Can jitter sound “nice”

[Frank Dernie]

Jitter is obviously bad in an objective sense but HiFi News Have given rave reviews to the Metronome Kalista CD player at £32000 and the even more expensive separate Transport/DAC combo at £72000. The jitter measured on the former with CD is 720 psec and the latter 1205 psec.
These are very high compared to everything else they have tested, with Chord products being typically 10 psec And dCS typically less than 50 psec.
Is this level possibly high enough to actually be audible, and if so is there any form it could take which may sound “nice”?
I fear it could be just the pretty styling and reassuringly high price which is influencing their expectation bias but...

 

[SIY]

Jitter measured where, exactly? Some jitter matters, some doesn't. The only jitter that matters (assuming things aren't so horrible that lock is lost) is at the clock input of the DAC chip.

Your fear is likely accurate. I think I'm safe in assuming the HiFi News didn't do any kind of level-matched DBT.

 

[DonH56]

At 20 kHz and 16-bit playback at CD rates 1 lsb of jitter is about 243 ps. 1 ns is about 14-bit resolution. It's just too low to be heard... Correlated jitter can add non-harmonic spurs that tend to be more audible, but you're still talking some 80+ dB below full-scale output.

https://www.audiosciencereview.com/...al-audio-jitter-fundamentals.1922/#post-50586
https://www.audiosciencereview.com/...igital-audio-jitter-fundamentals-part-2.1926/

 

[SIY]

Don, that assumes that you're measuring jitter at the DAC clock input. I am dubious that the reported measurements reflect that, but maybe Frank can clarify since he has the article?

 

[RayDunzl]

A little footnote from ESS on the sound of jitter:

Technical Details of the Sabre Audio DAC
Martin Mallinson and Dustin Forman, ESS Technology Technical Staff

http://www.esstech.com/files/4314/4095/4318/sabrewp.pdf -- footnote 15


"The noise that jitter induces is not easily described: it is not a harmonic distortion
but is a noise near the tone of the music that varies with the music: it is a noise that
surrounds each frequency present in the audio signal and is proportional to it.
Jitter noise is therefore subtle and will not be heard in the silence between audio
programs. Experienced listeners will perceive it as a lack of clarity in the sound
field or as a faint noise that accompanies the otherwise well defined quieter
elements of the audio program."

 

[gvl]

At these prices they should be advertising a revolutionary proprietary jitter profile specially tuned for listening pleasure based on years of experience and subjective experiments.

 

[Frank Dernie]

Jitter measured where, exactly? Some jitter matters, some doesn't. The only jitter that matters (assuming things aren't so horrible that lock is lost) is at the clock input of the DAC chip.

Your fear is likely accurate. I think I'm safe in assuming the HiFi News didn't do any kind of level-matched DBT.

I don’t know where they measure it. They quote jitter in the little measurement box of all digital kit. It is probably on Paul Miller’s website but I found it painful to navigate that and gave up years ago.
They have a standard set of measurements for every type of kit they review and, like Stereophile, the review does not necessarily praise kit that measures well and vice versa...
I have subscribed for about 50 years on and off, my preferred Hi-Fi mag.

 

[SIY]

At these prices they should be advertising a revolutionary proprietary jitter profile specially tuned for listening pleasure based on years of experience and subjective experiments.

At these prices, I should getting orally serviced by the staff member of my choice while I listen to music.

 

[gvl]

At these prices, I should getting orally serviced by the staff member of my choice while I listen to music.

I heard that's coming in the mk II version pending new hires as the current staff was unable to perform to provide the experience expected at this price level.

 

[nm4711]

A lot of people like the sound of vinyl. And there is jitter in it. The motor does not have the same speed in each moment, and when the vinyl is a little bit of center, the play speed goes a little bit up and down during each turn. So this kind of jitter (in combination with some other things) sounds nice for some people.

I like some kind "jitter" too: In electronic music repeating sounds are often perfectly even spaced, wich I dont like. Music played on instruments sounds a lot better to me, probably because the sounds are not perfectly even spaced (and they do not sound the same every time).

But in the analog domain the speed changes smoothly over a very long period (a half turn of the vinyl) compared the digital domain (from one sample to another), so I dont think you can really compare this.

 

[RayDunzl]

Ceiling fan "jitter"...

https://www.audiosciencereview.com/forum/index.php?threads/close-in-jitter.1621/page-15#post-41686

 

[sergeauckland]

I don't think that analogue jitter, or wow and flutter, as it's more commonly referred to, is ever pleasant. If audible, it gives a 'warbling or slurring' effect, especially on longer notes, that is always disturbing. Whilst harmonic distortion can add 'body' to the sound, W&F has no redeeming features.

Another unpleasant effect is noise modulation, where the noise floor rises and falls with the signal. That too is unredeemingly unpleasant, especially noticeable in things like solo flute or violin where it's not masked by other sounds.

S

 

[nm4711]

I am not a fan of it either. If so, I would probably not be a member of this fourm. But some people do like it.

 

[RayDunzl]

Another unpleasant effect is noise modulation, where the noise floor rises and falls with the signal.

We were listening to some pipe organ recently, during the otherwise very quiet parts, somebody/something was gain riding the record levels, to an extreme degree - guessing 10 to 20dB, totally annoying.

I don't know if I've ever heard noise floor modulation itself - without somebody hamfisting the record levels. Haven't noticed it, anyway.

 

[sergeauckland]

We were listening to some pipe organ recently, during the otherwise very quiet parts, somebody/something was gain riding the record levels, to an extreme degree - guessing 10 to 20dB, totally annoying.

I don't know if I've ever heard noise floor modulation itself - without somebody hamfisting the record levels. Haven't noticed it, anyway.

Analogue tape machines do it, especially pre-Dolby, but it's normally masked by everything else that's going on. Your organ recording will show it if it's an old recording, but yes, it could be manual gain riding, or a poor compressor.

S

 

[DonH56]

Don, that assumes that you're measuring jitter at the DAC clock input. I am dubious that the reported measurements reflect that, but maybe Frank can clarify since he has the article?

No, that is from the aperture time error of a data converter (ADC or DAC). If you have a perfect DAC and add that much jitter at the output, no matter the source, you will degrade SNR by an amount related to the jitter's amplitude and output signal frequency. Note that for uncorrelated jitter the clock frequency (sampling rate) rate drops out of the equations; only signal frequency and resolution matters. The charts in the other threads show SNR vs. jitter vs. signal frequency, not clock frequency.

For correlated jitter, the fundamental relationship to aperture error applies, but conversion to SNR/SINAD/SFDR does not because it is no longer a random noise source. I have not really done any threads about correlated jitter sources; it gets complicated, and I am lazy. I have to deal with a dozen or two different types of jitter in my day job and just haven't felt like spending my free time on it.

 

[SIY]

Would you see an aperture error of over a nanosecond?

 

[DonH56]

Would you see an aperture error of over a nanosecond?

I don't know what you are asking, sorry. In my day job, using a DSO worth more than my house, we "see" errors in the femtosecond range. When I was designing GHz data converters, most certainly, and 1 ns errors would get me in trouble. For audio, I doubt I could hear a 1 ns error, but maybe if it were correlated and all the stars aligned... Still, my hearing has rolled off someplace around 12 kHz last I checked, so probably not. At lower frequencies it takes increasingly larger jitter to be heard.

I do not have much recent experience. A few years ago when I was AVR shopping I noticed some had 50~100 ns of jitter specified especially on HDMI inputs. These days I see hardly any jitter measurements aside from Amir's and most everything seems to be much better and much less likely to be audible. Again I'll note that Amir's measurements show that most DACs provide better performance than most amplifiers and speakers can produce.

Going back to the original question, uncorrelated jitter just sounds like noise. Correlated jitter adds distortion-like spurs that are, or would be if high enough to be audible, much more annoying than random noise. That said, I'd guess that power supply products and nonlinear distortion terms (particularly IMD) are more likely to be audible than jitter in today's systems. @amirm and others probably have more audio-scale insight on that than I.

 

[SIY]

Well, if the magazine is reporting over a ns in jitter, my question was to confirm my understanding- that it was unrelated to the DAC jitter and likely measured somewhere upstream, since aperture errors are (to the best of my knowledge) orders of magnitude smaller. Does that seem like sound (ouch) reasoning to you?

 

[DonH56]

Sorry, what article? I was just skimming, as usual during lunch, so did not read any articles. What is upstream of a DAC is all digital to me (unless you push all the way back to the ADC). Jitter can come from a number of things, e.g. power noise, EMI/RFI, crosstalk, various noise sources in the DAC and other circuits like filters and buffers, clock feedthrough, etc. The DAC's output includes its environment so saying where and why a particular is there is beyond my ken without knowing a lot more detail.

I have not looked at specs lately. A few years ago a number of products, particularly AVRs and such vs. stand-alone DACs, exhibited jitter on the order of 100's of ps and higher. It "feels" lower these days based on what I've read but to my mind is becoming like amplifier THD

Sorry, busy today, feel like we are talking past each other... - Don