|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
98.95.190.28
A very interesting article on Jitter was in the August Pro Sound News, by Dave Hill, "A Matter of Time: The Audibility of Clock Jitter" P34-35, cont'd on page 39.
A copy of the article is at:
http://cranesong.com/A_Matter_%20Of_Time_The%20_Audibility_Of_Clock_Jitter.pdf
Files of the same music track with varying amounts of CLOCK jitter added are at:
http://cranesong.com/jitter_1.html
You are supposed to download the files, and burn a CD of them to listen on your main system, not necessarily to listen at the computer.
Even more interesting, are tracks with the DIFFERENCE files of the various tracks versus the "original", are at:
http://cranesong.com/jitter_2results.html
This page includes the "reveal" of the various A thru E tracks at the 1st page, so if you want to listen to the A thru E tracks "blind", don't go to the second page till listening to the files downloaded from the 1st page.
The second page also has an interesting graph, as well as PDF documents of the test set-up, etc.
For your interest and by way of a challenge, I was able to hear definite differences on one track versus others, and thought that I could hear differences on two of the other three tracks. I also successfully picked out what I thought was the reference track versus the "altered" tracks.
The very interesting part is that the levels of jitter involved are all well below what most naysayers tout as being audible. By at least an order of magnitude, and more.
Listen for yourself and see what you think. Remember, this is simulated master clock jitter, and does not include the contribution from your own playback system. This then raises the question of how can we possibly hear such low levels of jitter, if our playback system jitter is higher and masking the amount in the jittered file?
Jon Risch
Follow Ups:
Ive made a series of product improvements where I have measured the P-P jitter from one to the next and these were all in the range of 100-200 psec incremental changes. These measurements were real-time using a 7GHz sampling scope and jitter measurement S/W.The difference in SQ of these changes were all obvious in my system. I have yet to make an improvment in jitter that I can easily measure that I cannot hear.
One example is a USB converter with the output transformer 250psec and without 100 psec. This is averaging the gaussian curve, so there is more to the picture than one number.
In my measurements, the tails of the Gaussian curve are the P-P numbers and these may only occur every 10 or 20 seconds. I can actually watch these update as the measurement progresses. These are definitely not audible. This is why using one number to describe jitter is nonsense. One must look at the shape of the Gaussian curve to get a feel for the real audible effects. I believe this is even more important than the spectrum of the jitter.
Steve N.
Edits: 11/16/13
What type of jitter was added, random or signal-correlated, and how much?
And what are those levels that most naysayers tout as being audible?
Klaus
It would help tremendously in answering your questions, to simply go to the links I provided.
As for the jitter levels, I think that your question will answer itself once you see the levels he is talking about, and which ARE audible if you listen to his audio files!
If after reading the content of the links, you still have such questions, then I would begin to doubt that I could possibly be able to explain it to you. But I will try, if that is what you need.
Of course, if you read both links, the audio files will no longer be blind, but I guess it depends on whether or not you can invest even a few minutes of your time to check it out and listen first before going to the 2nd link.
If it helps to know ahead of time some of the details, then:
The jitter is a band of noise at center frequencies of 300 Hz, 200 Hz, and a band of noise below 10 Hz.
The amount is 300 pS (200 Hz and 300 Hz)and 800 pS (below 10 Hz band).
The reference file is supposed to have jitter levels around 2-4 pS.
The test set-up is shown at:
http://cranesong.com/JITTER_TEST_SETUP.pdf
Jon Risch
“The jitter is a band of noise at center frequencies of 300 Hz, 200 Hz, and a band of noise below 10 Hz.
The amount is 300 pS (200 Hz and 300 Hz)and 800 pS (below 10 Hz band).
The reference file is supposed to have jitter levels around 2-4 pS.”
Is that random or signal correlated jitter?
What are the lowest jitter levels "most naysayers tout as being audible", what are
the perception thresholds, and what is the research those thresholds come from?
Jitter is jitter. Discussing which type it is is irrelevant and moot. Either type you mention can be heard. Most levels are perceptible, as this example clearly shows. There is no real threshold, it's not a signal, it's a state of condition of the signal. It manifests itself as distortion of the signal.It's effects are also cumulative or additive. So system jitter will be added to the jitter on the recordings, so the relative differences in the recordings will still be perceived.
Where there is a threshold is in the resolution of the system to be able to resolve the details so they can be heard. A cheap budget system will not resolve jitter well (or music), where a more refined system will show its effects more greatly.
Edits: 10/16/13
Thank you John for the interesting link.
In my setup (Northstar CD and DAC (32bit) it was not easy to hear diffences on the Test CD - but like the subtle differences in signal cables (excellent vs. normal)- the reference track has better clarity, no harsh (more room ?)in total: more natural. I could not hear the differences between the bad tracks - but I´l repeat the test together with my friends, who are able to differentiate und detect "The best"
I'l keep you informed
Michael
...was that 16/44 sounded the same as High-Rez. Somebody even had the test results to prove it! And now you're saying jitter is audible????
Oh, the humanity!! :)
Hey thanks Jon!
I'll check that out... as soon as I figure out WHY MY LEFT SPEAKER ISN'T WORKING!! I'm tellin' ya what, it's just one thing after another.
Sorry - didn't mean to unload on you.
:)
The first time I heard a CD play with bad jitter, I thought it was a tape deck with a flutter problem!
Edits: 08/19/13
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: