offline bounce different than realtime bounce

[grant howarth]

Hi all,
I have just done my first mix in protools 11. I did a real time bounce down as i normally would and then i figured i would give the offline bounce a try. When i brought both files back into protools i flipped the phase on one of them to see if i could hear any differences. (i believe this is called a null test).
I was surprised to hear some guitar during the playback. Specifically the tracks that I used the "waves CLA Guitar" plugin on. When i removed this plugin and did the same null test the guitar disappeared.
With a careful blind listen to both mixes i was able to pick out the mix that had the off line bounce as the guitars in it sounded just slightly quieter.

So i guess i am wondering, is this a known issue with protools 11? Would this be a bug in the waves plugins? Any info would be greatly appreciated. Thanks!

[JCBigler]

There was a thread a few days ago about this very issue including Waves plugs. Not quite sure yet if it is a Waves bug or something else. I use a lot of Waves plugs, so I hope it's something else.

[tamasdragon]

I don't use Waves, and in my tests offline bounce was accurate.

[JCBigler]

This may not be the same exact issue, but he was using Waves plugins.

http://duc.avid.com/showthread.php?t...Offline+bounce

Fwiw, I did the same test with one of my sessions with waves plugs from the NPP, and did not have any issues.

[BobbyDazzler]

Were you using a chorus or phase setting on the guitar? That could account for the difference.

[JFreak]

Chorus and every random process makes it impossible to perform a null test. I believe that the Waves plugins' "analog" setting is one such random process, which cannot null if you have two different bounces -- even if you don't change a thing.

That is why back in the analog days we took three different "bounces" because reverbs are random and sometimes some gear had not warmed up properly, thus changing the sound between "bounces".

Just live with it or don't use random processes. Null tests of real mixes IMO are worthless. The null tests are only useful if you really know what you're testing and why. In real world it's just as useful as comparing pink noise with levels 0.1 decibels higher and lower.

[Drew Mazurek]

Print the guitar track with its effects and do it again.

As others have said, those effects are random and will never null.

[feliponk]

I don't like Waves

[grant howarth]

Thats a very good thought about the guitar effects containing chorus ext. They do. I only did null tests to make sure i had confidence in doing off line bounces. It never occurred to me that some effect ext could be random. Thank you greatly for the responses.