unsexy answer from me:

midi means : "open options" (standing still)
bounce to audio : "comitment" (moving forward)

working with audio files has so many posibilities over midi, and you have so much more control aswell.

i too ussualy have 2 project files, one with all the midi still there, if i want to go back and make some major changes.

2nd project file is all audio, where i go in and do lots of edits and massage the arragement inplace before putting on the mixengener-hat on, and do that process with that mindset as focus.

on the question about "should i bounce the track with efx/sends on or not"
it depends. are you contend/happy with the sound with the efx on on? are they crucial for the track to get "that sound" ? if the answer is yes, then bounce the track with efx/sends on.
If the answer is "not sure, or no" then bounce the track "dry" - or make a dry version + bounce a audiofile with only the efx/sends aswell, then you keep what you've done, but can still manipulate the efx/send sepperatly.

 

i always mix with bounced stems for several reasons, already told by others here before

1. looks much cleaner to me in the arrangement window and my focus is full on the mix
2. have all my cpu ressources back and dont have to worry about glitches and cpu crackles
3. still can edit some stuff. depends always on style, but when i cut an audio file the sound is gone. if i would be still in arrange mode, some reverbs, delays or FX might still be fade out. with audio i can make good and stronger drop parts with this for example

so what im doing is, to make a real real rough mix in the arrangement part (set volumes to work with and get a feel how it sounds later) do most of the automation stuff, add fx and all that is needed to shape the sound (distortions, chorus whatever) maybe a bit of compression already when its needed by a sound.

then i bounce every track to a single file. reload the stems in a new project and give the mixdown the final volumes and shapes with compression and eq'ing. put all drums to a bus to do bus compression and maybe shape the drums with more distortion or something like this. same for lead synths...whatever you need. i still might do some fx stuff in this stage when its necessary.

in the end i bounce a pre-master of this stem mixing session which i open up in another project just for mastering.
so i have 3 stages for a final master and this way i will be able to control every stage in the best way (imo or workflow at least )

 

I'm not an expert on this topic so I read with interest.
What I've found in the past is that when I export from the midi project directly to WAV
it chances "the mix" At least in Live it does. Drums disappear in the background or bass that
was at the right level becomes over loud. Since then I've given up on ever producing a decent-sounding song.
Never came to my mind to make waves/stems first. Could that solve my problem?
And would it be done with effects turned on or of?

And maybe a little bit of topic but what does mastering actually adds to the final result?
Like the original poster I'm more of a composer so finishing a sounds so it can be heard by the rest of the world
has never taken place.

Thanx for the input.

 

Haven't been home yet.
Will test it tonight.
Oke, if exporting directly from midi to wave (as a complete) song should not alter the mix then I've contact ground-control since we have a problem.
I will try it with different wave files per instrument tonight to see if that makes a change later today.

I'll have to dive into mastering I guess.
I've used Izotope a few times in the past. With the mastering assistant, I can hear what is happening afterward (louder and brighter)
and although I see modules pop up (Like EQ, Limiter, etc) I haven't got a clue of why, what and how.

There is a long Easter weekend ahead of us so how now I might find some answers ;)
Have a Happy

 

Always bounce if youre in dance music. Post bouncing, verify is your drums, bass, sequenced synths aligned(zoom in to see the transients and the grid)? Pay atention especial to transients. Dont trust Daws. Doesnt matter for live recordings of rock, jazz or any non-perffect-alignment music(aka live feel).

 

I'll stick to writing then. Since by that time I'll almost be in my grave..... (lol)

 

You always think you can get decent mixes for the type of equipment and lack of room treatment that you have until you get a vocalist with a whistle in his voice. No matter wash you do, he gon have that whishtool when he shpeaksh.

 

No. I'm not bouncing unless I have to for reasonably cpu issues. Reason? There's a lot of modulation on most of my tracks that i need to be choice open on the last minute when mixing...

But I have no fear of bouncing if I need to do it. Its just that I don't have any extra benefits doing normally on my workflow.

Things could change tomorrow, so its just my opinion TODAY.