[LAU] RME analog out of phase to ADAT connected Behringer analog out

Paul Davis paul at linuxaudiosystems.com
Tue Jun 14 11:22:54 UTC 2011

On Tue, Jun 14, 2011 at 6:48 AM, Ralf Mardorf
<ralf.mardorf at alice-dsl.net> wrote:
> On Mon, 2011-06-13 at 09:14 -0400, Paul Davis wrote:
>> On Sun, Jun 12, 2011 at 10:37 PM, Ralf Mardorf
>> <ralf.mardorf at alice-dsl.net> wrote:
>> > Hi,
>> >
>> > I made a sound check, that isn't a valid test, just a test to get a
>> > coarse impression, but it's positive that for the 'RME analog out / ADAT
>> > out test' the signals were out of phase.
>> I have no idea why you would expect two analog signal paths with
>> independent D/A conversion would ever be in phase.
> Thanks, that was what I wanted to know, so it's not broken. I didn't
> expect that, I was told that when synced "NOTHING" would differ. I
> claimed that I don't believe this and seemingly I didn't mistake. It's
> important to be aware of this ;).

no, you did make a mistake. your test (to the extent that i understand
it) didn't sync anything. you used two different D/A converters
without a common word clock. even with the same word clock, they would
not necessarily be synced because they may buffer different number of
samples internally in front of the actual D/A circuitry. but this is
sort of irrelevant because you didn't actually sync anything together
at all as far as i could tell.

> No valid test, just a test to hear, if at least the sound quality of a
> (less good?) CD player is reached, fortunately it is reached, even with
> the elCheapo Terratec, 44.1 KHz it's quasi equal for the RME, Terratec,

 first of all, CD players don't do direct digital audio playback. the
audio you get telling the disc to play an audio CD is different than
you get when reading it as a digital storage medium. secondly, you
appear to have made the cardinal error of trying to do psycho-acoustic
testing without using even blind, let alone double blind testing. this
means that your assessment is meaniningless. if you don't believe me,
try watching (at least the first 2 speakers) of this:


(a video that anyone with any pretensions to being an audio engineer
needs to watch, assuming they can listen to english).

please note that i did not that say that your assessment is *wrong*.
but the way you did it is. your conclusion about sound quality might
be right, but its equally likely to be wrong.

More information about the Linux-audio-user mailing list