21 Sample Delay on External Pres via ADAT

[Ben Chase]

I'm bringing in external preamps via ADAT and there's a 21 sample delay on those vs. the internal preamps of my audio interface. I can slide the clips over after the fact to get them to line up, but am wondering if there's a way to compensate for this on the way in. Any insight would be greatly appreciated. Thanks!

[David Baay]

If everything is going to be recorded via that preamp, you can add that 21 samples to the Manual Offset for record latency. But if you'll also be recording from your interface's direct inputs (hopefully not simultaneously), you'll have to manually compensate or change your Manual Offset for each case. Having one of your nudge values set to 21 samples will simplify doing it manually.

[Byron Dickens]

21 samples? Can you even hear that?

[Ben Chase]

8 minutes ago, David Baay said:

If everything is going to be recorded via that preamp, you can add that 21 samples to the Manual Offset for record latency. But if you'll also be recording from your interface's direct inputs (hopefully not simultaneously), you'll have to manually compensate or change your Manual Offset for each case. Having one of your nudge values set to 21 samples will simplify doing it manually.

Hi, David. Thanks for the response. I'd only ever be using those external preamps in conjunction with the internal ones. I like the idea of setting one of my nudge values to 21 samples. Are you saying, however, that the manual offset is global; i.e., can't be set independently for the internal and external inputs?

[David Baay]

That's correct, the same Manual Offset is applied to all recordings, regardless of the source. I have the opposite issue that ADAT input from synths is over-compensated by the Manual Offset that's configured to make up for the unreported 42 samples of A/D conversion time for analog inputs. It has been requested a few times over the years to have per-input Offsets, but not enough demand to make it a development priority. 

[Ben Chase]

Thanks, David. I appreciate the info.

[msmcleod]

3 hours ago, bdickens said:

21 samples? Can you even hear that?

It's a fair comment.

Sound travels at 1.13 feet per millisecond.   21 samples @ 44.1K is 0.476 milliseconds.

So it's the equivalent of moving a mic approx 6 inches further from your guitar cab, or you moving your head 6 inches further away from your monitors to listen to them.

[Xoo]

Ah, but phasing might kick in depending on what you have in other tracks.

[Byron Dickens]

Really? At less than half a millisecond?

[Xoo]

I did say "might" 🙂  I know some people move drum samples around by tiny increments to get the "right" sound, so they might notice (or think they do!).

[Byron Dickens]

That's why I ask the question " Can you even hear that?" Because I bet you can't. It takes longer for the sound to get from your guitar amp speakers to your ears.

People waste enormous amounts of time fussing over "problems" that didn't even exist before DAWs allowed them to listen with their eyes instead of their ears.

 

Edited September 27, 2022 by bdickens

[David Baay]

Yes, 21 samples is a tiny discrepancy that you will only ever hear as a phase error, but you definitely will hear it in the right circumstance. In my book, it's always best to have your record latency dialed in to the sample as a starting point so that when you get weirdness like PDC failures or parallel compression phasing you know it's not your hardware setup.

That said, there are many unavoidable sources of latency and sync errors (especially for MIDI guys using hardware synths like me), and at some point you do have to ignore things that you can't hear (or control). But it's amazing what a little phase change can do to the stereo image or the attack sound of a layered/doubled part when one signal is consistently early or late relative to the other.

Edited September 29, 2022 by David Baay