Printing VI's before mixing

[Simsy]

Hi

I was just wondering how many people here print their MIDI Virtual Instruments to audio before mixing? And how they go about doing that?

For instance, say you have a very large template consisting of 7 VE Pro's each holding 16 Instruments. Do you create a VE Pro with 16 outs and send each out to a audio track. Then print all the audio for each instrument before mixing the track?

Does this give any workflow benefit against either - just having a stereo VE Pro and controlling the internal instruments volumes via MIDI (CC7 or CC11) - or - having a 16 channel VE Pro and sending each out put to an aux track to mix / add effects?

I know that a lot of composers do their MIDI and scoring in DP but print all their tracks to Pro-Tools for mix down. I am not sure about the benefit of this personally, plus to get the audio into pro-tools would involve adding audio interfaces to two machines?

Any thoughts or shedding of light on the subject is greatly appreciated!

Many thanks!

[FMiguelez]

Hello, Simsy.

I ALWAYS print all my VI tracks. Not necessarily before mixing, though...

The way I have it right now is the workflow I've liked the most (after trying almost every possible scenario), and it requires a well planned template set in a special manner.

I can either, mix everything still in the MIDI stage, or opt to do it after I print the tracks. Either way, the result is the exact same, and this is the beauty of that workflow.

The VIs, being routed to audio tracks with input monitoring enabled, can use plug-ins at any stage of the production process, so I can opt to use MIDI CCs, plug-in automation, and regular audio track automation as I see fit (this works for heavily processed VI MIDI tracks when you want to deal with the finished sound as if it were "printed to audio" already).

That means that by the time I print my tracks, they can be already mixed, or I can mix them after I print them.
I usually print all the tracks when I know some director won't ask for changes, or when I'm happy with the currently sounding premix. I can always tweak it later, if I need to.

But my favorite thing about this whole workflow is that if you need to go back to MIDI after printing (as it's typical), it still sounds exactly the same, and any changes you do in MIDI will be already "mixed". All the MIDI tracks will sound identical to the printed audio tracks! All I have to do is press 3 again to re-print any changes in the track. It's a piece of cake to go back and forth reprinting tracks like this!

If you are interested, I could give you a couple of detailed links on how to achieve that (at least the idea behind it).

[Simsy]

Yes, I would appreciate any links you have.

Two questions though

- if you have the MIDI plus now an audio track for each MIDI track that doubles the session track count? How do you find your machine copes?

- also on a few films I have done I have had the budget to get it professionally mixed. So I had to print all the parts to send to the mix engineer. I found that my hard drive couldn't cope with that many tracks so I would print say 20. Then delete those tracks in DP and print the next 20. How do you find the hard disk usage?

[FMiguelez]

This is the link I told you about:
http://www.motunation.com/forum/viewtop ... 26&t=50004

Yes, I would appreciate any links you have.

Two questions though

- if you have the MIDI plus now an audio track for each MIDI track that doubles the session track count? How do you find your machine copes?

In a way, yes. Almost every MIDI track has its corresponding audio track where I can print it at any time.
This is where template organization becomes critical (hello DP folders and chunks).

But you could opt to "catch" in these audio tracks more than one instrument at a time, i.e., all flutes instead of subdividing them into piccolo, flute 1, flute 2 and flute 3. That's up to you. But I can tell you that the more "independent" or atomic your session is (separate audio tracks), the easier it is to mix and make changes of any kind. It's just much more flexible.

If you use VE Pro you could even do some submixing there, but then that would add yet another layer of complication, and I wouldn't recommend it.

I've found that when deadlines are tight, you better have everything set and ready, and it's better to delete stuff you won't use as you advance in the project, rather than looking for & setting sounds from scratch.

I used a similar template with a G5, and it was always a struggle. But now, with my new Mac Mini, I haven't been able to make it choke. Nowhere near it! It's a little beast! So smooth and reliable.

- also on a few films I have done I have had the budget to get it professionally mixed. So I had to print all the parts to send to the mix engineer. I found that my hard drive couldn't cope with that many tracks so I would print say 20. Then delete those tracks in DP and print the next 20. How do you find the hard disk usage?

I find that workflow a little convoluted and easy to make mistakes with.
It's faster to record-enable as many tracks as you will print (by clicking on a record button and sliding down) in one gesture, one time only, rather than having to keep track of what you printed, what you haven't printed, etc. And if there are changes afterwards?

These days HDs are so cheap. Get a dedicated and fast (7200 RPM) 2 TB HD for your DP projects and audio files and you won't even need to worry about that. Even if you print hundreds of audio tracks, that shouldn't be a problem. After you finish your project, just before you archive it (in a different HD, of course), you can delete unused stuff, or even use the Compact Project command, if disk space is a concern)

I just want to add, Simsy, that my recommendations are based on my own experience and idiosyncrasy, but I know for a fact that other well respected members would not agree or even like my method.
Make them share theirs, so you can get to compare them and see what you like more.

[Shooshie]

Sometimes I print my audio, and sometimes not. I often make incremental bounces to see how things are coming along. In those cases I don't print the audio, but merely bounce everything together, "live."

Once I'm through with MIDI, completely, I usually print the tracks and start treating it as I would any other audio recording. At that point I'll possibly do automation, effects, and maybe even some audio editing. By the time I'm ready to do a CD master, the MIDI has been long forgotten.

However, I have projects I've worked on in my spare time for years, or even new projects that require going back to the MIDI a lot. Those remain in the MIDI domain, and any recording I have of them was bounced straight from MIDI, often with effects and automation on Aux returns from the instruments in V-Racks. In these cases, I use Auxes rather than audio tracks. Personal preference, but sometimes recording is a problem if you've got audio tracks, and you have to keep up with which tracks are record-enabled, and all that. Aux tracks eliminate any concern about inadvertently recording or erasing track audio every time you record another MIDI track. It all remains in the MIDI domain until you're ready for the big switch to audio.

I don't have hard-core rules about any of this. Each situation brings its own challenges or short cuts, so I try to feed off the situation and do what seems logical at the moment. Also, I try to avoid ending up with folders of dozens of abandoned tracks. Sometimes audio tracks seem more logical. Other times Auxes do. Sometimes I just let the instruments in V-Racks output directly to the Main Outs, or at least to a Submaster track where I do the main mastering for the entire song.

Some don't like the word "Submaster" or "Sub" for what is essentially a master or stem track. I do. One reason is because DP has "Master" tracks which you can add to a chunk, and they are different from regular Aux tracks or audio tracks. I prefer to save the words "Master Track" specifically for those tracks. On a big board I probably wouldn't use the same terminology, but in DP it saves me from confusion, especially returning to a project years later, which often happens.

I haven't had dropouts from bouncing MIDI directly to disk since many versions ago. It's all been working pretty well for a long time now, but my instruments may be the reason. I've tried to eliminated Kontakt and some other VI's from my setup, though sometimes I still use them. If I'm working with instruments in my V-Rack in which I heavily use Control Point automation, especially synths and heavily modified samples, then I'm more likely to print those to audio before bouncing.

I'm sorry if this sounds jumbled and disorganized. My main points are that I've limited my primary VI selection to those that work well with Bounce to Disk, and that I try to be flexible. No method is off-limits for me if I understand its pros & cons. When I add complicated instrumentation (heavy on the controller automation in V-Racks), then I'm more careful about B to D, and more likely to print audio first.

Shooshie

[Shooshie]

A little background on Aux vs. Audio tracks for V-Rack returns:

Not too many years ago, if you wanted to listen to audio patched through an audio track, you had to record-enable that track. So, imagine that you've got 80 or 100 audio tracks acting as returns for that many instruments in a V-Rack. You have them all record enabled, and you're listening to your MIDI "live," when you decide to add another MIDI track. Of course, you record enable your MIDI track and hit "Record." Once you're done, you notice that you've recorded audio in every audio track!!!

So, in order to record MIDI, you had to disable recording in those audio tracks if you wanted to avoid recording audio into them. But... that creates a problem, too. Now you can't hear your MIDI tracks which were patched through those audio tracks! So, that led to all kinds of workarounds, but the easiest and most elegant was the Aux track. You could hear audio patched through an aux track without having to worry about any of the above.

Fast-forward to the present day, or at least a few years ago. At some point MOTU added another button to the audio track. It's an input monitor or "patch-thru" button which enables you to hear the inputs to your audio tracks, whatever they may be. That means you can now patch-thru your MIDI to audio tracks and monitor them without having to record enable the track. And monitoring your MIDI while recording another MIDI track doesn't create any problems with superfluous audio recordings.

This means that now we have much simpler choices. You can use an audio track to monitor MIDI, or you can use an Aux track. Audio tracks make it simpler, because when you decide to commit to audio, you don't have to add any more tracks. The Aux track, then, becomes much less essential to the process.

My hundreds of projects all still have Aux tracks in them, so I won't be leaving behind the Auxes any time soon, but I've found that I don't use them nearly as much as I used to in new projects. The input monitor button has eliminated the need for dedicated monitoring tracks. Just watch those record enable buttons!

So when DO you use an Aux track? For me, the aux is perfect for mastering. I want a track to which I can add effects and automation, but I don't want to record audio into it. Why? Because I'll be recording many such audio tracks, perhaps comparing them via SOLO, and I don't want to mess with their plugins. So, create one aux for a stem or master (submaster, I usually say, for reasons I explained in my previous post), and do one set of plugins on it, then produce as many audio tracks as I want from it.

The Input Monitor button on the audio tracks has radically changed what used to be the simplest workflow, and it has eliminated the need for an enormous number of superfluous tracks. Now, Aux tracks can function as they do on a real board, not as stand-in audio tracks for each input. (Audio input = MIDI output)

There... I hope that sounded a little more focused than my previous post, and a little more informative.

Shooshie

[Simsy]

Hi

Thanks for all your thoughts, it has been very enlightening! And I now fully understand why you choose audio over aux tracks, thanks Shooshie.

I have 7-8 VE Pro's all running on slave machines. I forgot to say that a lot of my work is more orchestral, so I am running a lot of Kontakt.

My VE Pros

- Woodwind
- Brass
- Percussion (with aux tracks split - percussion, timp, mallets, harp, piano)
- Hans Zimmer Spitfire Percussion
- Chamber Strings
- Orchestral Strings I
- Orchestral Strings II
- Synths

Plus I normally have a few other synths, drums etc VI's inserted.

I usually spend a lot of time on CC data (CC1, CC11, CC7) and get my mix balance perfect that way. If I need effects on a specific instrument I add it inside of the VE Pro.

I am just wondering -

- do you find that your mixes turn out better as you are solely focused on audio and mixing? Is this the real benefit of printing all your instruments?

- how much extra time do you think doing this adds to the process? Or is it in fact faster? As then mixing in the audio domain is faster than doing the final tweaks to MIDI?

The extra time is a bit of a concern if there is any. Most of my projects are so tight. I had a BBC series to do that I had to write 69 cues in 3 weeks.

Thanks for all the input so far it has definitely help me think about the process better.

[Shooshie]

Actually, I don't focus on audio unless I record a project in audio. When I do MIDI projects, I stay in MIDI until it's pretty much perfect (or more accurately: until I give up), then if I record it in audio it's for the purpose of mixing and mastering a recording. I view the two as separate acts: creation of the music and recording of the music. When I'm working in MIDI, I may have Audio track returns or Aux track returns from the V-Racks, but they're just returns; I'm not working in recorded audio. I have to leave the MIDI operational as long as it is possible, until I've got my phrasing, vibrato, timbres, etc. the way I want them.

THEN I may record it [print the audio] and mix and master in audio.

Shooshie

[FMiguelez]

Hi
- do you find that your mixes turn out better as you are solely focused on audio and mixing? Is this the real benefit of printing all your instruments?

Given the choice, I'd rather mix (or improve the MIDI mix) in audio. There's something about seeing everything printed that screams> This is progress! This is the official audio mixing stage. It's definitely less distracting and more efficient. This may be a psychological thing on my end, though.

But no. I wouldn't say that's even one of my top 3 reasons for printing the VIs.

- how much extra time do you think doing this adds to the process? Or is it in fact faster?

I don't know... How fast can you press a record button? That's as long as it takes me to record one or all my tracks at the same time, everything being pre-routed and ready.

As then mixing in the audio domain is faster than doing the final tweaks to MIDI?

Probably. It's definitely more elegant.

The extra time is a bit of a concern if there is any. Most of my projects are so tight. I had a BBC series to do that I had to write 69 cues in 3 weeks.

There's no extra time involved. On the contrary. It makes things faster, more streamlined and organized PROVIDING you have a great thought-out template.

[HCMarkus]

I find MIDI is great for getting parts just right - internal note dynamics, quantization/timing adjustments and klinker fixes - but Audio is usually a lot easier to work with when editing and mixing, especially if sustain pedal or continuous controller messages are part of the MIDI stream. That stated, a lot of MIDI parts I do don't get converted to Audio until the final mix. It's just a matter of doing what the project requires; different projects demand different approaches.

[frankf]

The thing that has kept me using auxes and not audio tracks is automation. Although I do a lot of MIDI CC automation, I often do section orientated volume and some panning in the aux track. I then print with this automation to an audio track. I've found if I place automation in an audio track, then print to that audio track I must remember to disable automation playback when listening back to the audio track(s) or I hear the automation doubled. FM and others, how do you deal with this?
Thanks,



Frank Ferrucci

[Shooshie]

The thing that has kept me using auxes and not audio tracks is automation. Although I do a lot of MIDI CC automation, I often do section orientated volume and some panning in the aux track. I then print with this automation to an audio track. I've found if I place automation in an audio track, then print to that audio track I must remember to disable automation playback when listening back to the audio track(s) or I hear the automation doubled. FM and others, how do you deal with this?
Thanks,



Frank Ferrucci

Well, if I record the output of a VI into an audio track, I'm recording the same thing I've been automating, pre-automation. That is, the VI's output is still original. That's what gets recorded into the audio track, and the automation then affects it in the same way that it affected the VI. Unless I'm not thinking right today, there should be no problem with that.

On the other hand, if I record a track summation through a Master track with automation and plugins, and if that summation then is output through the same Master track, THAT is a problem.

I usually go through my routing, literally drawing it with my finger on the screen, just to make sure I haven't done something like what you described. My new setups are a lot simpler than the ones I used to make, so it's not the problem it used to be. I mean, I used to have MIDI tracks, Aux tracks, and Audio tracks for every instrument and part. Then I mixed down to stems, and all those tracks were archived in folders. I've been trying to simplify these days, but I still use Aux tracks for some purposes.

One thing I like about aux tracks is that I can see all the automation in the Tracks Overview and Sequence Editors without any fuss over layers or views. It's very easy to copy/paste automation between auxes in the Tracks window. It just depends on the circumstances. I use a little from column A, and a little from column B.

Shoosh

[FMiguelez]

The thing that has kept me using auxes and not audio tracks is automation. Although I do a lot of MIDI CC automation, I often do section orientated volume and some panning in the aux track. I then print with this automation to an audio track. I've found if I place automation in an audio track, then print to that audio track I must remember to disable automation playback when listening back to the audio track(s) or I hear the automation doubled. FM and others, how do you deal with this?

When you record into an audio track with input monitoring enabled, whatever automation or plug-ins there are in the track do NOT get printed.
That is the reason this method gives me the best of both worlds! That's exactly what is so cool about it.

What I don't understand is why your automation gets doubled?

For instance. Say I printed an audio track for a piano solo part. Then I insert some EQ plug-in, and I automate its audio volume for a few sections.
But afterwards, I decide to go back to the MIDI track and change a couple of notes. For this, I simply enable input monitoring, correct the MIDI note, and re-record the relevant parts in that same audio track. The EQ and and volume automation I had done before are still there, and I can hear them, but they will not affect the recorded audio track as it gets printed (think of the signal it gets as "pre-mixer" or pre-everything). The resulting audio file will be 100% pure.
Once I disable input monitoring for this track (so I can hear the printed track and not the VI), it will be EQ´d and I will hear the volume automation, but only ONCE, and only because it's still there. IF I delete all that, the recorded sound remains pure.

Did I get your question right, or am I missing something?

[cbergm7210]

I work exactly how FM is describing. It's invaluable especially at the end when I am mastering the project and need to open songs to make small tweaks. I just open and tweak away without having to load all the VI's. (I host all VI's in VEP) Cuts down on work tremendously, and it is nothing to arm all the tracks and print VI's when I'm 95% done working with the MIDI getting ready to mix.

[frankf]

When you record into an audio track with input monitoring enabled, whatever automation or plug-ins there are in the track do NOT get printed.
That is the reason this method gives me the best of both worlds! That's exactly what is so cool about it.

What I don't understand is why your automation gets doubled?

For instance. Say I printed an audio track for a piano solo part. Then I insert some EQ plug-in, and I automate its audio volume for a few sections.
But afterwards, I decide to go back to the MIDI track and change a couple of notes. For this, I simply enable input monitoring, correct the MIDI note, and re-record the relevant parts in that same audio track. The EQ and and volume automation I had done before are still there, and I can hear them, but they will not affect the recorded audio track as it gets printed (think of the signal it gets as "pre-mixer" or pre-everything). The resulting audio file will be 100% pure.
Once I disable input monitoring for this track (so I can hear the printed track and not the VI), it will be EQ´d and I will hear the volume automation, but only ONCE, and only because it's still there. IF I delete all that, the recorded sound remains pure.

Did I get your question right, or am I missing something?

FM,
Thanks for the explanation, and now I understand your workflow and goals better, and I see how you don't double automate. When I print a track or stems here in a film music workflow, I may and usually do want to print the automation, plugs, etc, depending on how the film music mixer works. Or if there's no music mixer, the film mixer him or herself. Also depends on the time available to do the music or film mix and if I can be there as producer. Unless there is time to give to respect the music, I need some control to present the music as I hear it. The mixer can adjust the stems from there. Artist tracks I would not automate unless the engineer I'm working with ask me to, so there I can see how I can possibly use your workflow and will try it. Different strokes. And DP allow us to choose



Frank Ferrucci