Questions about Bussing Midi to Audio

[cmrees]

I use East West's Play software and virtual instruments with Pro Tools 12, and I have a question about bussing multiple Midi tracks to audio. I have several Midi tracks linked to a single instrument track with Play loaded, and each Midi is assigned a different channel in the Play software. I want to be able to create an audio track for each Midi track in order to put effects on the sounds. The issue that I run into is that I am unable to bus an individual Midi track to the Audio track, as I have to bus the Instrument track to the Audio Track. Unfortunately, I have all of my midi instruments loaded on the single instrument track so I cannot place effects on them individually. Do I just need to create and new instrument track for every Midi track and open a new Play window in every single one or is there a way to specify which Play channel on the instrument track that you want bussed to the Audio track?

[Darryl Ramm]

Create a separate instrument track for each different instrument, stick the plugins you want on that track.

Or follow this multi-output tutorial... https://www.youtube.com/watch?v=KVVKspP4jpQ

What workflow is best depends on what your preference/what are doing.

[cmrees]

That video was exactly what I needed. Thanks a lot

[musicman691]

As Master Yoda would say: there is another (way). Say you have 4 instruments in Play being hosted in a PT Instrument tracks. One Play instrument gets it's MIDI from the Instrument track and the other three from three MIDI tracks. Go to the Mixer page in Play and on fader one have that output to Master and the audio for that instrument goes to the Instrument track. The assign fader two to output 3/4, fader three to output 5/6 and fader four to output 7/8. For the other three Play instruments create aux or audio tracks in PT; route their inputs to plugin/Play 3/4 for one PT track, 5/6 for another PT track and 7/8 for the last PT track. That way you don't need to have multiple instances of Play running and eating up system resources.

One thing to keep in mind with this process you may not be able to render out one Play instrument at a time with the offline method. I don't have PT12 so maybe that can do it. Using real time render you can do one Play instrument at a time.

People tend to forget about the mixer and routing capabilities in the Play engine.

[DonaldM]

How you do this will depend on what you are doing inside Play. If all the MIDI tracks are playing different voices of the same Play instrument (Goliath, say), then one instance of Play is fine. However, if you're using different Play instruments for each MIDI track, then definitely set up a new instance of Play on a separate instrument track for each MIDI track. Otherwise, you could have all sorts of issues. That's generally a good rule.

Assuming you're using one Play instrument for all MIDI tracks, you can assign the AUDIO output of each separate patch of the Play instrument within Play. Create a separate AUDIO track (stereo or mono, depending), and then select the corresponding input for the audio track from the correct Play output. Use the mixer in Play to control the input level for the audio track for each track.

When you've got your levels good, hit record and lay down the tracks, or, if you're using PT 12, you can just use the Commit function. Either works just fine.

[DonaldM]

Originally Posted by Darryl Ramm

Create a separate instrument track for each different instrument, stick the plugins you want on that track.

Or follow this multi-output tutorial... https://www.youtube.com/watch?v=KVVKspP4jpQ

What workflow is best depends on what your preference/what are doing.

Personally, I would not set it up the way he's showing in this video. I'd much prefer to have 1 instrument track (instread of the multiple ones he's showing) with the instance of Play, separate MIDI tracks for each part (ie strings, bass, etc), and then route the audio output for each part in Play to separate outputs to a PT AUX or AUDIO track for further processing. With PT 12, I normally just use AUX tracks to get everything sounding the way I want it, and then use Commit to create the audio tracks I'll use for the final mix.

I know he said use one instance of Play for multiple instruments instead of separate instances of Play for different Play instruments...but on the EW Forum, they were saying the exact opposite. I guess it depends on your system and how you have your sample streaming set up. My EW sample libraries are on an external HD connected via USB3, so its pretty fast and I've not had any issues with multiple instances for different instruments with Play.

[musicman691]

Originally Posted by DonaldM

How you do this will depend on what you are doing inside Play. If all the MIDI tracks are playing different voices of the same Play instrument (Goliath, say), then one instance of Play is fine. However, if you're using different Play instruments for each MIDI track, then definitely set up a new instance of Play on a separate instrument track for each MIDI track. Otherwise, you could have all sorts of issues. That's generally a good rule.

Assuming you're using one Play instrument for all MIDI tracks, you can assign the AUDIO output of each separate patch of the Play instrument within Play. Create a separate AUDIO track (stereo or mono, depending), and then select the corresponding input for the audio track from the correct Play output. Use the mixer in Play to control the input level for the audio track for each track.

When you've got your levels good, hit record and lay down the tracks, or, if you're using PT 12, you can just use the Commit function. Either works just fine.

Just for giggles I set up an instance of Play with 4 different strings from EWQLSO (using 4 separate instruments), 2 instruments from Ra and 2 instruments from StormDrum 2. Each fed by separate MIDI tracks except for the instance of Play which is hosted in a PT Instrument tracks. All routed in Play to separate PT aux tracks. Works just fine. Where it gets hairy is the Play gui doesn't change with the different parts being played but that's not a big deal. You just have to keep your wits about you when editing things in the Play engine.

I ordinarily wouldn't work this way but tried it just to show it can be done.

Originally Posted by DonaldM

Personally, I would not set it up the way he's showing in this video. I'd much prefer to have 1 instrument track (instread of the multiple ones he's showing) with the instance of Play, separate MIDI tracks for each part (ie strings, bass, etc), and then route the audio output for each part in Play to separate outputs to a PT AUX or AUDIO track for further processing. With PT 12, I normally just use AUX tracks to get everything sounding the way I want it, and then use Commit to create the audio tracks I'll use for the final mix.

I know he said use one instance of Play for multiple instruments instead of separate instances of Play for different Play instruments...but on the EW Forum, they were saying the exact opposite. I guess it depends on your system and how you have your sample streaming set up. My EW sample libraries are on an external HD connected via USB3, so its pretty fast and I've not had any issues with multiple instances for different instruments with Play.

My samples come off a Thunderbolt connected 7200 rpm spinner. I tend not to stream things but load everything into ram as much as I can and with 32 gig I've yet to run out of ram. Sure it makes for a longer session load but it's what works best for me. I could stream if I want to but I'd rather not (at least not with PT).

[DonaldM]

Originally Posted by musicman691

Just for giggles I set up an instance of Play with 4 different strings from EWQLSO (using 4 separate instruments), 2 instruments from Ra and 2 instruments from StormDrum 2. Each fed by separate MIDI tracks except for the instance of Play which is hosted in a PT Instrument tracks. All routed in Play to separate PT aux tracks. Works just fine. Where it gets hairy is the Play gui doesn't change with the different parts being played but that's not a big deal. You just have to keep your wits about you when editing things in the Play engine.

I ordinarily wouldn't work this way but tried it just to show it can be done.

Yes, it can be done. But I find it can be a bit confusing. My preferred way is a different instance of Play for each instrument.

Originally Posted by musicman691

My samples come off a Thunderbolt connected 7200 rpm spinner. I tend not to stream things but load everything into ram as much as I can and with 32 gig I've yet to run out of ram. Sure it makes for a longer session load but it's what works best for me. I could stream if I want to but I'd rather not (at least not with PT).

Now that you say that, that's what I do as well. Everything gets loaded into RAM first. But its really fast! I have 24gig, which has seemed to be more than enough.

[musicman691]

Originally Posted by DonaldM

Yes, it can be done. But I find it can be a bit confusing. My preferred way is a different instance of Play for each instrument.




Now that you say that, that's what I do as well. Everything gets loaded into RAM first. But its really fast! I have 24gig, which has seemed to be more than enough.

I don't remember - do you load off of a spinner or ssd? I'd go ssd but I need well over a terabyte and that ain't cheap in an external drive.

[DonaldM]

Originally Posted by musicman691

I don't remember - do you load off of a spinner or ssd? I'd go ssd but I need well over a terabyte and that ain't cheap in an external drive.

Its a spinner. West. Dig, Black, 7200 RPM, external, 1TB. USB3 connection.