[1] Measure the jitter of two DACs. They will be identical in every way. Put them side-by-side, they cannot be distinguished apart. Same volume level.
[1a] You point to the one that you are convinced sounds better by quite some margin. It turns out that it is the one that was earlier measured as having higher jitter. What is going on?
[1b] It is the nature of the jitter that has to be measured. Jitter is frequency related, jitter is in fact noise and it has a frequency response. So it matters WHAT frequency that the jitter shows up. At certain frequencies, jitter matters more than showing up at other frequencies.
[1c] Jitter close to the fundamental (low frequency) is very difficult to measure and it has been shown to be the worst form of jitter.
[2] But what happens when we multiply the Blue area by 100? This and it is scary: Look at how below 50 Hertz the rise in random noise means an increase in very low-frequency jitter.
[2a] It goes from a decent -130dB to worse than -90dB and if further magnification it would likely be worse than that.
[3] This LF jitter is what causes the kind of "digital sound" that many have complained about.
1. Good.
1a. What's going on is that firstly, if one has a higher measured jitter, then obviously they are NOT "identical in every way". What other way are they not identical? Secondly, even If we assume they are identical in every way except jitter, then the jitter artefacts have to be at least equal to, or above, the threshold of audibility otherwise "what is going on" MUST be some cognitive bias/perception error, there is NO other rational alternative!
1b. That depends on how you define "matters". "Matters" in terms of a measurement or "matters" in terms of an audible difference?
1c. That depends on the frequency of the fundamental and what you mean by "low frequency". If the fundamental is at say 3kHz then that would be potentially the "worst form of jitter [noise]" as that's where human hearing is most sensitive. If we're talking about frequencies below about 500Hz though, then that would potentially be the best form of jitter noise, as human hearing is progressively less sensitive to low frequencies. I say "potentially" because obviously it doesn't "matter", if it's below audibility.
2. There's obviously two huge problems with your assertion:
A. Firstly, that's a hypothetical question. What happens if we multiply a wasp's size by 100 times, is that scary? Sure, in our imagination but of course there's no need to be scared in reality because no such wasp exists. We might as well say "what happens when we multiply the blue area by" a few trillion, that would be particular scary as it would probably kill you!
B. Secondly, even if it were not a hypothetical question and there were DACs which multiplied the blue area by 100 times, it still wouldn't make any difference! If we take jitter noise (or any other noise) that is inaudible at 3kHz, increase it's amplitude by 100 times and change it's frequency to "below 50Hz", then it is still INAUDIBLE! Because, human hearing is about "
100 times" less sensitive to frequencies below 50Hz. Science has known this for over 85 years (Fletcher-Munson) and it's been confirmed by countless hearing threshold tests ever since! In other words, even if there were DACs that produced 100 times more jitter noise below 50Hz (at -90dB), that would still be way below audibility!
2a. How is -130dB "decent"? It's not "decent", it "preposterous"! Given any reasonable listening volume, -130dB noise cannot even be reproduced by transducers and so the question of audibility is obviously mute!
3. No it's not, you just made that up! The "digital sound that many have complained about" is the lack of analogue distortion and in some cases, also a deliberate choice made during mixing and/or mastering. This has been thoroughly well established for many decades, even when digital audio was first released to the public!
As there are cheap DACs on the market ($100 and lower) which have jitter artefacts throughout the spectrum which peak no higher than -120dB, then "What is going on" appears completely obvious: The old audiophile marketing ploy of taking some inaudible "problem", lying about it's magnitude and/or audibility and then offering some expensive audiophile solution which (falsely) claims an audible improvement!
[1] So what interests me? Doing my job!
[2] BTW, I know about the Stereo Review magazine. That was 32 years ago and if you are doing the same as they are, then I am not surprised. We have thankfully come a long way since then.
1. Which is to sell an expensive audiophile solution which claims an audible improvement!!
2. Agreed, we've come an awful "long way since then", in terms of the audiophile marketing of snake oil, although I'm not in the least be "thankful" for that!
[1] A large body of experiential evidence along the lines of dac a sounds different to dac b . Counterpoint a large body of empirical evidence demonstrating difference outside of audible limits
[2] "When you have eliminated the impossible , whatever remains , however improbable , must be the truth" Sherlock Holmes , Arthur Conan Doyle
1. As castleofargh pointed out. "no" there is no reliable evidence, let alone a large body of reliable evidence. In fact, the reliable evidence indicates no difference. There also is no reliable evidence which demonstrates anything beyond audible limits, given the conditions of listening to commercial music recordings at a reasonable level with consumer equipment (consumer headphones/speakers).
2. Absolutely! But conversely, what if you haven't "eliminated the impossible", what if you've eliminated something that IS possible? Surely, "what remains" (as it's improbable) is unlikely to be the truth. What if you've eliminated something that isn't just possible but is likely? Surely, "what remains" is extremely unlikely to be the truth. And, what if you eliminated something that isn't just likely but is proven/demonstrated to occur almost continuously and is responsible for the very existence of music in the first place? Surely, "what remains" has no chance, and even less chance still if "what remains" isn't even applicable! You've effectively applied the quote backwards; your "what remains" should have been eliminated and what you've erroneously eliminated should be the "what remains" (and it's not even slightly improbable)!!
G
