The solution is in your hands. Buy some test gear. Learn to use it. Pay somebody to teach you if necessary. Do your own tests. Post them on here.
What's your problem with that?
w
What's your problem with that?
w
It doesn't always give the results you want. That's the nature of actually doing measurements.
That's one of the advantages of being totally non-commercial- one doesn't have a stake in the outcome and can let the chips fall into the mayonnaise.
That's one of the advantages of being totally non-commercial- one doesn't have a stake in the outcome and can let the chips fall into the mayonnaise.
Tests have already been published here & scope shots presented with the variables controlled! These were rejected by you, by Sy. This again is the usual tactic - I criticise the validity of the tests, you fall back on the "well show us how to do it argument". If you can't defend the testing method without resorting to this then it really doesn't give much credibility to it.
Last edited:
I have shown tests phase noise spectra included - and not only.
Of the Hiface device modded/ not modded, and other devices (not cheap chinese sh..)
The Hiface has ~an order less data correlated jitter. The modded Hiface betters it because of the eliminated residual bounded uncorrelated jitter.
All this is measured directly on the transmitter output, not through a totally undefined / uncalibrated quality receiver, or even worse, a full dac.
A low quality dac, in addition.
By the way, here it is the inherent jitter of the dcx2496, used by You, Sy, in those tests of yours:
http://esken.net/ergo/blog/wp-conte...arta-48k-samplerate-yelloworig-greenergos.png
The original is the yellow trace!
Of the Hiface device modded/ not modded, and other devices (not cheap chinese sh..)
The Hiface has ~an order less data correlated jitter. The modded Hiface betters it because of the eliminated residual bounded uncorrelated jitter.
All this is measured directly on the transmitter output, not through a totally undefined / uncalibrated quality receiver, or even worse, a full dac.
A low quality dac, in addition.
By the way, here it is the inherent jitter of the dcx2496, used by You, Sy, in those tests of yours:
http://esken.net/ergo/blog/wp-conte...arta-48k-samplerate-yelloworig-greenergos.png
The original is the yellow trace!
Wow, that's weird. I don't think I've seen anything like that bad out of my stock DCX. But I haven't tried 12Khz or whatever that is. Only up to about 5Khz.
Can you describe this measurement set-up a little better, please?
Can you describe this measurement set-up a little better, please?
The intermod performance is the most telling (on 9+10kHz and 19+20kHz). Incompetent design, no doubt about it.
Indeed, using this device in any "measurement" regime would seem to be foolhardy & badly misguided. I continually asked for this device to be proven in the set of measurements produced using it - my requests were ignored - I believe this vindicates why I was asking for such baseline measurements to be provided - to prove the equipment is capable of the resolution being attributed to it by SY - "exemplary" I believe was the word he used to describe this POS?
Last edited:
Sorry, I was wrong what SY said was not "exemplary" but "impeccable" as in
http://www.diyaudio.com/forums/digi...tenuators-jitter-reducers-40.html#post2355223
Thanks for the link- and I agree that the stock DCX output is nothing to write home out.
You guys are still missing the point. Still not understanding the fundamentals of your own request.
And I do not believe this.
You guys are still missing the point. Still not understanding the fundamentals of your own request.
No, not you.
This was opened supposedly as a thread about whether "FTT" is a valid method for determining sound quality. Probably this meant spectrum analysis using FFT.
It seems that the OP determined that it was not valid to his satisfaction. But now that there are spectrum analysis plots posted that might invalidate a measurement that he did not like, FFTs are valid!
Can't have it both ways.
This was opened supposedly as a thread about whether "FTT" is a valid method for determining sound quality. Probably this meant spectrum analysis using FFT.
It seems that the OP determined that it was not valid to his satisfaction. But now that there are spectrum analysis plots posted that might invalidate a measurement that he did not like, FFTs are valid!
Can't have it both ways.
Pano, keep in mind that the picture is NOT a spectrum of an 11kHz test tone, though that's implied. A spectrum of that signal through a DCX looks quite a bit better. If I read the website correctly, these are spectra of a special test tone designed to check the DAC chip's sensitivity to the jitter of the signal at its clock pin.
Equivocation. The original contention is that FFTs aren't particularly good at determining sound quality. But they are excellent at determining distortion - no dispute there. So the most recently linked FFTs, rather than invalidating a measurement (they may or may not, depending on the particular non-linearities which they reveal) rather invalidate the choice of piece of kit being used to make that measurement. That's because the justification for that choice which was offered was that it (the kit being proposed) was 'impeccable' - which has now been shown to be a butchery of the commonly accepted meaning of that word.
The 'didn't like it' thesis appears to be a red herring. It may or may not be the case that jkeny didn't like the measurement, but that's quite tangential to whether the measurement made was a valid one. So far, the jury's out on its validity.
Sorry, I thought you wanted to discuss FFTs rather than try to reopen a closed thread after being warned by a moderator.
I won't comment further.
I won't comment further.
You're absolutely right, I thought it looked familiar. 11,025 @ 44k1 & 12,000 @ 48k. I just wrote a program to generate these tones a week or so ago.
Incredible.
w
- Status
- Not open for further replies.