The 24-Bit Delusion

jkeny

Industry Expert, Member Sponsor
Feb 9, 2012
3,374
42
383
Ireland
John, your arrogant attitude and obfuscation & deflection tactics annoy me greatly. I am really sick of this. Why cannot you answer Amir's simple and direct question in a direct way? I would also like to know.

Too bad, Al - I believe you don't understand what deflection means - Amir as he was asked questions by more than me a number of days ago & didn't answer - so now he comes up with his counter-question tactic in the last number of posts - give me a break, please - typical deflection tactics of his.

Amir presents himself as the oracle of measurements here & elsewhere & yet shows fundamental misunderstandings with reading these measurements - I have already stated this a number of times - go read my posts!!
 

Al M.

VIP/Donor
Sep 10, 2013
8,689
4,478
963
Greater Boston
Too bad, Al - I believe you don't understand what deflection means - Amir as he was asked questions by more than me a number of days ago & didn't answer - so now he comes up with his counter-question tactic in the last number of posts - give me a break, please - typical deflection tactics of his.

Amir presents himself as the oracle of measurements here & elsewhere & yet shows fundamental misunderstandings with reading these measurements - I have already stated this a number of times - go read my posts!!

So why the hell cannot you directly answer the question.

Amir said: "The actual noise floor is not -140+ db. Oversampling is used in the measurements resulting in much lower measured noise floor."

What is wrong with that?

Answer the question, please. Don't p*** me off. Don't get arrogant on me, John.
 

jkeny

Industry Expert, Member Sponsor
Feb 9, 2012
3,374
42
383
Ireland
So why the hell cannot you directly answer the question.

Amir said: "The actual noise floor is not -140+ db. Oversampling is used in the measurements resulting in much lower measured noise floor."

What is wrong with that?

Answer the question, please. Don't p*** me off. Don't get arrogant on me, John.
Oh dear, Al - you need to get a grip!
 

Al M.

VIP/Donor
Sep 10, 2013
8,689
4,478
963
Greater Boston
Oh dear, Al - you need to get a grip!

Another arrogant reply instead of an answer.

As long as you do not answer, I have to assume that Amir's statement is correct. If you do not like this outcome, then you should provide an answer explaining why the statement is incorrect in your view.
 

jkeny

Industry Expert, Member Sponsor
Feb 9, 2012
3,374
42
383
Ireland
Not arrogant, Al - just some advice to calm down although you do seem to have drunk deeply of Amir's koolaid.
Al, if you don't understand what is so fundamentally wrong with Amir's statement you would do well to read this http://www.imeko.org/publications/iwadc-2007/IMEKO-IWADC-2007-F087.pdf but if you prefer to drink deeply of the koolaid then .......
DFT noise floor.jpg

I wait for Amir to fulfil his stated promise
 
Last edited:

Al M.

VIP/Donor
Sep 10, 2013
8,689
4,478
963
Greater Boston
Not arrogant, Al - just some advice to calm down although you do seem to have drunk deeply of Amir's koolaid.

No, John, I have not drunk Amir's Koolaid. I just want a simple answer, preferably one that I as a non-technical guy can understand. Instead you gave me arrogant responses.

You could do well to read this http://www.imeko.org/publications/iwadc-2007/IMEKO-IWADC-2007-F087.pdf
View attachment 31788

I wait for Amir to fulfil his stated promise

Now that is better -- perhaps. Is this an answer? If so, can you please decipher for me? Oh yes, you give the link to a highly technical paper that only a digital engineer can understand.

Why can't you give a simple answer? Or is it that, if I can't understand this, I am too hopelessly deep below your superior technical knowledge, unworthy of an ordinary answer?

If you have this attitude with customers of yours, why would anyone want to buy your products, John?
 

DaveC

Industry Expert
Nov 16, 2014
3,899
2,141
495
Also form Newton: "Where I find digital to fall down is the long term listening fatigue......"

Provided that the digital playback system in its totality ( speakers, amplification, etc.) is of high quality, and the production process commensurate, there should be no such perceived symptoms. The removal of preconception and dialled-in expectations will also help. I experienced no digital fatigue at all emanating from the T+A PDP 300HV and from other manufacturers, in my long quest to finalise my digital playback dilemma ( my extended comments are on the Aqua Formula thread ). Furthermore, since I NEVER play badly produced music ( the car and the portable suffice for this), I have no need to manipulate the original sources, digital or analogue. I do not "equalise" something which I perceive as not being "unequal"

However, if the edginess, aggression and agitation are intended and integral elements inherent in the music itself, they MUST be reproduced, regardless of being digital or analogue and regardless of causing fatigue or not! An apt example is Edgar Varese's "Arcana, Integrales, Ionisation" with Zubin Mehta (Decca), played at appropriate SPLs ( accurately C-weighted peaks of around 105 dB at listening chair) on a linear, distortion-free system. This is all, of course, on vinyl.

By the way DaveC, tour D4 cables have finally found their ideal partner - the T+A PDP 3000 HV, which I am receiving in a few days.

Cheers to all, Kostas.

Good post, I agree... perhaps digital does not smooth over imperfections in the source material, as well as the fact digital had, and sometimes still has some artifacts that may cause fatigue although this seems like it's less frequent with today's more advanced digital sources.

Congrats on the T+A player/DAC, it looks like a really nice piece of gear and very happy the D4s will be moving it's signal to your preamp! :)
 

jkeny

Industry Expert, Member Sponsor
Feb 9, 2012
3,374
42
383
Ireland
No, John, I have not drunk Amir's Koolaid. I just want a simple answer, preferably one that I as a non-technical guy can understand. Instead you gave me arrogant responses.



Now that is better -- perhaps. Is this an answer? If so, can you please decipher for me? Oh yes, you give the link to a highly technical paper that only a digital engineer can understand.

Why can't you give a simple answer? Or is it that, if I can't understand this, I am too hopelessly deep below your superior technical knowledge, unworthy of an ordinary answer?

If you have this attitude with customers of yours, why would anyone buy your products, John?
Sorry, Al from the pissed off attitude in your posts to me, I thought you were knowledgeable in this area & so I posted a link for you.

The thing is, Al, why be pissed at me when it's Amir that is spreading misinformation & refusing to answer. His ruse of pushing back a question to me is typical of his m.o. & one that most here are familiar with - I'm surprised that you are not. So, when you have no technical knowledge in the area why are you attacking me & falling for Amir's tactics?


It's simple, Al the FFT (or DFT) noise floor is determined by the DFT algorithm used in FFT(particularly the number of bins used for the analysis & hence the bandwidth of each bin but other things like windowing, etc) & not on oversampling - for instance, changing the number of bins, changes the "noise floor" seen in the FFT (nothing to do with oversampling) - everybody who deals with measurements of digital audio knows this about FFTs as it is fundamental to understanding the measurement. It's not the actual noise floor that you see in these FFT plots, it's the result of the FFT processing i.e process gain

When someone portrays themselves as a measurement expert makes such a fundamental mistake & spreads such misinformation, I call it out

Similarly when Amir spreads such fundamental misunderstanding & misinformation as "That a degraded USB signal causes the receiver to ask for retransmission of USB data. Therefore he thought by strengthening the USB signal through a Hub chip, would help" When he tries to claim that John Swenson (a well known & highly respected digital engineer) would state such misinformation as a cover-up for his own lack of such basic & fundamental knowledge, I call it out. http://www.whatsbestforum.com/showt...-microcomputer&p=394601&viewfull=1#post394601

Just to explain this - asynchronous USB (the protocol used in audio) does NOT retransmit the data - it is NOT part of the protocol & can't be done. A complete fabrication showing a severe lack of understanding of USB audio transmission & what's worse is trying to pin this misunderstanding on Swenson, as something that he said to Amir & Amir was just repeating it - jeez!!
 
Last edited:

jkeny

Industry Expert, Member Sponsor
Feb 9, 2012
3,374
42
383
Ireland
Good post, I agree... perhaps digital does not smooth over imperfections in the source material, as well as the fact digital had, and sometimes still has some artifacts that may cause fatigue although this seems like it's less frequent with today's more advanced digital sources.

Congrats on the T+A player/DAC, it looks like a really nice piece of gear and very happy the D4s will be moving it's signal to your preamp! :)

I think I read an explanation on WBF but maybe not on this thread - my experience is that digital seems to be more affected by varying ground noise, I surmise, mainly because this noise is likely to be signal correlated due to the current draws on the PS at high frequencies when digital signals are being processed by the digital chips.

Psychoacoustically, we seem to be more tolerant of fixed noise compared to variable, signal correlated noise - in my experience we seems to notice the effects of signal correlated noise, not the noise itself (fixed noise we phychoacoustcally separate out into a different auditory stream which makes it easier to ignore).

So, maybe 24bits change the granularity of the processing cycle i.e going from 0v to 1V in many more steps than 16bits - hence changing the pattern of current draw on the PS - smaller but more frequent current draws to map the same analogue output signal thus changing the resultant PS noise spectrum - more like a fixed noise level. Of course the answer is a highly stable PS, low noise immune to ground noise fluctuations due to current demands across a wide range of high frequencies.

BTW, typical FFTs won't show on their plots such dynamically changing, signal correlated noise.
 

Al M.

VIP/Donor
Sep 10, 2013
8,689
4,478
963
Greater Boston
It's simple, Al the FFT (or DFT) noise floor is determined by the DFT algorithm used in FFT(particularly the number of bins used for the analysis & hence the bandwidth of each bin but other things like windowing, etc) & not on oversampling - for instance, changing the number of bins, changes the "noise floor" seen in the FFT (nothing to do with oversampling) - everybody who deals with measurements of digital audio knows this about FFTs as it is fundamental to understanding the measurement. It's not the actual noise floor that you see in these FFT plots, it's the result of the FFT processing i.e process gain

Thanks, John, for the explanation. So if I understand correctly, the graph posted in #8 of this thread is an FFT (fast Fourier transform) plot, and the noise floor in the plot is not the actual one. Then:

1) what is the actual noise floor?
2) what is the use of such a graph if it doesn't give you direct read-out of the noise floor and thus the fold-bit resolution?

The next figure in the Stereophile link, figure 12, seems to give more information about the real resolution, do I read that correctly?

The text reads: "With a dithered 1kHz tone at –90dBFS, increasing the bit depth from 16 (fig.12, cyan and magenta traces) to 24 (blue and red) dropped the noise floor by 24dB, indicating that the Vivaldi DAC has at least 20-bit resolution, which is the state of the art."
 

jkeny

Industry Expert, Member Sponsor
Feb 9, 2012
3,374
42
383
Ireland
Thanks, John, for the explanation. So if I understand correctly, the graph posted in #8 of this thread is an FFT (fast Fourier transform) plot, and the noise floor in the plot is not the actual one. Then:

1) what is the actual noise floor?
If you look at Opus post he calcs the actual noise floor based on a 1Hz FFT bin width - comes out at -101dB. This is based on Amir's statement that the bin width is "less than 1Hz" which we are still waiting to hear from him on - where he got this 1Hz bin width from? His <1Hz statement would make the actual noise floor worse than 17bit resolution ( -102dB)
2) what is the use of such a graph if it doesn't give you direct read-out of the noise floor and thus the fold-bit resolution?
FFTs accurately show the amplitude level of narrow band signals because their energy is confined to a small number of FFT bins - the problem is with wideband signals (like noise) whose energy stretches across many bins (in the case of noise floor, is in all bins).

The next figure in the Stereophile link, figure 12, seems to give more information about the real resolution, do I read that correctly?

The text reads: "With a dithered 1kHz tone at –90dBFS, increasing the bit depth from 16 (fig.12, cyan and magenta traces) to 24 (blue and red) dropped the noise floor by 24dB, indicating that the Vivaldi DAC has at least 20-bit resolution, which is the state of the art."
Yes, comparisons between FFT noise floors can be made once the exact same FFT algorithm is used for both plots (which is the case here, I presume) - so the FFT "noise floor" is -130dB for 16bit measurement & -154dB for 24bit measurement - a difference of 24dB or 4bits (approx 6dB per bit) which is where he gets the 20bit resolution from.

Note Atkinson never states that the "noise floor" is -130dB & goes down to -154dB, just that it has dropped by 24dB - with different number of bins & therefore bin widths, the absolute measures would have been different but the relative difference would still have been 24dB.

It would serve you well to not blindly accept Amir's postings here as they are often technical half-truths & agenda driven. He has been exposed in this agenda a long time ago.
 
Last edited:

jkeny

Industry Expert, Member Sponsor
Feb 9, 2012
3,374
42
383
Ireland
Now Al - why don't you show to Amir the same ire that you came on to me with in your posts?
After all he has not answered any of the questions he was asked, despite numerous requests for explanations.

BTW, Al, my customers buy my products because of their sound, which is the result of my experimentation/understanding of where the weaknesses are in digital audio & my continual striving towards knowledge & improvements in this area same as the Schiit people you quoted.
 

Al M.

VIP/Donor
Sep 10, 2013
8,689
4,478
963
Greater Boston
If you look at Opus post he calcs the actual noise floor based on a 1Hz FFT bin width - comes out at -101dB. This is based on Amir's statement that the bin width is "less than 1Hz" which we are still waiting to hear from him on - where he got this 1Hz bin width from? His <1Hz statement would make the actual noise floor worse than 17bit resolution ( -102dB)
FFTs accurately show the amplitude level of narrow band signals because their energy is confined to a small number of FFT bins - the problem is with wideband signals (like noise) whose energy stretches across many bins (in the case of noise floor, is in all bins).


Yes, comparisons between FFT noise floors can be made once the exact same FFT algorithm is used for both plots (which is the case here, I presume) - so the FFT "noise floor" is -130dB for 16bit measurement & -154dB for 24bit measurement - a difference of 24dB or 4bits (approx 6dB per bit) which is where he gets the 20bit resolution from.

Note Atkinson never states that the "noise floor" is -130dB & goes down to -154dB, just that it has dropped by 24dB - with different number of bins & therefore bin widths, the absolute measures would have been different but the relative difference would still have been 24dB.

Thank you, John, for the explanations! Now I understand much better.

It would serve you well to not blindly accept Amir's postings here as they are often technical half-truths & agenda driven. He has been exposed in this agenda a long time ago.

Obviously, since you now provided adequate explanations, you also exposed Amir's mistake. I just wanted to get an answer in order to further my understanding; I did not operate under the assumption that Amir had to be right.
 

Al M.

VIP/Donor
Sep 10, 2013
8,689
4,478
963
Greater Boston
Now Al - why don't you show to Amir the same ire that you came on to me with in your posts?
After all he has not answered any of the questions he was asked, despite numerous requests for explanations.

John, if from the start you would have provided an intelligible answer to a simple question, as you finally did, all this could easily have been avoided. Your initial apparent refusal to do so was what drew my ire. We may have misread one another. If that is the case, I apologize.

BTW, Al, my customers buy my products because of their sound, which is the result of my experimentation/understanding of where the weaknesses are in digital audio & my continual striving towards knowledge & improvements in this area same as the Schiit people you quoted.

I understand that. However, I perceived in this thread an arrogant attitude. I may have been mistaken.
 

jkeny

Industry Expert, Member Sponsor
Feb 9, 2012
3,374
42
383
Ireland
Thank you, John, for the explanations! Now I understand much better.



Obviously, since you now provided adequate explanations, you also exposed Amir's mistake. I just wanted to get an answer, I did not operate under the assumption that Amir had to be right.
Well it certainly appeared like that - you thanked & praised him for his post "Thanks, Amir, informative technical comments as always!"
And came onto me like a rabid dog - I have to say that it is a strange reaction for someone who really didn't understand Opus & my questions to Amir or indeed Amir's question of me (his usual tactic) which you bullishly got pissy with me for not answering.

It might be best to stay out of technical discussions that you have admitted you don't understand?

Based on your lack of technical knowledge, accusing me of arrogance is pretty off, don't you think? I also don't appreciate your style in bringing my business & customers into this discussion.
None of this have you apologised for - I see you have now - accepted!
 

jkeny

Industry Expert, Member Sponsor
Feb 9, 2012
3,374
42
383
Ireland
John, if from the start you would have provided an intelligible answer to a simple question, as you finally did, all this could easily have been avoided.
No, this would have been avoided if you didn't fall for Amir's tactic & asked him to provide an answer to the questions he was asked, instead of posing a question to me. You got pissed at me when you should have looked for an answer from him & all this would have been clarified. Instead you waded in out of your depth & then got ratty because I wasn't throwing you a lifeline.
Your initial apparent refusal to do so was what drew my ire.
This was your mistake as you don't understand the technical area.
We may have misread one another. If that is the case, I apologize.

I understand that. However, I perceived in this thread an arrogant attitude. I may have been mistaken.
OK
 

Al M.

VIP/Donor
Sep 10, 2013
8,689
4,478
963
Greater Boston
Well it certainly appeared like that - you thanked & praised him for his post "Thanks, Amir, informative technical comments as always!"

Because that is what I perceived at the time. At least he gave an answer that, even though it seems to be wrong, I could understand.

And came onto me like a rabid dog - I have to say that it is a strange reaction for someone who really didn't understand Opus & my questions to Amir or indeed Amir's question of me (his usual tactic) which you bullishly got pissy with me about not answering.

It might be best to stay out of technical discussions that you have admitted you don't understand?

Why? This is a forum for everyone, not just for technical experts. If you explain something you should do so in terms that are intelligible for a wider audience. Otherwise you should post on the measurement-based forum. And when Amir asked a simple question what was wrong with his statement, a question that was also of great interest to me, your refusal to answer it directly at that time was not appropriate -- even if, as you claim, you had already answered it.

Based on your lack of technical knowledge, accusing me of arrogance is pretty off, don't you think?

I already said that I may have been mistaken. And the appearance of arrogance in your posts had nothing to do with my lack of technical knowledge, you did this to yourself.

I also don't appreciate your style in bringing my business & customers into this discussion.
None of this have you apologised for - I see you have now - accepted!

Thanks.
 

Al M.

VIP/Donor
Sep 10, 2013
8,689
4,478
963
Greater Boston
No, this would have been avoided if you didn't fall for Amir's tactic & asked him to provide an answer to the questions he was asked, instead of posing a question to me. You got pissed at me when you should have looked for an answer from him & all this would have been clarified. Instead you waded in out of your depth & then got ratty because I wasn't throwing you a lifeline.

This was your mistake as you don't understand the technical area.

Forget it, John. You keep pretending as if all this was just my fault, and you always keep pointing the finger towards others, in this case me and Amir. If you cannot accept the fact that also you played a role in the unfortunate development of this discussion then this is too bad, and indeed revealing with respect to attitude. Have a good day.
 

jkeny

Industry Expert, Member Sponsor
Feb 9, 2012
3,374
42
383
Ireland
Because that is what I perceived at the time. At least he gave an answer that, even though it seems to be wrong, I could understand.



Why? This is a forum for everyone, not just for technical experts. If you explain something you should do so in terms that are intelligible for a wider audience.
It was a question to Amir, not you or the general readership. Ignore it if you don't have the technical chops.
Otherwise you should post on the measurement-based forum.
Amir posted technical info - why don't you tell him he should have posted it only on the measurements area as you didn't understand it (although you thought you did)? He was erroneous - should I or Opus not correct him? Opus asked him the question which Amir ignored - I simply repeated asking the question
And when Amir asked a simple question what was wrong with his statement, a question that was also of great interest to me, your refusal to answer it directly at that time was not appropriate -- even if, as you claim, you had already answered it.
I had already answered it for anybody who knows the technical area so he was simply playing you & others as fools who would fall for his questions of me - as I said his usual tactic


I already said that I may have been mistaken. And the appearance of arrogance in your posts had nothing to do with my lack of technical knowledge, you did this to yourself.

Thanks.
Sure it did - I spoke to you as someone who seemed to understand the technical arguments - you never said until well into the exchange that you didn;t have th etechnical knowledge. So my answers appeared to you to be arrogant but it was your failure to lay out your technical level that was the problem.

You may still blame me for my answers & your lack of technical understanding - there's nothing I can do about that but my advice still stands - best to stay out of technical discussions & don;t be so easily fooled by Amir's tactics
 

jkeny

Industry Expert, Member Sponsor
Feb 9, 2012
3,374
42
383
Ireland
Forget it, John. You keep pretending as if all this was just my fault, and you always keep pointing the finger towards others, in this case me and Amir. If you cannot accept the fact that also you played a role in the unfortunate development of this discussion then this is too bad, and indeed revealing with respect to attitude. Have a good day.

You still drink the koolaid & fail to see his tactics - that's a shame!
We look forward to his answers as I'm sure you do!
 

About us

  • What’s Best Forum is THE forum for high end audio, product reviews, advice and sharing experiences on the best of everything else. This is THE place where audiophiles and audio companies discuss vintage, contemporary and new audio products, music servers, music streamers, computer audio, digital-to-analog converters, turntables, phono stages, cartridges, reel-to-reel tape machines, speakers, headphones and tube and solid-state amplification. Founded in 2010 What’s Best Forum invites intelligent and courteous people of all interests and backgrounds to describe and discuss the best of everything. From beginners to life-long hobbyists to industry professionals, we enjoy learning about new things and meeting new people, and participating in spirited debates.

Quick Navigation

User Menu

Steve Williams
Site Founder | Site Owner | Administrator
Ron Resnick
Site Co-Owner | Administrator
Julian (The Fixer)
Website Build | Marketing Managersing