opaqueice Wrote: > Sean, could you verify if the the pk-pk value is the "latest" (most > delayed) sample minus the "earliest" sample?
Right. It's the "width" of the sample set and is not terribly meaningful. More important is the shape of the distribution. > > In the latter case you can't compare at all, unless all jitter spectra > are non-Gaussian in the same way... That's not the least of it. Stereophile's tests look at the spectrum of the DAC's output (sidebands around a pure test tone) which is IMHO a rather roundabout way of measuring jitter and may mask or exaggerate other factors. Of course if you are measuring a whole system including the DAC's jitter sensitivity (as opposed to a digital signal) this is what you need to do. I am measuring it by looking directly at clocks or the s/pdif signal in the time domain, using a 10GHz scope and a specialized jitter software package. With this system I have made repeatable measurements of phenomena as small as 2 picoseconds (for example, the jitter introduced in a single gate delay, or a few centimeters along a PCB trace). I also have a dScope analyzer which can generate jitter, and also decorrelate jitter from the s/pdif input and show its spectral content in the 0-96Khz range. It's not very interesting unless the jitter is highly correlated to some non-gaussian source in the audible range, which I've never seen except in contrived cases such as when using the jitter generator (and then only at very large levels, ie 1ns+). As a side note, the guys at Lecroy chuckled when I told them what I wanted to use their jitter tools for - they considered it massively overkill. :) -- seanadams ------------------------------------------------------------------------ seanadams's Profile: http://forums.slimdevices.com/member.php?userid=3 View this thread: http://forums.slimdevices.com/showthread.php?t=25188 _______________________________________________ audiophiles mailing list [email protected] http://lists.slimdevices.com/lists/listinfo/audiophiles
