Latency: Optical in vs. Analog in

[Larold]

I have an 002. I have percussion going in the lightpipe and Bass going in analog.

Sounds fine going down, but when I playback, the perc track is always about 33ms. behind. In fact, all the Optical tracks are late.

Anyone else notice this? All the sounds are coming from one machine.

Any help???

Thanks!

[therecordinghouse]

i would like the final word on this too. should we be nudging the tracks that come in on the digital ports?

[therecordinghouse]

i am going to do a test tonight.

snare drum hit in to my octopre connected to the adat in of my 002r.
analog out of the octopre connected to analog 8 of the 002r.
same hit recorded on two different tracks. one via digital in, one via analog in.

[therecordinghouse]

ok, did the test a little while ago.

snare -> sm57 -> focusrite octopre -> 002r adat in 7/analog in 7 simultaneously.

the adat input is consistantly 4 samples later than the analog in on both 48k and 44.1k.

did the same test for spdif

snare -> sm57 -> art DMPA -> SPDIF in L/analog in 7 simultaneously.

the SPDIF input is consistantly 2 samples later than analog in at both 48k and 44.1k.


hmmm

[DigiTechSupt]

What you're seeing is normal - The 002 can compensate for it's own A/D conversion process, but can't for external A/D convertors, since it doesn't know how many samples it takes for the external convertor to complete it's A/D conversion.

Larold - which convertor are you using? 33ms seems like an excessively long A/D conversion time....

The 2-4 samples that therecordinghouse is getting is much more in line with typical conversion times.

[tele_player]

Digi support person - why not pass the word up that it would be good to allow the user to measure and configure the conversion delay for third party connected devices, so that this would work correctly?

[Jsswtpmp182]

2-4 samples is nothing... in no way, shape, or form is 2-4 samples going to effect the sound

[tele_player]

2-4 samples isn't a problem, but I don't think you'll find many interfaces with 2-4 sample conversion delays. If I recall correctly, just using a loopback cable on the Digi 002 S/PDIF introduced an error of dozens of samples (compared to tracks done on the internal converters), and that's obviously not counting any conversion delays.

In short, this is a delay which you'd really want to compensate for, if there's any correlation in the signal on the S/PDIF and the internal converters.