Work flow question - Composing / Mixing

Alright, I've seen several different threads / posts about work flow for creating music cues, and they've been most helpful.

 

Here are my 2 main questions:

 

1. What work flow do you use to address revisions?

2. If you're composing in a MIDI environment (with or without audio tracks), are you mixing in that environment, or exporting all tracks as audio to another session?

 

 

I have been composing in Logic, and love it. I've also started mixing in Logic, and love it. However, it's a real pain to address revisions with my current workflow, especially if I need to replay a part in Logic in a session with mix plugs engaged. My audio buffer is set to about 256 when composing, but with plugs in, I'm bumping that up to 1024 to handle the extra processing power required to bounce it out.

 

When opening the session to go back to address a revision request, I can't play in my new ideas with all the plugs on the channel strip because of the large buffer size / latency, and low-latency playback engaged results in unpredictable sync issues.

 

I really don't want to have to go back and pull off all the plugs, play my revision ideas in, and have to remix the whole thing... especially for a small change or 2.

 

What type of work flow do you use for this type of scenario? I'm open to making a change in my process to address this, but it has to be efficient for addressing revisions, without interrupting the creative process.

 

Thank you in advance for your responses!

if u want to try new ideas, bounce the project to an audio file and open it in an empty project.... than if u get something useful, save that project and import those tracks into your original one...

 

in my case, with 1028 buffer i get realtime latency from fx send channels... so i always work in 128

Are you recommending bouncing out stems and mixing in another session to mix, and keeping the midi session separate?

i process the tracks with plugins then bounce it to audio to free CPU... i leave the original, bypass the plugins and hide the track in case i need to go back to it....

When I think I'm ready to start mixing, I bounce everything that needs to be bounced, create a new project, which I'll usually name something like "Song X MIX" and import everything into the new project. Things can get pretty messy when you're in the creative process, and I prefer to start afresh.

 

I usually use the data import feature to import markers, time signature and such. If there's anything else I need which hasn't been bounced, it can also be imported at that time. For example, if I have an audio track w/ a very cool combination of plug-ins on it which I want to use in the mix but which may need adjustments, or a comp track, I'll import those w/ or w/o their assigned plug-ins/busses, as needed.

 

Then, before I start adding any plug-in or making any changes, I'll save the project w/ a new name, like "Song X MIX Balance". Because the balance is usually the first thing I'll work on. This way, I can always revert to the raw version if needed. Same thing when I start adding plug-ins or automation - I'll save w/ a new name. And so on.

 

If I need to make revisions or re-record something - and that happens all the time - I simply go back to the previous project incarnation that has the track I need to work on. Since I apply the same versioning principle when I'm putting the song together, it's pretty easy to go back to the exact version of the project I need to work on.