I wonder how you MIX your tracks when working with VSTi.

- Do you bounce your VSTi into audio tracks before mixing the track?
I think I will start doing this (again). Main reason: I focus on the MIXING - not the composing, which should be done by then.
Later it can also be useful to have the audio file of a VSTi.

- What is the audio spec for those files?
I'd go with WAV 24bit/48khz (I'm on PC).

- Do you record the VSTi signal to audio without the VSTi internal effects (e.g. reverb, delay, etc.)?
Often these effects are very important to the sound. without those effects the sound is much different.

What are your thoughts about this topic?

 

I render all channels to 24bit/44.1 kHz audio tracks. Then I open them into a new session for mixing.

 

I personally mix as i go along, unless CPU is being a total bitch, which then obviously makes sense to stop lag and latency issues. The reason I make tracks like this is because once something is exported, it is usually a hassle to change the original thing again. For some people this is better, for reasons you have already mentioned but for me personally, i prefer doing lots of micro edits, soloing specific elements and slowly building them up to make sure that the mix is gelled better. For me its almost a process of reverse engineering a composition and then cleaning up any mucky elements. This goes on for about 2-3 iterations, because my soundscape has changed by then, so somethings may be lacking. However, for micro edits for basses and other super complex operations, e.g. neuro basses with crazy modulation, i tend to bounce out the vsti version of the track, and then post process that exported file even more. It depends on what you want to focus on, a composition you are more pleased with, or a track that is well mixed but might be lacking elements. From my perspective, the composition is fundamental, as your mixing skills will improve over time! Anyways good look with your mixes!
BTW: 24bit or even 32bit at 48kHz is my preffered export due to audiophiles wanting quality (services like bandcamp and masterers like this :D)

 

i usually do a light mixing while producing so i'm sure that everything sound already good from the start. something like pick up a kick and a base eq them and sidechain. once they sound good i move to lead eq sidechain(if needed) and start sending them to reverb fx channels and delays. once the production side is done i move to proper mixing: adjust eq here and there, apply compression and bus compression, stereo imaging, reverbs delay. Once i'm satisfied with the result i do a quick master and start playing the track from ipod to car, everywhere i can.

VSTi: When i was starting with FL and doing the first track i was able to not bounce steams, because i wasn't using too much vsti.

Now i use tons of vsti + vst effects but is not a problem because i'm using Studio One wich makes thing easier. i usually transform midi to audio when i'm happy with the sound i'm looking something like bass, lead,arp and then trasform, riser pianos and other cool stuff and the trasform again. in this way i can reduce my cpu usage and in anytime i can switch back to midi and start editing again. the daw trasform the file @32 bit,44100 hz (wich depends on song settings) working @48000 the highs are slightly better but increase the cpu amount.

when i finish mixing a trasforming a track a usually hit something like 60-65% of my i3.

if you are using other daws check out freeze or transform cabapilities, i friend of mine just bought a cool i7. if you can get it because sometimes avoid you to bounce file at all

 

i depends on you Computer CPU Power i Mix with all My Virtual instruments With My Mac Pro all i Bounce is the is the Finished Stereo mix then i Master the Stereo Mix separate in T-Racks of some times in my DAW

 

This does depend on a couple of things I guess. First the genre. With electronic music a lot of times the mixing is part of the arrangement and even the sound design, so you really can't separate the two. In other more acoustic genres you can probably get away with bouncing down your dry stems and then start mixing from there.

I personally mix as I compose and arrange. This helps me get a feel for how things everything is going to sound like when I'm all finished and also gives me space to get creative as I go alone and a lot of times helps me finish an arrangement faster.

If you are going to bounce all the stems then 24bit/44100 is more than enough but make sure you use dithering if you're bouncing something down that's different than those specs.

It's all a matter of personal perferance but yes for me I do find that mixing as I compose and arrange is the more efficient way to go.

 

- Do you bounce your VSTi into audio tracks before mixing the track?
if my i7 can handle it, i dont bounce, but if its a big project, ill bounce any "hungry" track (like any track with kontakt )...

- What is the audio spec for those files?
if i bounce something, will be at 24 bit/44.1 kHz...

- Do you record the VSTi signal to audio without the VSTi internal effects (e.g. reverb, delay, etc.)?
maybe, if the fx fit in the mix, and sounds good, why complicate things? deactivating and recreating the same fx (which already sounds good) in another vst, running the risk of dont get the same feel... so, if the internal fx sounds good as hell, dont touch it, but if i cant get what i want with the internal fx, ill deactivate it and put my fx in the mix console...

-What are your thoughts about this topic?
In my personal oppinion with this topic: "some people say bounce its better, some people says direct mix the vsti its better , so i let my ears and brain decides whats best... if i like what i hear..."

PS: before i updated my pc last year, for like 5 or 6 years ive been working in a core2duo with only 4 gb of ram, so bounce was a MUST, not a maybe