or Connect
Head-Fi.org › Forums › Equipment Forums › Sound Science › Jitter Correlation to Audibility
New Posts  All Forums:Forum Nav:

Jitter Correlation to Audibility - Page 2

post #16 of 361
Originally Posted by Theta Alpha 1 View Post

Sorry, I thought it was 0.76ns, noted.


Yes, it's for one receiver, but the flaw seems to relate to something within HDMI itself and the Yamaha costs $1,899 USD with 7.6ns jitter, so I'm inclined to think cheaper HDMI audio has weak SNR and bit depth due to jitter, so much for DVD-Audio then?


To be honest, I'm kind of curious about the test, procedure, and spectrum for the other devices. It's pretty unusual to get spikes that far out, from what I've seen.


How is the timing usually reported for jitter, anyway? Max difference of one cycle and another? Average difference between average time and min, between average and max? Some kind of rms value? Anyway, with a 10 kHz signal, a cycle takes 100,000 ns. Even if 10 ns were to indicate that some cycles take 100,010 ns and others take 999,990 ns, you'd expect to see spikes or spreading at 9990 - 10010 Hz, right? What are those spikes doing at 7 kHz and 13 kHz then? That's a bunch of orders of magnitude away. What's causing it to only happen there and that strongly? Does it have something to do with the HDMI receiving itself? Even if it does, why is the number reported in the single digits ns?


Maybe I should think about this after going to bed.



By the way, those are spectrum from J-test: 11025 Hz single tone (quarter of 44.1 kHz, tested at that sampling rate) with LSB toggled. It's supposed to be a worst-case scenario. All that spreading next to the 11025 Hz tone and all those spikes out from the noise floor are caused by the jitter. Well, you can see the noise floor in those graphs, but those are jitter tests, so it's weird to describe those as DAC noise floor measurements. The general noise floor is from things other than jitter of course.

Edited by mikeaj - 6/19/13 at 10:16pm
post #17 of 361
Originally Posted by mikeaj View Post

Those figures are well known (and often cited by some), but keep in mind they're theoretical and based on conservative calculations for what should be masked and inaudible based on other research on masking effects.


It looks like those levels of jitter are what would be needed for the jitter products to be under the noise floor, and they ignore masking and other limitations of human hearing. Otherwise, why would 20-bit audio need 15 times (= ~4 bits) less jitter, when under realistic conditions (not extremely high volume, etc.) people cannot even tell it apart from 16-bit ? I guess by the same logic 24-bit samples would further reduce the "audibility" threshold to 0.5 ps ?

post #18 of 361

Originally Posted by robertsong View Post

For instance, Empirical Audio uses two oscillators that are both specified at 2psec RMS jitter. The two oscillators sound radically different to me when used in a re-clocker in a resolving audio system.


I guess this was not a blind test ?

post #19 of 361

Originally Posted by mikeaj View Post



spectrum images (Click to show)

Xonar Essence STX sound card, output signal -6 dBFS


Note that this was a 16-bit test, so many of the peaks in the spectrum are actually from the JTest signal itself (due to the low frequency square wave used to toggle the least significant bits). For comparison, the 16-bit JTest spectrum can be seen on this graph in yellow.


By the way, at the first link in my signature there are some files for blind testing (for example, with the foobar2000 ABX plugin), that include audio recorded from cheap sound card DACs. Try these on a "resolving system" vs. the original sample to find out if you can hear the jitter, or any other artifacts. That is, of course, assuming that the A/D conversion with a separate device did not somehow remove the jitter normal_smile%20.gif

Edited by stv014 - 6/20/13 at 2:26am
post #20 of 361

Originally Posted by Theta Alpha 1 View Post

"This means the Yamaha creates reduces your signal to noise ratio from 96db for 16-bit audio to 80db or 13 bits of resolution. You decide if you want to pay to get 16 bits of quality or 13."


That is not necessarily true. Jitter level likely decreases with lower signal level (so it is more like 0.01% non-harmonic distortion for -80 dB total jitter products), while noise remains more or less the same. That is why noise can be heard in quiet parts of the music, when it is not masked by a much louder signal.


By the way, 13-bit (dithered) resolution is actually not as bad as some people may think. You can test what it sounds like compared to a 24-bit original sample here: it is possible to hear a difference, but it is likely smaller than you would expect.

post #21 of 361

And not a word about the very low frequency jitter of a vinyl LP turntable that most audiophiles seem to love.

post #22 of 361

I'm certainly no expert or anything, but when I see the audio "authority" quoted is involved with a company selling a USB filter for $200, I question motives immediately.  But maybe I'm just paranoid.

post #23 of 361

lots of phono playback geometric errors can be interpreted as "jitter" equivalent, not just low frequency wow and flutter: http://www.edsstuff.org/photos/phonograph_tracking_distortion.pdf

post #24 of 361
Originally Posted by robertsong View Post

From Steve Nugent (a leading authority of digital audio). It's a little dated, but there's some fantastic info here. I "bolded" what I found to be the most significant parts.



I don't see any meaningful significance at all. It's just the same old hand-waving we've been getting for decades now. "I believe," "IMO," etc. It all boils down to "There's a problem! No one has demonstrated that it's a meaningful problem (i.e. actual audibility). But we have the cure! And it can be yours for $$$$."





post #25 of 361
Exactly. Jitter is a convenient rabbit hole that non-ethical audio experts can send folks down. Every time I've seen the issue raised it goes round and round with numbers and complications until someone tries to apply the numbers and types of noise to audibility, and then the whole thing becomes moot.

Jitter is a hoodoo.
post #26 of 361

So we have theoretical tests with special test signals and listening tests with music. The latter show that jitter isn't audible unless you go over several thousand ps.


The "high" jitter in HDMI receivers also was fixed quite some time ago and some can achieve incredibly low values of just a few tens of ps.



=> No point in spending more money to get less jitter.

post #27 of 361
Originally Posted by xnor View Post
=> No point in spending more money to get less jitter.


Yeah. I get less jitter with my decaf mocha, but it costs the same as the caffeinated mocha. biggrin.gif



post #28 of 361


post #29 of 361

But... But... But...


Without (playback) jitter, what else would people worry about needlessly? What would the audiophile manufacturers claim that their product does better than everybody else's?

post #30 of 361

You have to distinguish between manufacturers that really want to push the limits of technology, which I can understand if somebody wants to pay more for, and manufacturers which design their stuff by ear with unscientific tweaks and magic.

The latter might produce a DAC which includes a super duper extremely low jitter high frequency oscillator but if they didn't do proper measurements the implementation is usually subpar resulting in not much better performance. Sometimes such devices perform even worse than consumer electronics. It's like putting a 10-speed gearbox into a 1950's VW Beetle.

New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Sound Science
Head-Fi.org › Forums › Equipment Forums › Sound Science › Jitter Correlation to Audibility