heavy automation project with EQ: writing the current parameters to automation when stopping ?

Hey guys.

 

How would you edit automation for heavy automation projects like a movie ? I am particularly concerned about automating EQs and compressors (and all their parameters, which would be cumbersome to edit in the automation curves) on each track across the whole movie.

 

My ideal workflow would be to highlight a section I need to work on, and then tweak the EQ (as a track effect) as there wouldn't be any automation and then when I am ok with the settings, copy/transfer those to the automation tracks for that section.

 

For a moment I thought the WRITE mode was the answer, but for some reason I don't get, if I stop playing before the end of my selection, it writes an additional point there at 0 value. In this mode I really need to play the selection entirely. That would kinda work but very prone to error. READ does not work since that would mean editing directly in the curves. TOUCH & LATCH do not work since you waste energy just trying to catch & grab the moving EQs. If I put it in other words, I would need a LATCH where the ending value propagates to the entire selected range.

 

I also would avoid region based processing in order not to overload the computer with toooo many plugins.

 

Or any other workable workflow. Any tips ? Thanks & keep it up ! Cheers.

It's pretty rare when you need to automate EQ or Compressor parameters, as those are most of the time set to static values. Are you trying to go for a special effect? Or do you have different regions on the same track that require different compressor/EQ settings? In that latter case I'd most likely move them to different tracks.

Hey David,

 

As we all know, the process you describe fits well for mixing a song.

 

Here I am talking about mixing a film, not a song.

The plugins setting will vary for each character (say the amount is C) and scene location (say the amount is L). Therefore that means that only dialogs, I might encounter C*L different situations. If I dedicate one track per situation, that will escalate quickly out of control.

 

Plenty of films can handle dialog on about 8 tracks by checkerboarding the different characters' lines. Therefore on each track you will have different characters and locations. That's what I am trying to handle. + also matching reverbs between production sound and ADR.

 

I must say that I don't quite get the workflow ...

 

Cheers

I see...

 

Would destructive editing work for your work flow? If yes, then you could use selection based processing...

I thought of that, but there should be a better way

I am just trying to figure out how it's done in real world films. And surprisingly, that answer is not easy to find ... I already looked around quite a bit :-/

So if anybody has some insight, that would be very appreciated !

Thanks guys!

As far as insight...

 

First, normally each character gets their own track.

 

Second, I've never had to choose different EQ and compressor settings for each location. If anything, even within a single location, each scene, in fact even each word from a single scene may have different frequency and dynamic qualities, but that's generally expected...

 

When something is really off I've resorted to Match EQ.

Logic is not the greatest DAW for post production to be frank...

In pro tools you can automate plugin *presets*, which means you can have a couple of settings ("scenes) you can switch on the fly. Which is basically what you are hoping to achieve here.

Logic is not the greatest DAW for post production to be frank...

Agreed, not the best tool for the job.

First, normally each character gets their own track.

 

I also tend to think it's a good idea. Yet, I've seen huge projects where that was absolutely not the case at all.

 

Second, I've never had to choose different EQ and compressor settings for each location. If anything, even within a single location, each scene, in fact even each word from a single scene may have different frequency and dynamic qualities, but that's generally expected...

Maybe I am overthinking it. Still there are instances where it's quite needed.

 

The missing piece of the puzzle might then be something along the lines of what Ploki said : plugin presets that you can switch and automate on the fly in Pro Tools.

This plugin preset automation is not possible at all in Logic then ?

 

R.

This plugin preset automation is not possible at all in Logic then ?

You could try using performances... but it takes a little bit of time to load a new channel strip setting, so that's not 100% instantaneous. Still that could work and worse case scenario you'd use two tracks (for when two audio regions needing two different settings are too close to each other). (See: http://help.apple.com/logicpro/mac/10.4/#/lgcp35966da6)

it is indeed needed when doing serious post work, especially when fine-tuning reverb room-cuts. (preset switching here is a *dream*) Doing that in logic is pretty horrible.

 

top that with 256 track limit, a large post production project can easily be overwhelming.

 

Closest i came was creating custom Smart Control settings where "one knob" controls many parameters, but its a terrible pain in the butt to setup and later to adjust.

 

That said, it is indeed possible to deliver completely professionally done post projects from logic, it's just not the best tool for the job unfortunately.

There’s a workaround to plugin preset automation : load several instances of the same plugin with different presets and automate their bypass state. You can even use this in conjunction with automating sends (in a situation where sends are set tu unity, the sending track where the audio sits is set to ‘no output’, and level mixing is done with Gain plugins). Something like this :

 

You have one track per character, with static eq and comp fx on a per-character basis (or a few sets per character, bypass-automated). Then location-specific treatment is done on some aux returns(one per location, or a few per location) and then you send a given character to a given location (or location « alternative »), and you automate that send. So everything is static (or mostly so), and you automate bypasses and sends. I don’t know if this would work (I’m not a movie mixer so I may just be writing something very stupid here) but that’s how I would try and approach it.

There’s a workaround to plugin preset automation : load several instances of the same plugin with different presets and automate their bypass state. You can even use this in conjunction with automating sends (in a situation where sends are set tu unity, the sending track where the audio sits is set to ‘no output’, and level mixing is done with Gain plugins). Something like this :

 

You have one track per character, with static eq and comp fx on a per-character basis (or a few sets per character, bypass-automated). Then location-specific treatment is done on some aux returns(one per location, or a few per location) and then you send a given character to a given location (or location « alternative »), and you automate that send. So everything is static (or mostly so), and you automate bypasses and sends. I don’t know if this would work (I’m not a movie mixer so I may just be writing something very stupid here) but that’s how I would try and approach it.

 

Yes Arnaud that's the routing I am going for. And I will try what you suggested about having different static EQs on the channels and just automating their bypass state.

Thanks !

Cheers

Great idea Arnaud!

Yes, you can do it that way. Pro Tools was built for this kind of post production, with its write to automation, rehearse and other functions (HD of course), but you can still make pro mix in Logic with some workarounds like these. I gave each character a folder (track stack), so I could keep it organised, and inside it copied tracks when necessary to change eq's or noise reduction. I also used bounce in place quite a lot when I knew it was sorted, and freed up processing power and number of tracks, by having the bounced regions all on one track. In the end with about 25 characters, I ended up with about 120 tracks, all organised into folders, which Logic comfortably coped with.

Although others may frown on it. I have been doing some 'destructive editing'.. I will take sections of audio, change their volume, slight EQ.. This way, later on down the road, you can still play your piece, even if your plug-ins change, or you move to a different system, or platform.

 

Of course you can easily 'paint yourself' into a corner. I make 'channel strips' of everything, and save them in a folder. and use the processed audio for tracks. If I find I need to change, I can go back to the channel strip make changes, and rectify anything. I still use a moderate amount of automation..

 

I have a number of times in the last few years, once I get up to over 130 tracks, and have made a LOT of changes, (talking bout 100+ hours on a piece) the project will become corrupt.. I can still import the corrupt project into a new one, leave out automation, and bring it back to life.

@ronfya, David,

 

Thanks for your comments, happy to have been helpful, hopefully