Pro Tools internal bouncing issues(phasing)

[DWatson]

Long story not so short.

I tracked a bands drums for a demo a while ago and that was supposed to be the end of it. Well sure enough, they asked me to mix the drums for them and send over the individual tracks so their guitarist could mix the rest from there.

No big deal I thought. So I did a mix for them to get the drums where they needed to be, and I did my usual bouncing process in PT(I'm using 9 by the way). What I've always done is use the internal bussing. For example, I'll route the Mono Kick Drum bus out to a mono bus(bus 3 for example), and create a new mono track with the bus 3 as the input. I'll make the Mono Kick Bus output set to unity gain and then match the track I'm bouncing to at the level that the Mono Kick Bus was before raising it to zero.

I bounced down everything, including my parallel bus for the drum kit and even the Verb return. I noticed after listening back to the bounces, that something didn't quite seem the same, especially on the Kick and Snare, a light phase issue and even the verb seems not to sit right.

After trying to figure out the issue, what I did that seemed to fix it was bounce down the individual tracks first, THEN bounce the Verb and the Parallel drum Bus. I think it's because the sends going to the Parallel Bus and the Verb were on tracks that were being bounced as well. and since PT bypasses ADC on tracks that are being recorded..that's where the phase issue was coming from. It really made a huge difference.

I've rarely had to bounce down drum kits with processing for other people, so it definitely through me for a loop.

I guess my question is to everyone, is this typical? And if so, is this how you have fixed it? I may be over thinking it a bit, but when you spend all of that time during tracking to get the phase right, the least thing you need is something like that to screw it up.

Any other tips for how you guys go about doing similar bouncing is appreciated too!

 

[crillemannen]

I don't trust PT's latency correction one bit, that has been proven to me time after time.The best way to get it phased align would be to do as you say bounce every individual track down, one by one after you mixed it with a transient peak on grid so you can phase align that with all the tracks. But who has the energy to do that? Just to be aware of that could help you though so if the problem becomes to apparent you can fix it, but as long as it sounds good, well it is good.

 

[BLUElightCory]

Is delay compensation turned on?

 

[DWatson]

Is delay compensation turned on?

Yes the delay compensation is turned on. No tracks were in the red for the compensation, so all should be OK.

Here are some examples if you would like to hear what I'm talking about.


Original tracks before bouncing

Tracks with everything bounced at the same time

Tracks with the parallel bus and verb bounced after the individual tracks


I can hear a big difference between example 2 and the rest. On the snare especially.

Another test I did this morning was I would solo out the bounced track and the original, put them at the same level and output, and then flip the phase on one of the tracks. All of the tracks completely canceled out except the Toms, Kick, and the parallel bus. One thing common between the Toms and the Kick is that they both had the trigger plug in on the individual tracks, and then I sent the kick to a mono aux and the toms to a stereo aux to apply processing to both the natural and the trigger sound. The parallel bus was fed from a pre-fader send on the Kick, Toms, and Snare Tracks. I do it pre-fader so I can adjust the individual tracks without messing with my threshold on the parallel bus comp.

Take a listen to the Kick tracks inverted.

Kicks Inverted

and the parallel bus inverted.

Parallel bus tracks inverted


Like I said, none of the delay compensation meters were in the red, so I wouldn't think anything major would happen.

 

[amarshism]

yeh i dont usually trust the adc esp when using hardware inserts. i print a sin wave and realign then use the sample gap to input the comp manually

 

[Brett - K A L I S I A]

It is totally normal for triggered tracks to not cancel out since samples are randomly triggered so two passes will never be the same.

 

[DWatson]

Good Point Brett. I think issue is when you bounce to a new track, the delay compensation gets turned off on the track feeding the record enabled track. But, if you have your tracks going through an aux first, then being bussed to a new track, the delay compensation engine stays on. So if you have a mix of those types of set ups on a drum kit, some tracks will be delay compensated and some won't.

This video explains the issues, even though it's for pro tools 11 and I'm running 9.

Delay Compensation issues and solutions