There are a lot of misinformation out there and there are a lot you can do about it. 1) Learn to filter them out. 2) Spread the words. 3) Challenge it when you see it.
@ Jakob2, I shuffled the sequence for clarity sake.
Given the upper quote, shouldn't your statement pointed out in different color be "as explanations for possible audible differences"? Unless such sensory experiment is already done and reported.
Posters use the phrase "IMHO" or "believe" all the time. Problem is when people confuse speculation with evidence.
Please define what correct result is in this context.
Can you quote or link to such post? I want to know if you are just speculating or talking facts. Thanks in advance.
He has a lot to hide for business reason.
You missed the existing body of data, especially on DACs. DAC has been a matured technology for over 20 years. Even the cheap ones are audibly transparent and their specs are available for observation. All you need to do is compare the specs when some internet poster claims of superior sound quality from new DAC or modded DAC. The reason why you won't reveal your listening comparison method or measurement is because your claim can't withstand the scrutiny of objectivity.
When people have an audio device in front of them then there is the option to both listen and measure. When people don't know how to measure and all they have is listening experience, then what? The brain automatically constructs the best story it can from the available data. By default (which is to say, without effortful intervention of system 2), the story is believed. That is the default mode of thinking most of the time even if the available data is incomplete and or of poor quality. The links to WYHIATI should make it clear (What You Hear Is All There Is).
Not a bad point, but is the end purpose of the audio device to satisfy human listening, to sound like real music? It is going to be used as a measurement device? Some of both?
I certainly take some measurements, and I have written about investigations with scope, spectrum analyzer, shortwave receiver, signal generator, DVM, FFT, etc. Also have written about experiments involving things that are not so easily measured or that would require more expensive test equipment than my budget allows. Mostly haven't made a habit of posting pics, but the results are still what they are.
Would agree most of what I talk about is related to SQ (according to human perception) since that's what I'm more interested in. Other people are free to talk about what they prefer. Of course it could become an issue if someone wanted to claim FFT_measurements=SQ, that caps are transparent because they don't show much HD, etc., but that's simply because such things aren't exactly true, at least not IME.
I certainly take some measurements, and I have written about investigations with scope, spectrum analyzer, shortwave receiver, signal generator, DVM, FFT, etc. Also have written about experiments involving things that are not so easily measured or that would require more expensive test equipment than my budget allows. Mostly haven't made a habit of posting pics, but the results are still what they are.
Would agree most of what I talk about is related to SQ (according to human perception) since that's what I'm more interested in. Other people are free to talk about what they prefer. Of course it could become an issue if someone wanted to claim FFT_measurements=SQ, that caps are transparent because they don't show much HD, etc., but that's simply because such things aren't exactly true, at least not IME.
Would you for once stop spilling that nonsense, please. I’m sorry that you are obviously deaf, but not all members of this forum have this condition.
I do not believe in cable or 5 cm silver wire sound, boutique DAC or speaker cables, nor hear any difference from ordinary good quality ones.
However, two DACs that measure stellar and are very close, according to ASR, to my average ears sound very different without any doubt. One has distinctively clearer and stronger treble, more detailed soundstage and more details in the sound reproduction. Jet, they have distortion that differs by 0.00002% and SINAD is 121 for both. All FFT artifacts are more than 130 dB below signal for both.
So, would you look at those specs and help all us clueless by explaining the difference, as you seem to be an expert on DAC specs.
Audio device, yes. What about audio reproducing device?
I remember seeing a picture of your modded board with wires dangling all over the place but not the picture of your measurements.
Would that be from subjective listening sessions?
And still they go into hard clipping when you try to play peak sample normalized music recordings through them at 100 % volume, except for some exotic models liked by audiophiles, such as NOS DACs, the DACs of Benchmark Media and a certain dirt cheap DAC built into car radios.
in general in front of the speakers....
It's too complicated or too simple for you?
In post 277 Hierfi compared SMSL DO100 and SMSL SU-9n which both have very low distortion and SINAD at 121dB according to ASR measurements. However looking at the FFT of SMSL SU-9n I can easily believe that it may sound different to some other DACs with similar performance numbers. Noise skirts are wide and high and higher order harmonics may even be audible. IMO distortion and SINAD numbers do not tell anything by themselves.
Looks like a quite strong assertion; could you please cite some sensory experiments (i.e. listening tests) on DACs that will give evidence for your assertion?
@mocenigo,
different povs are needed to have a discussion. "Ad hominem" posts do not help in discussion on these topics
There are many types of listening tests done. Which type do you prefer to read about?
As for the definition of correct result I brought up on post #301 goes...?
How is any of that going to be audible? It's buried completely within the noise. The only reason you can see them there is because of the FFT process used to extract them from the noise itself. They are all almost 140dB below the fundamental, this is ~20dB below the noise floor of the DAC.
deleted some posts and their colateral. keep the personal cracks out of the forum.
D
Deleted member 537459
FFTs are pretty useless since they are time-invariant measurements.
TWSDAC-1862 and TWSDAC-1541 aren’t transformer coupled, they share the same I/V converter, folded cascode with zero negative feedback. Simulation gives around -85dB of THD.
At the end they sound much better than the -120dB Sabre DAC. Which confirms that the FFTs fail.
TWSDAC-1862 and TWSDAC-1541 aren’t transformer coupled, they share the same I/V converter, folded cascode with zero negative feedback. Simulation gives around -85dB of THD.
At the end they sound much better than the -120dB Sabre DAC. Which confirms that the FFTs fail.
That was not the point. The point was that from 10kHz sine on test vinyl you will measure 30kHz H3 and from 15kHz you will measure 45kHz H3, just as a result of distortion. Is such “ultrasound” info any useful?
Do you really believe in audibility of something that is all below -130dBr?? Have you tried to calculate such differences into SPL? What would be the mechanism of hearing such thing? Have you done any DBT to support your claims? Because usupported opinion does not count.
@PMA If I understand you and @billshurv correctly, you claim that anything above 20 kHz on a record is distortion. That doesn't make any sense to me, because as far as I know, there is no brick wall filter anywhere suppressing signal components above 20 kHz, assuming no digital recorders with a low sample rate were used. Of course you have roll-off, but it will be much smoother than for CD. (You may have some notches due to analogue tape recorder gap length, but those will be far above 20 kHz.)
You can't measure the response above 20 kHz with a test record unless you have one with test signals above 20 kHz.
The fact that some of the signal above 20 kHz will be distortion is true but irrelevant. The same holds below 20 kHz: from 1 kHz you get 3 kHz third harmonic distortion and from 100 Hz you get 300 Hz, so by your logic the part below 20 kHz is of no use either.
Whether a smooth ultrasonic roll-off is any better than a sharp roll-off in the first place is a different question. I expect it sounds more natural to my cat. I'm not so sure about that for humans, but considering the popularity of high sample rate recordings (1), apodizing filters and the whole MQA format and everything Hans van Maanen writes on his temporal coherence website, there are people who believe it to be an advantage for human listeners as well.
(1): High sample rates also reduce the clipping on intersample overshoots, so that could also be a reason for their popularity. That and plain old marketing.
You can't measure the response above 20 kHz with a test record unless you have one with test signals above 20 kHz.
The fact that some of the signal above 20 kHz will be distortion is true but irrelevant. The same holds below 20 kHz: from 1 kHz you get 3 kHz third harmonic distortion and from 100 Hz you get 300 Hz, so by your logic the part below 20 kHz is of no use either.
Whether a smooth ultrasonic roll-off is any better than a sharp roll-off in the first place is a different question. I expect it sounds more natural to my cat. I'm not so sure about that for humans, but considering the popularity of high sample rate recordings (1), apodizing filters and the whole MQA format and everything Hans van Maanen writes on his temporal coherence website, there are people who believe it to be an advantage for human listeners as well.
(1): High sample rates also reduce the clipping on intersample overshoots, so that could also be a reason for their popularity. That and plain old marketing.
Last edited:
- Status
- Not open for further replies.