Sound Design Stack Exchange is a question and answer site for sound engineers, producers, editors, and enthusiasts. It only takes a minute to sign up.
Sign up to join this community
Between these two phases of work, how does your workflow or the "normal" workflow commence?
Say you are editing backgrounds…you create backgrounds. As an editor, do you volume graph/automate them to the scene cuts/events and then hand that product to the mixer? The other way is to create the backgrounds, put them on the project and hand that to the mixer for him to discern.
I've taken a few days to think of the best answer for this question, and the best I've come up with is there really is no "right or wrong" here, only "quicker or longer".
But first, lets take a walk into the past.
Back in the "old" days, the division between editor and re-recording mixer was very defined. With the sound for film originally being edited on magnetic film, there wasn't much the sound editor could do to manipulate the sound other then a partial fade by actually dissolving the magnetic oxide off of the film with acetone. Even adjusting the levels during the transfer process weren't a great option because if you were too low, there would be too much hiss, or on the other side if you were too loud it would create distortion. With the sound editor only being able to actually hear a few tracks played back at once, their primary job was to find the sound and place them in sync (not an easy task).
The re-recording mixer had the luxury to listen to all the tracks together and was their job to balance them all together. If there was a problem with a track, or something needed to be added, the mixers hands were tied. The tracks would be taken off the machines and sent back to editing.
As we walk a bit forward in time, multitrack tape machines made the sound creation job easier, and more creative. The overall track counts started to increase. Unfortunately on the dub stages, the re-recording mixers had to start pre-dubbing tracks in order to be able to play them all back at once during the mix due to the available real-estate on the consoles. This led to bigger consoles and the invention of mix automation.
Fast forward to today - (what in the heck is tape?)
With both editorial and the mixer using the exact same gear the lines between the two positions are becoming very blurred. DAW's have given sound editors the ability to literally create hundreds of tracks of audio. They have also given the re-recording mixer the ability (along with new digital consoles and control surfaces) to play them all back and manipulate them without extensive pre-dubs.
This is where quicker and longer comes in.
For the sound editor it obviously makes their job longer to automate volume, pans, and sometimes even reverbs to their tracks. Most editors I know prefer this because they have been working closely with the Director and the sound supervisor anywhere from 4 to 12 weeks. When the session leaves them, they know the tracks will be at least listened to once the way they have intended them to play. While the re-recording mixer has the ability to change anything, they can focus on the balancing of the elements to work on the emotions of the film and make sure it flows properly.
If the tracks have no automation and is left up to the re-recording mixers to weave through them, the final mix will take longer. Longer on a mix stage equals money.... a lot of money. You can generally get one week in an edit suite for the cost of one day on a mix stage.
Like I said at the beginning, there is no right or wrong. As a re-recording mixer I can see it both ways. If the tracks come to me with volume and pans, its great. (My wife likes this as well because I'll be home for dinner earlier) If the tracks come to me with nothing though, I also enjoy it because I get to really mix a lot! (Mixing is fun...and that's why we all do this right?)
Some people don't want me messing with their stuff, others gladly welcome it. Like I said there is no right or wrong. All I ask is that it's in sync! :)
Where a rerecording mixer is involved, i'll try to balance things as well as i can before i hand it over.
I'll cut dialogue, using fill to smooth the edits. If it's very noisy, i'll hard process with some NR, but i'll leave the original region muted on a worktrack below.
I'll balance level on the FX and whatnot. Maybe i'll do some complex volume or pan automation moves if i need them to sell the FX to the director; then i'll ask the mixer if they prefer me to leave them in or drop them.
Another thing that i think is very important for the handover is to make sure that the soundtrack will play well with all faders at 0. To do this, i make a session copy before the handover, and use the Gain Audiosuite plug-in to adjust the level of all the regions so that, if the mixer decides to delete all automation, it still sounds like a film.
Finally, i like to clean up the session so it doesn't have anything unnecessary in it. I make a new session, then "import session data" from my edit session, and consolidate with 2000ms handles. Always make a backup copy before doing any of this, though. The last thing you want in this industry is to have to answer "i need X" with "i don't have it anymore".
I think this is a good process with which to approach any film, even if you're both editing and mixing. It's very hard to do fader moves on an atmos track that's sitting at -60dB, and it makes for a nice neat session that lets you approach the film with a fresh perspective.
EDIT: All of this is subject to the preferences of the mixer, of course.
I think that nowadays, what it really comes down to is communicating with the supervisor and the mixer. With digital workflows, it seems that there's no hard-and-fast answer for almost ANYTHING any longer. Having experienced both receiving tracks that were somewhat of a PITA on the stage, as well as delivering those, I've realized that really what it comes down to is never assuming anything, and always being able to backtrack. And quickly. With people glaring at you. And a knife at your throat. Etc.
This is a great question. I am very curious as to what everyone does here.
I once did have a misunderstanding when freelancing where I was asked to do a dialog edit....So, I cleaned the dialog with the necessary noise floor, made clean edits, set the dialog to general broadcast tech specs, and processed the voices to match space and perspective in the mix (Some of the processing truly just made some of the edits possible that otherwise would not fly.). I was greeted later by the mixer who said in a confused voice "Wow, you went ahead and mixed it." I apologized and asked what I could do to make up for the misunderstanding. I don't think she minded but she was surprised. She wanted me to put in noise floor and use simply a fade and cut technique to smooth edits. Dope....
In all other cases, I have always been asked to create a stem within a session that contains all elements individually available to the re-recording mixer. So, I mix all of the sound effects, music or dialog relative to its own elements in a manner that I think best describes the program with the understanding that this will make the final re-recording process easier and allow the re-recording mixer to change any of my preliminary processing and/or levels. This means the mix will play decently well with all stems at unity. Sometimes the mixer will all together remove elements and other times I will find my SFX very forefront in the final mix with music barely whispering in the background.
What do you the rest of you guys do?
As a sound editor you need to balance your material, otherwise how do you know each element is actually contributing? (but I am not a dialogue editor - that requires a different approach to ambiences/FX etc) especially since an ambience for a location may have dozens of layers contributing to it. That balance is presented to the predub as the starting point, but listening in an edit room is a lot different to listening in a full size dub stage.... And it is the mixers perogative how they then approach it. But if they were to remove my level automation then the amount of work they have to do would increase ten-fold!
Being aware of what level the production audio will play at & what backgrounds come with it (that will need to be matched/masked etc) is also crucial.
As far as panning goes, I have multiple tracks feeding each stem dedicated to C, LR and LsRs (so eg current project AMB A = 4c + 4LR + 2LsRs) but where my projects are mixed it is not in the box, so each track goes out a dedicated hardware output...
