Obviously you are not forced to, it's your choice. I may be rough at times (since my toleration to BS is very limited) but then why can't you just ignore me, instead of waving the ignore list flag, like I should care about?
May-be it is not here that lie audible differences ?
I 'm still asking-me questions why I can hear differences of character between 2 DACs that both measure so flat between 20 to 20 000Hz and down under what we consider totally inaudible thresholds (noise + HD).
Last edited:
Good choice of using the ellipsis. Readers can insert an attribute of their choice; I guess you could easily guess my choice.
I'll go back and check. I do have a concern with Pavel's ABX result as it shouldn't be detectable. OTOH I would happily build that line stage if I needed one. Nothing wrong wtih going OTT for **** and giggles.
Some people can clearly hear if a CD were made without dither. Some people can hear smaller things than that. As Jakob pointed out, published hearing threshold numbers are not limits at all. Also, what people can or can't hear is very dependent on test conditions. I was able audibly discern distortion in music reproduction by adjusting ES9038Q2M HD2 and HD3 compensation registers slightly away from measured levels around -120dB. Soon other people will be able to try it as well, when Allo Revolution dac goes on sale. Users will be able to tweak the same H2 and H3 distortion compensation registers.
What's annoying is that you chose to not even consider the possibility that you may want to hear a difference in "sound character" (whatever that is).
Second annoying thing are your excursion into engineering, and when caught in missing EE 101, you become defensive and pull out the "differences in sound character" as an universal backup argument.
I've said this many times, I have way more tolerance to clueless people, than to proudly clueless people; we have a few remarkable samples of the latter here.
Calling me an alcoholic is the last on the list of annoyances; that's a matter for the moderators on duty, not for me.
EDIT: Forgot to add to the list of annoyances the complete lack of sense of humor, as proved by your reactions to messages including
.
Last edited:
Maybe no point posting those recordings then, just continue with the blah blah blah
No recordings of that dac were made. AK4499 is still being worked on, so various possibilities might exist for it. I'm still undecided on the matter, but its looking more and more like no one will ever come to see for themselves first hand. Even those who could do it don't seem to have any interest. Only interest is in arguing.
Look, Matt, what if I post some recordings that sound different, people will say I am lying about how they were made or there is some pathological condition I am not admitting to. So what then, just more wasted effort and more arguing.
Never heard of that. You are claiming that you can play a CD and someone can tell if dithered or not. I have heard about hearing dithered vs not for same sample. But that is 30dB above the number Bonsai posted.
No measurements so you don't know what was actually changing and by how much.
One of the interesting aspects of the correction technique that Pavel has developed Digital Distortion Compensation for Measurement Setup is that the distortion correction really only works at a specific exact level and frequency. Change the level or frequency and you need completely new corrections. I would not be surprised that the ESS technique has some similar limitations.
What would be interesting would be to use a known good ADC (as in useful for recording music) to capture the playback from each DAC and then compare them. Another classic tric would be to loop the same trach through one DAC several times which should amplify its "quirks" and do the same with the other. If there are subtle audible differences looping through several times should amplify those differences i would think.
...And thousand more useful, for people like me who are no more in the position to can listen to various DACs, to get an idea of the real differences they can have that to fight about the way one or an other evaluate them by listening.
IIUC, that method pre-injects multiple specific frequency, phase and level sine waves to cancel out measured harmonics.
Seems more likely simply it bends the D->A transfer function slightly according to a 2nd or 3rd order polynomial.
The transfer function through the origin is probably more important than full scale. I've suggested before that plotting the residual from a notch filter down to near vanishing output is possibly revealing. The attitude seems to be plug it into the AP or nothing. The cry for "new" measurements is pointless they are already there.
I was more like hoping to get to your ignore list, any chances you could add me pretty please 😀.
- Status
- Not open for further replies.