khz get us over the hump so it never sounds different than higher rates?

Ah, who cares? We've market forces for 192, 384 and the 768 khz DACs (are there any 768 khz ADCs for recording yet?).

Maybe we need the thread I've had in mind about whether upsampling can offer any measured benefits. I'm pretty sure such benefits are exhausted long before 768 khz.

Frankly, I really see no reason for the downsampling and upsampling to exist at all if aim is high quality reproduction. While downsampling can theoretically be done right, upsampling even in theory cannot restore the information that is not there any more - it can only try to interpolate it but that is of course guesswork. And that is the reason why I fully agree with you when you said "record at 88 or 96 to leave no doubt".

Yes, we have market forcing 192, 384 and even the 768 khz DACs, but the truth is also that market is mostly uneducated about the real effects of such high bitrates (as it is for the rest of the technology) and manufacturers obviously don't seem to think it is to their best interest to educate the market as it is far easier to catch fishes in muddy water than to make things clear. Unfortunately, it seems that community of the professionals is also not doing a very good job to educate the market as even a few of us who are willing to search for the truth (and as Mulder said we all know it is out there) we are having hard time to find it, as the tests are either not freely available or are outdated, and some of them even doesn't seem to be performed in a professional manner. And that is why we will continue to have market pushing for insane bitrates while still believing that not only the difference between modern DACs but also swapping the power cord can be heard.
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,374
Likes
234,475
Location
Seattle Area

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
4,600
Likes
3,065
Location
Zg, Cro
The problem with producing high-res music is that what is captured there is frequently noise and garbage. See my analysis in these three samples:

Why so? It's Saturday night so I can't go through your texts so I'll ask you instead: recorded analog signal is not of good enough quality or HiRes AD converters are not up to the job?

If I have to bet i would put my money on the former..

But even if that is the case I don't really think upsampling is a credible answer to the HiRes challenge to produce material of a proper quality.
 

Frank Dernie

Master Contributor
Forum Donor
Joined
Mar 24, 2016
Messages
6,445
Likes
15,781
Location
Oxfordshire
You mean that the audible difference may in fact be due to the downsampling method, and that better downsampling would work better and not be audible? It's a possibility, of course.

EDIT: but if they use standard downsampling and this downsampling is audible, then their results would still have significance for the real world. If they use some exotic downsampling on the other hand, not so much.
If I remember correctly Werner posted some samples of files which had been re-sampled using more than one different re-sampling programme on another site some years ago.
I could hear the difference between them, though it was slight.
This is rare for me. a 96/24 file downsampled to 44/16 then upsampled back to 96/24 using the software used by the BBC sounded the same to me, despite any information the original had above 44/16 was discarded.
I think both re-sampling software and the way a DAC deals with different file types can influence whether files sound the same or not.
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,374
Likes
234,475
Location
Seattle Area
Because there are a lot of ultrasonic noise sources. Since they are not heard, they get recorded and pass through final distribution of content to us.

In addition, there is no analysis done to see if higher sampling rate was needed for the content they wanted to encode. So what gets captured is just noise.
 

restorer-john

Grand Contributor
Joined
Mar 1, 2018
Messages
12,580
Likes
38,283
Location
Gold Coast, Queensland, Australia
Out of audible bandwidth noise is undesirable. I don't want it in my source material, my replay device or passed through preamplifiers, amplifiers and tweeters.

Stability, oscillation and intermodulation spuriae, increased noise etc, are all by-products of ultra-wideband digital encoding.
 

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
4,600
Likes
3,065
Location
Zg, Cro
Because there are a lot of ultrasonic noise sources. Since they are not heard, they get recorded and pass through final distribution of content to us.

Doesn't the low pas filter stage after DA chip cut those?
 

Theo

Active Member
Joined
Mar 31, 2018
Messages
288
Likes
180
the assumption that we detect such frequencies in an unconscious manner.
What would be the human part that would sense the ultrasonic frequencies for our unconscious brain to "hear"? Could someone point me to scientific papers which show that these frequencies can reach the brain?
 

oivavoi

Major Contributor
Forum Donor
Joined
Jan 12, 2017
Messages
1,721
Likes
1,934
Location
Oslo, Norway
What would be the human part that would sense the ultrasonic frequencies for our unconscious brain to "hear"? Could someone point me to scientific papers which show that these frequencies can reach the brain?

There are some studies which claim that this is the case. Here's a recent one: https://www.autonomicneuroscience.com/article/S1566-0702(17)30243-6/abstract

I would like to see proper replications from other research groups before I buy fully into it, though. Most of these findings seem to originate from the same lab. There could be confounding variables in the experiment. (do the ultra-sonic frequencies intermodulate lower frequencies, for example? Etc)
 
Last edited:

Wombat

Master Contributor
Joined
Nov 5, 2017
Messages
6,722
Likes
6,459
Location
Australia
There are some studies which claim that this is the case. Here's a recent one: https://www.autonomicneuroscience.com/article/S1566-0702(17)30243-6/abstract

I would like to see proper replications from other research groups before I buy fully into it, though. There could be confounding variables in the experiment. (do the ultra-sonic frequencies intermodulate lower frequencies, for example? Etc)

Ultrasonic frequency modulation of other frequencies can be easily measured. No confounding there. o_O
 

oivavoi

Major Contributor
Forum Donor
Joined
Jan 12, 2017
Messages
1,721
Likes
1,934
Location
Oslo, Norway
Ultrasonic frequency modulation of other frequencies can be easily measured. No confounding there. o_O

Yeah, but the question is whether they've measured/checked for that or not. Haven't read the paper yet; only read the abstract, so I don't know. "Confounding variable" is scientific jargon for the possibility that there are other things that explain the outcome, not just the variable they were interested in (i.e. the perceptability of ultrasonic frequencies).
 
Last edited:
Top Bottom