Originally Posted by arnyk
The first problem we see here is that the referenced paper is more than 20 years old.
I think I am going to stop believing in gravity since Newton thought of it in 1726
Amir apparently wants people to believe that Audio is a very static area, and that we face the identical same problems in the audio systems that are before us as Dunn and Hawksford were worrying about in 1992.
Our ears have not changed from 1992, nor have mathematics of jitter. Implementations have changed. Unfortunately, there have been further developments in other areas. In 1992, a high performance laptop probably cost $3000 and was over an inch thick. Today, it is $300 and one third as thick. Any wonder that you hear noises from your hard disk in your laptop? With a big case and $3,000 budget, we could make sure the audio was far away from the digital circuits. And spending $1 on better analog parts would have been justified. Not now. Folks want to cut every corner. Since GHz sell, and no PC ever specs jitter performance, we are stuck with a crapshoot. And you hearing your hard disk.
One can see a specific example of this problem in figure 25 on page 21 of the cited document. It shows a scheme for de-jittering a "Musical Fidelity" DAC with a "Jitter rejection unit". Now, 20 years later virtually every AVR contains the circuity shown in the "Jitter rejection unit" as a standard feature.
Say what? There is no de-jittering block in that picture:
As we see, it is a test fixture to inject
jitter, not eliminate it. They used it to determine that the Musical Fidelity’s jitter filter stopped working at 5 Khz and lower. From the paragraph right above that diagram:“This technique was used to determine the PLL cutoff frequency of the Musical Fidelity DAC at approximately 5 Khz, imply that any jitter components below this frequency will be unattenuated by this particular unit and will contribute to jitter error at the DAC.”
How can you not tell the difference between a fixture that *injects* jitter and confuse that with a jitter reduction circuit Arny? And then proceed to claim such a circuit exists in all AVRs? Really? AVRs inject jitter in the system?
The above highly relevant example demonstrates the dangers inherent naively reading 20 year old technical papers, as Amir continues to suggest.
Or trust the readers digest of the same from Arny
Note also that the paper Amir refers to references standard AES 11-1991. The current standard is AES 11 -2009, and contains significant updates.
What updates? This is from AES 11 – 2009:“220.127.116.11 Clock jitter
Jitter noise may be either random or in the form of modulation, which at frequencies less than sample rate will cause a timing error to accumulate according to the amplitude and frequency of the modulation waveform.
NOTE AES3 [AES3-4-2009] defines limits for jitter on the digital audio interface.”
AES3-4-2009 is almost entirely dedicated to jitter, showing the allowable levels by frequency for both transmitter and receiver as did the 1991 version (with a tiny change in one graph). More below.
The text of the referenced paper points out that versions of standard AES-11 were marginal for TOSLINK interfaces prior to AES 11-1991. That standard is now totally obsolete and no doubt includes additional updates that further refine industry standard use of TOSLINK interfaces.
Your read of that paragraph is incorrect Arny. The statement regarding Toslink violating the audibility requirement for jitter is based on the simulations in their paper and threshold of distortion audibility. It has nothing to do with AES recommendations. They show that interface time constant has to be below 40 ns which the Toshiba Toslink transceiver they examined failed by having too low of a bandwidth (40ns time constant = 4 Mhz bandwidth).
The paragraph does make a reference to AES11-1991/1992 but that is with regards to a different point related to the rise and fall time of the AES transceivers. I just checked that part of the spec against the 2009 revision and there is no change:2009:“The rise and fall times, determined between the 10 % and 90 % amplitude points, shall be between 0,03 UI and 0.18 UI when measured across a 110Ω resistor connected to the output terminals, without any interconnecting cable present.”1991/1992:“The rise and fall times, determined between the 10% and 90% amplitude points, shall be between 5 ns and 30 ns when measured across a 110-Ω resistor connected to the output terminals, without any interconnecting cable present.”
The updated document states the same values but correctly makes them relative to the sampling rate using the “Unit Interval” instead of nanoseconds. The UI for 48 Khz channel in AES is 162 nanoseconds. Multiplying that by .03 and 0.18 we get 4.9 and 29.2 nanoseconds respectively which when rounded up is the same as what was stated in the 1991/1992 version (5 and 30 nanoseconds).So in neither case has anything changed in this regard Arny.
I hope that the above shows the folly of naively referencing obsolete technical information, which Amir habitually imposes on this forum.
Well, so far it looks like the shoe is on the other foot
* You said the paper was old but didn't demonstrate what from their finding is not accurate.
* You confused a test fixture for injecting and measuring jitter in a DAC with a "de-jittering" circuit in every AVR.
* You said that the update to AES recommendations to digital audio had completely obsoleted the old revision whereas they are essentially word for word the same with regards to the points you picked on.