question for you sound guys
As a producer in an edit session I have a question. At the end of a session and the mix is locked how long should it take to "bounce" the tracks?
For example, say I have a 23 minute show. I need a full mixed track, and also a split track version (say 1 track VO, 1 track sfx, 1 track music) Would it be 23 minutes for each one of these, so 4 x 23 minutes, or could it be faster? I just wonder because I feel like I am getting banged for extra time while this process happens.
They are working on Pro Tools by the way.
Thanks for an enlightenment you can give me.
The answer to your question is that bouncing takes as long as it takes. Different mixers have different methods of achieving the final mix, and skipping steps is a pretty risky action. After all, the time it takes to double check a track is really insurance against running into problems once you leave the mixing theater. Discuss possible time-saving techniques with your mixer, but trust their judgement on the matter.
That said, here's what I do to save time.
I record sub-mixes of the elements as I go, that way at the end of a mixing pass I already have my stem mix files. My mix sessions basicall look like this (with 8 tracks of Dialogue, 4 tracks of ADR, 2 tracks of narration and voiceover, 8 tracks of effects, 8 tracks of atmospheric effects, 16 tracks of foley, 8 tracks of music):
DIAL 1..8 [INPUT=none / OUTPUT=bus 1-2]
ADR 1..4 [INPUT=none / OUTPUT=bus 1-2]
VO 1..2 [INPUT=none / OUTPUT=bus 1-2]
EFX 1..8 [INPUT=none / OUTPUT=bus 3-4]
ATMOS 1..8 [INPUT=none / OUTPUT=out 3-4]
FOLEY 1..16 [INPUT=none / OUTPUT=bus 3-4]
MUSIC 1..8 [INPUT=none / OUTPUT=bus 5-6]
DIAL PREMIX (stereo aux) [INPUT=bus 1-2 / OUTPUT=bus 7-8]
EFX PREMIX (stereo aux) [INPUT=bus 3-4 / OUTPUT=bus 9-10]
MUSIC PREMIX (stereo aux) [INPUT=bus 5-6 / OUTPUT=bus 11-12]
DIAL STEM (stereo) [INPUT=bus 7-8 / OUTPUT=out 1-2]
EFX STEM (stereo) [INPUT=bus 9-10 / OUTPUT=out 1-2]
MUSIC STEM (stereo) [INPUT=bus 11-12 / OUTPUT=out 1-2]
The Premix tracks are stereo aux tracks. These are where you'll put global effects (eg. reverb, compression) or fader adjustments that you want to affect the entire premix. The Stem tracks are just record-enabled stereo tracks. Rather than just playing the film as you mix, punch into record and record your mix changes as you go. Watch where you punch in, because you can inadvertantly clip off reverb tails if you're not careful (use quickpunch and preroll a few seconds before punching into record). That way, at the end of the pass you'll already have your stem mixes ready to go. Next, disable and hide all tracks except the Stems, and create a new track for the final mix. After you change the inpute and outputs for the Stems, the mix will now look like this:
DIAL STEM (stereo) [INPUT=none / OUTPUT=bus 13-14]
EFX STEM (stereo) [INPUT=none / OUTPUT=bus 13-14]
MUSIC STEM (stereo) [INPUT=none / OUTPUT=bus 13-14]
MIX (stereo) [INPUT=bus 13-14 / OUTPUT=out 1-2]
Make sure that the stems are no longer record-enabled, and record-enable the MIX track. Now, sit back and enjoy your final double-checking mix pass. At the end of this pass, consolidate all the regions in your stems, and relable the new regions as appropriate. Go into your Audio Files folder and locate the final stem audio files and the mix audio files, and your ready to rock.
If you use this procedure, you'll shave 92 minutes off of your studio time. However, the first time you give this a shot, double-check the elements against your picture: this arrangement can get pretty confusing pretty quickly, and its awfully easy to misassign an output.
Very impressive. Can we make that part of the FAQ?
Do you do calculus and the NY Times crossword puzzle at the same time?
Over what period of time did it take you to build to that sort of complexity?
Ty Ford's "Audio Bootcamp Field Guide" was written for video people who want better audio. Find out more at http://www.tyford.com
All mixing aside, the bounce down time is dependent on the software used along with computer speed.
Some applications, like MOTU's Digital Performer for example, can bounce faster than real time.
Others like Pro Tools, bounce in real time. Meaning if you have two different 23min mixes, it will take each mix 23min to bounce down.
If you think it might help, go ahead and put it in the FAQ.
I used to always bounce the tracks individually, which is a whole lot of fun when you're faced with multiple language versions of the same program (4 x stem tracks, M&E mix, english mix, farsi dub, etc.) A few years ago I sat in on a short film mix where all the tracks were being premixed through Aux's and then recorded to stem tracks, which blew my mind. The practice takes some getting used to, but it's saved me countless hours of bouncing.
Also, whenever I bounce a track, I always felt anxious about shipping it without first checking it back. By recording the mix to a track group within the mix itself, I can verify the integrity of the new file as I go.