Hi masters,
On a mix with both audio and midi-virtual-instruments files, do you export the vst and reimport in form of audio files - so that you have the whole mix in audio files - or you mix the song using midi-vst inserts and go hybrid?
I personally keep my VSTi tracks live until my PC can’t handle the load anymore, then I’ll render some to audio and mute/hide the VSTi track (but I keep it around in case I need to make changes and re-render).
So for me at least, the bottom line is that sometimes I eventually run out of processing power in my older/slower computer. Playing back a pre-processed audio track takes way less system resources than driving a VSTi along with a chain of VSTs loaded in the mixer inserts.
For me VSTi performance can vary big time depending on the plugin, and sorts of patches/programs they have loaded. I.E. With something that’s just playing back samples that match the project sample-rate (I.E. Halion Symphonic Orchestra or Garritan GPO) where a central high end reverb plugin or two is ‘shared’ (I.E. on an FX track with sends), I can get a heck of a lot of tracks/voices with dedicated dynamics processing plugins on each track before I need to start rendering things down. In contrast, if I’m working with several heavily processed tracks generated by something a like CPU intensive granular synthesis plugin, I may well need to start rendering down sooner.
In short, I’d imagine your optimal work flow will have something to do with how powerful and compatible your ‘hardware’ is with all the software you like to throw at it. Personally I just keep it simple and keep everything live until some performance issue forces me to render some VSTi stuff down to pure audio. When I start rendering VSTi tracks, I’ll go for the most processor intensive tracks first and work my way down the list until I get the head-room needed to keep going.
When I’m all done with my total composition and mix, I’ll go ahead and make a fresh copy of the project and render EVERY TRACK to audio in their ‘dry and unprocessed’ states. I next unload ALL the plugins. I mainly do this along with my original project file/folder for archival purposes…I.E. in case I decide to revisit the project years in the future, possibly in some other DAW, under different plugins/versions/etc. and find that my initial project file is broken for some reason (I.E. Won’t load because of a really old plugin that won’t have or can’t use anymore). Note, such archives can take a lot of disk space…so for me it’s just a ‘back-up’, and it goes on slower archival media (these days I just use cheap external platter based hard drives meant for backing up files).
I keep them live as long as possible, usually right until I’ve finished the mix. But, when I’ve mixed and exported the song I go through rendering them all so that I have WAV files if I need to remix in the future, because indeed there’s no guarantee that you’ll be able to run the same instruments in future DAW versions…
Mike.
I do much the same… I keep VSTi’s playing and usually don’t render. However, I’m finding that sometimes a sound I want to create is easier accomplished with an audio file. I also sometimes will Render parts when I’m attempting to build a more complex layered sound and want to use the same instrument’s sounds. So, rather than loading four instances of the same instrument, I’ll render, change the patch, then render again, and so on. until the full sound is generated.
Excellent tips here, thanks.
P.S., I think of myself as a beginner in Cubase having only been using it for about a year and half – two years in May! 