Hi Folks,

Just thought I would ask my fellow musicians what their processes are regarging mixdown.

In the past I have always bounced my tracks down to audio, reset faders to zero and build my final mix - this was mainly down to habit and the fact I had to freeze most of my tracks due to low RAM. However, with the project I am working on now (with 16gb of RAM - goodbye freezing!) I have been mixing as I go. I feel the track is now ready to be exported as WAV to be sent out for mastering.

Is this wrong? Should I ALWAYS start afresh, converting my tracks to Audio and resetting the faders? Interested to hear your thoughts.

 

I used to bounce out, but too many times I would have to go back into the mix to fix things and rebounce, it became a pain. And there is really no reason NOT to just master inside the actual mixdown project, unless you are just out of CPU. Once I get my mixdown correct, I just make a set of new busses with zero'd faders and route my audio to them accordingly. Its like bouncing inside your project. lol. This way if you need to fix something that can only be fixed in the mix itself, you don't have to go through all that trouble.

*note - I work with mainly virtual in the box synths and samplers, so sometimes things have to be fixed within the midi programming itself, etc. However, if you are mixing tracked recorded instruments, it wouldn't matter since you really cant "fix" things other than having the band member replay the part, so either way, bouncing out doesn't make a difference once your in the digital domain. Basic rule of thumb is, whatever is easiest and quickest without losing sound quality is the best way.

 

Last edited: Nov 22, 2015

Cheers for that. I think what I was asking is - is it bad practice to export my mix as is (i.e. midi)? I have my faders just where I want them and just want to export now rather than bounce each track then re-do all my level settings.

 

Theres two schools of thought.
One is that mixing and production are two different and separate stages so you should commit therefore you can't backtrack (so like basketball where once you cross the half with the ball you can't go back).
The other is that, you can fix issues with the mix from the source. So for example instead of compressing to bring out little ghost notes of a piano, tweak the midi.

I prefer the latter because I'm a control freak, and I believe in getting things right at the source (although I hardly ever mic anything which is where the phrase originates from).
How ever I do change my mind state. When I'm mixing I'm done writing.

 

I think with modern tech and the ability to mix live, it is always better to mix live, as you can get more finesse then "attempting to polish a turd". The only thing you may want to do is bounce some major specific elements of a track for the phase alignment of tracks as a whole, but that check can be done in the mix live in most cases. This is my response coming from an electronic music standpoint. Otherwise your doing your fixes with a limited toolset; you cant get into the guts on your tune and fix the problems where they begin, whether is be midi alignment, velocity arrangement, eqing and comp fixes on specific elements. Only if your PDC fucks up because of a plug should you bounce out but then imo you are more consolidating rather than bouncing (kinda the same thing but small difference).

 

i mix during recording / writing. nothing too complicated, only eq-comp-saturation-sends, a nice compressor-limiter on the master to impress

 

That's why StudioOne's track transform frature is so great, you can go back in the previous state anytime, even "bounce to no track" a midi clip including FXes without setting up the new track manually first.
This type of workflow "freezing" speed/convenience should be made a standard among most DAWs.
P.S. sorry for the OT.

 

yeah I personally no longer see the necessity to bounce audio for mixing with the performance of computers these days...

I think it's also about your mixing technique and how efficiently you mix, how you place your tracks into busses and how you handle what gets compressed, processed and layered together in the composition...

I find with good technique, you can use less elements and therefore save CPU space as you can make alot from one well mixed sound :D

 

I always bounce all my VIs to audio before mixing anything, but then again, I often prefer to work with individual drum tracks and mono audio for certain things that would "really" be in mono (guitars, basses, kicks, snares, etc.). There's definitely a difference in sound. Now, for anything I do with a tight deadline, like film scores, I just bounce certain instruments and then mix everything else live. It doesn't sound as good to me as just mixing straight audio, but either way gets the job done.

Really, at the end of the day, is anyone listening to your mixes and saying "This doesn't sound as good as it would if you had bounced everything to audio first"? However you personally arrive at a good mix is the right way to mix.

 

For me, I do the composing, mixing, recording, mastering and final wav export in the same project because I have the ram and cpu!

Just do what you are comfortable with. A good sounding track is the most important. If you find your current way of working comforts you or thinking your tracks sounded better then I guess you go for it.

 

OR you could go the lazy way or the "modern way" of mixing. bouncing (fewer) audio "stems" With grupping things into 1 audio steam
(drums stem, bass stem, synth/keys stem, efx stem, vocal stem)

Anyway,
Depending on the project im working, i do a "killer mix" when im writting. mixing along as i go. more or less.
I aim at making it "sound great" - not "fantastic" or "perfect" - but good enough that i could send it off to someone else to mix or master it, more or less.

when completed, i bounce each instrumenttracck to seppeerat audio file, and startup a new projectfile build on my "Mix template" and i start working on mixing it (again) - but my mind is in "mixmode" whitch frees up my brain to focus on the big picture - The song/tune

Also, never forget the power of audio files. You can do so much more with it, than with a midi file.

 

I forgot for a minute there about bouncing my kick and bass to mono! Think I will just bounce down then mix the audio as I have done in the past. I suppose I was just looking to make the process a bit quicker, also was happy with the levels but it won't take overly long to get them back to around what I have now.

Now, onto a bit of a sidenote - in the past when I bounced to audio I bounced down with all effects and automation included and made sure each channel was -0.1bB. The only thing I took out was volume automation then reapplied that when mixing the audio. Do you guys bounce down dry audio then add the effects back in or bounce down with effects?

Edit: I should also add that I make my final mixes -6dB when I send my tracks for mastering. Just added that incase anyone thought I sent the complete Wav as -0.1dB.

 

Last edited: Nov 23, 2015

I do bounce my track multiple time.
I work with 32 bit 44.1 and it eat a lot of my small i7 and 6gb ram......so i have no choice.
Also ,i use a lot the cpu hungry plugin like Nebula and its textures librarys so i have no choice.
I bounce 4-5 time ( with seperate save file) during composition.
I bounce one mix
Then i bounce my master.

I would usualy use my Mix bounce as a 6 only different group where drum, perc and bass are mono and pads, synth and vocals are stereo .
Those would be the 6 file i send to the mastering engeneer + my own crappy master + a Flac copy ofa reference master im trying to achieve for a general idea.

This way its easy for him to correct and i dont have to send files when asking for correction.

I noticed over the years that my composer ears are biased when it comes to mix and it became obvious when reading notes regarding my mix and listening to tracks i mixed 2-3 years before .

Anyways, thats my workflow

 

I do everything in 1 project, most of the time I freeze the beat and the vocals ( I mainly do Hip-hop)
so when I go back I only have to reactivate the plugins of that part.

My master track is heavy enough to push my cpu to it's edge by itself,
I could higher my buffer when I got into the final mixing stage but I like my meter response

 

side comment on the WET/DRY issue.

I'd say it depends on what sound your working with and what element it "is".

also, it depends on how you use your reverb/delays. Are they on AUX channels or as inserts?

but when in doubt, always simplfy. Ask your self (simple) questions.

ala "is this Reverb sound thats on my synth sound crucial? (to the sound emotionel impact)

If yes, then bounce "as is"
if no, then bounce it dry,

or use time as your meter
eg. "i spend 1 hour making the perfect Reverb sound on my synth" verses - "spending 2 secs on slappying a 100%wet reverb on the track"

 

Have decided to just do what I have done in the past. Bounced each individual track out wet at -0.1dB. Forgot how long this takes! Each track taking 2 and a half mins and I have 40 tracks!

 

Believe it or not but some plugin elements respond better to (bounced) audio files. Compressors for example and peak control with limiters or distortion units.When a sound comes straight from an instrument (with midi notes playing etc.) and goes through a mixing chain it won't result in the same "tightness" in comparison with processed audio files. Not always noticeable in the mix of course but still. I 'discovered' this by accident btw when A/B'ing. It's also interesting to know that when you add plugins directly after the instruments (and you didn't check the VU inside the instruments) they will heat up the mixing tools which can result in small artifacts. When you do this with bounced stems the volumes are already lowered (you did that with the faders remember).

Dividing the production and mixing stage (and maybe go even further with groups and final mastering) works for me. But I have this crazy thing to create each track at least 2 times. I always start with a loop with all the main sounds in it. When creating the arrangement and adding automation and additional sounds I explore my ideas. There's also some essential mixing going on. Most of the time it ends up in a mess so when I think it's 'finished' I start over again and make everything descent from scratch - even with a nice layout - without the mixing chains. I'll do that with the bounced stems. I guess that I don't trust myself so it's also a 2nd opinion to see if I made any mistakes. And there's always room for improvements.

 

For me, it depends on the genre and, of course, the amount of tracks being used. As I mainly work in "Jazz/Hip Hop/Fusion", every single one of my compositions requires a different modus operandi. So, at times, "bouncing" tracks to audio is my preferred method.

 

I don't until cpu issues arise. I'm not sure about other daws but what I love about studio one 3 is the ability to convert a track to audio and it removes the instrument so as to free up cpu and then I can with one click, convert back to midi again and it reinstates the instrument and all settings.

 

Something is def. went wrong here, you are entitled to your own way of handling stuff. but something def. went wrong here.
I can get 100% phase cancellation in live, fl studio and cubase when comparing bounced tracks vs un-bounced.

you clearly did something wrong, or are using the wrong daw.
I strongly suggest to reconsider your findings.

I also want to ask the community to chime in on my reply, wrong facts can be a work-flow killer.