Phase alignment using different paths to record?

[Drew Mazurek]

I'd like to tap the braintrust here on something regarding a specific recording scenario. (paging Propower and Rich Breen)

Has anyone tested the time alignment of a signal that travels from an Avid HD I/O or an Apogee Symphony box directly into PT vs. a signal that goes into a Presonus Digimax LT and gets into PT via a lighpipe connection?

Im trying to figure out if I can replace a Sytek 4 channel pre (that I rarely use--basically overflow pres) with a Presonus box that I also have. Just trying to simplify my setup and conserve A/D.

[Rich Breen]

Originally Posted by Drew Mazurek

I'd like to tap the braintrust here on something regarding a specific recording scenario. (paging Propower and Rich Breen)

Has anyone tested the time alignment of a signal that travels from an Avid HD I/O or an Apogee Symphony box directly into PT vs. a signal that goes into a Presonus Digimax LT and gets into PT via a lighpipe connection?

Im trying to figure out if I can replace a Sytek 4 channel pre (that I rarely use--basically overflow pres) with a Presonus box that I also have. Just trying to simplify my setup and conserve A/D.

Hi Drew - haven't tested this particular kind of setup. Would be pretty easy to check by just multing a click into the two paths and comparing the two signals against each other. Certainly you'll be adding the Presonus' A/D and transport delays to the equation, so it won't be phase aligned with the other inputs (though it may not matter in your app) Why not check it and post here?

best,
rich

[Drew Mazurek]

Of course Rich!! So simple. Will try that tomorrow.

[propower]

Originally Posted by Rich Breen

Hi Drew - haven't tested this particular kind of setup. Would be pretty easy to check by just multing a click into the two paths and comparing the two signals against each other. Certainly you'll be adding the Presonus' A/D and transport delays to the equation, so it won't be phase aligned with the other inputs (though it may not matter in your app) Why not check it and post here?

best,
rich

Exactly. But the next thing is how do you fix it? In PT you have to either manually add the right delay comp number to the track(s), manually move them, or add a plugin to do it (Metaplugin for instance - used to be a Digi Time Delay plugin for this if it's still around). What if the Presonus box is faster? The method has to support adding a negative number too

[Drew Mazurek]

Presonus through an ADAT input is 36 samples later than Avid HD I/O. The pic shows a square wave, but I used a few different sound sources.

[Rich Breen]

Originally Posted by Drew Mazurek

Presonus through an ADAT input is 36 samples later than Avid HD I/O. ..[/IMG]

Cool, good to know. That's at 44.1k?

Originally Posted by propower

...the next thing is how do you fix it?...

That's what Time Adjuster is for, though unless multiple mics on a single instrument/ensemble are being split across interface types (generally a bad idea), the 36 samples *may* not matter to Drew.

[Matt Hepworth]

Little bit older thread, but I just did some testing (and solutions) for this with various scenarios.

I'm not certain the DLA plugin is available in Native like it is in TDM, since I track exclusively in TDM.

Solution:

Measure delay for converter.

Insert DLA plugin on the the fastest converters (Apogee or HD I/O)

Set it to the number of samples difference you measured

Turn OFF compensate for converter delay on input.

Use the ADC bar in the mixer view and insert the delay plus 3 samples (TDM delay) as a negative on the other converters.

The reason we do that as well is so you hear both recording and playback in phase. Otherwise, playback would be in phase, but not recording/monitoring or vice versa.

I figured this out recording drums and noticing the variation in phase when using my HD I/O with a 192 (symphony is on the fritz...).

This works perfectly. The simpler solutions will play back fine, but you'll hear phasing/comb filtering if you're splitting across converters as you monitor. Same thing if you leave the compensate for recording delay box checked.

Hopefully this helps others. You can even mult/parallel without comb filtering across various converters.

BTW, 192 is 48 samples later than HD I/O. Mackie Onyx was 6 (4 connected to the 192 instead). Apollo was 73, IIRC.