[time-nuts] Clocks for Audio gear

Attila Kinali attila at kinali.ch
Thu May 10 21:14:18 UTC 2012


On Thu, 10 May 2012 19:25:33 +0200
"Heinzmann, Stefan  (ALC NetworX GmbH)" <Stefan.Heinzmann at alcnetworx.de> wrote:

> Benjamin and Gannon, the first reference in Ashihara's paper, come to lower 
> figures for sinusoidal jitter with carefully selected frequencies relative 
> to the main signal, which is also sinusoidal. Their results reach down to 
> the single figure nanosecond range, and that can be regarded as the real 
> limit of audibility.
> 
> Of course, that still leaves those who claim to hear jitter in the 
> picoseconds range out in fairy-tale land.

I'm not so sure there. Having been active in the video coding scene
for quite a while i know that a trained ear/eye can be easily a factor
10 better than the average. Eg, after a couple of months of hunting for
bugs in the A/V sync code, i got sensitive to A/V desync down to 3-5ms.
Yes, i know this is below the frame rate. But i cannot otherwise explain
it as something started to feel odd, when the desync got over that limit.
>From somewhere between 5 and 10ms i could usually tell in which direction
the desync was. Yes.. still below the frame rate.

And i know that there are eyes and ears out there that are much better
trained than mine ever could be. So i wouldnt imediatly dismiss it, if
someone would tell me he could detect a A/V desync of 1ms (given proper
frame rate).

So, if someone proves that 1ns jitter is audiable for an average person,
i would definitly not decline the possibility that a trained ear can hear
100ps jitter.

But that is still a jitter level you can get with an crystall quite easily.
You need to take care of a few things to not introduce any jitter from
power supply or bad shielding, but nothing too difficult.

			Attila Kinali

-- 
Why does it take years to find the answers to
the questions one should have asked long ago?



More information about the time-nuts mailing list