Workflow: Keeping Track of Your Designs

Question

I'm trying to figure out a better workflow for designing more complex effects with multiple layers of heavily-processed sounds. So far I've just been using AudioSuite for processing or, if I have to automate a plug-in, I put it in an insert and then print it. All in my FX session. This works, but it's not a very fluid workflow and makes it hard to change things after the fact; if I use AudioSuite, I'll have to remember/store somehow what settings I used, go back to the unprocessed version and process it again; if I use dedicated tracks for each composite effect with inserts, I retain all the plug-in settings but end up with a huge number of tracks.

So I'm curious to hear what you guys do. Do you have a dedicated session for each effect, or one big master session, do you save plug-in settings so you can come back to them later, or do you just forget about it?

Answer 1

All my editing & layering happens in one big super session (ie all reels of the film, each reel starts on the hour etc) which doesn't have any live plugins running sicne I dont want to be relying on them when migrate to the dub stage for predubs (although I do have a few inactive source & destination tracks in that session for printing simple processing eg doppler, IRs etc)

Anything involving a lot of processing I do in seperate sessions (eg Weresheep Vox is one session, WereLamb another etc) so I have a chronological archive of versions of each session & the plugin setup/routing etc remains intact... But also that work is often done 96k whereas my main edit session is 48k....

As with score cues it can be very important to be able to revert to an earlier version, and potentially continue developing it in a different direction... Same goes for the super session eg when a scene is dropped, but then returns 3 cuts later.. Need to be able to conform between versions of the cuts....

Answer 2

Good question, I'm interested to hear how others chime in on this.

First off, I almost always have a build session and a stem session. The build is what I edit/create in, the stem is what I deliver to the stage for mix (a tidied-up build session). I'll talk with the super/mixer before I deliver in order to know just how much unwinding he/she would like to be able to do, and not over-deliver a bunch of tracks they'll never use that will only confuse them.

In the build session, if I have a heavily layered/processed sound that I anticipate either having to make derivatives of later down the timeline, or wanting the ability to roll back a step or two in the design process. I will cut the sound on "design" tracks, using all the plug-ins I desire, automate, and then print to an FX or BG track and then hide & make inactive those design tracks.

On the other hand, if it's an Audiosuite process, or rather a string of static (un-automated) plug-in instances on a single effect, I usually create and save a VST Rack preset in Soundminer and will name it using a "project name abbreviation - obvious description" scheme. For example "RoKS - Insect Feet", occasionally I'll append that with search terms as well as pitch speed so I know my source material later on. If I'm processing way outside the lines, I'll drop the source file unprocessed in the timeline below the processed one, cut both together and mute the unprocessed clip. The VST Rack Presets are great because they not only save the rack position of each plug-in, but also the state of each plug-in whether it's a saved preset or not.

Answer 3

When I work with audiosuite processing I rely heavily on track playlists and deactivating plugins.

Let's say, for instance, that I am working with a single piece of dialog that I want to expand and de-ess: I start out by inserting and programming the an RTAS version of each plugin on the track. I then "Duplicate Playlist" and name the new playlist something like "[DLG]proc1" to tell myself that this is the first round of processing done to the file. I then copy and paste the settings from the RTAS expander/de-esser into the Audiosuite verison and render the effects onto the track. Once this is done, I deactivate the RTAS plugins by COMMAND-CONTROL clicking them—I don't delete the RTAS plugs off of the track.

I'm drawn to this method because it's easy, not only to return to earlier versions of the audio, but also to re-visit the plugin settings that you used to get there.

In the track comments I sometimes put a note about the contents of each playlist, though this usually isn't necessary for most material. Still, it can be nice to know what you did on each pass, as well as the date, should the director or your supervisor wish to return to a particular version.

Best,
~Matt