Changing Buffer and Automation?!?

[tha]-[acksaw]

Every now and then I needlessly get it to a head trip and start over thinking the inter working of ProTools.

When mixing, I usually write my automation before I start to really use some of the more finalizez type plugins. And of course, the more sophisticated finalizer plugins can really bog down that last bit of available processing.

So, I mix and write automation at a buffer of 128.

Then when I start to work my mastering section, I usually will up it to 512 or 1024. Then I can finish my mix master without bogging my CPU.

What I'm wondering is... does automation I write at the 128 low latency actually happen off time should I move it to 512? Will it now happen 521 samples later then the part played? I hope I'm asking that right and it makes come sense.

Does the buffer delay effect the automation in any way?

I realize that if you write automation at a buffer of 512, for instance, you will actually be making your rides 521 samples later then where the audio lies in the sessions time. But I'm wondering if automation recorded at low latency, messes up when the latency is changed to a higher value.

Or is it just the final speaker output that's delayed by the buffer size.

Sorry for the long winded bout. Hope it made sense. And I hope everyone is having a nice Holiday season. Thanks up front for any feedback. And DUCers ROCK!!!! And I'm sorry if this comes across as a stupid question.

[albee1952]

While I follow your train of thought, I don't think that's the case(and if it was, can you hear that the automation is off by 512 samples??). Maybe a better idea would be to concentrate on mixing and nothing else, then bounce to disk. Do your "mastering" on the final stereo file(where all your worries would be gone)? To each, their own, but I prefer to leave any mastering or master treatment until afterwards.

Having said that, I don't think its a stupid question at all(just not sure it needed asking). Since you DID ask, maybe one of the Avid techs will offer some insight(but this is a user conference)

[tha]-[acksaw]

Thanks albee. I appreciate your feedback as always.

You are totally right about mastering in a separate session from the mix. And I feel like I'm about 80% there, in terms of getting the mix right before mastering. But often I need to fix a few minor things in the mix, once I start the mastering process. I'm getting there. Baby steps, right? LOL

I need to stop being lazy and just do a test. I'll get on that and report back real quick.

Thanks again for taking the time to respond brother. Much respect.

[tha]-[acksaw]

Ah yes, nothing better then some self induced learning...

So, I took a mono drum loop. Did a pan ride moving it from hard left to hard right at the middle of the loop. I made sure to do the automation by hand from a controller, and not with the pencil (not that it should matter). I did the pan ride with a hardware buffer of 128.

I then did a bounce to disc, into a stereo file at 128. Then I did another bounce at 4096 buffer. I loaded the two files back into the session.

The first cool think I notice is, the file done with buffer 4096 was not 4096 samples later then the other. So, during the bounce, or more likely the latter conversion, it accounts for the buffer setting and moves the audio back in time accordingly. WONDERFUL!

Then I put a phase invert on one of the two stereo loops and play it back..... NOTHING. Perfect cancellation. All the way through the pan ride and everything.

So I guess it's official. Digi has thought of everything. Buffer settings don't effect already performed automation.

Another brilliant case of me just over thinking things. Carry on. LOL

[albee1952]

Nice! Nothing like empirical evidence Thanks for sharing