Audio vs Aux tracks in template

[Chris T]

Does anyone know if Audio tracks use up more resources in DP11 than Aux tracks?

I have a large template in DP11.31 running many orchestral libraries, synths VIs and video. I'm running a maxed-out MAC Studio (M2 Ultra). I run a lot of VIs within DP but also have VEPro7 running on the same machine and on an external Mac Mini for my orchestral libraries.

I ask because I currently have all my "audio returns" from VIs / VEPro coming into Stereo AUDIO tracks (in "monitor mode") in DP. When I'm ready to Mix, I record-arm all the necessary audio tracks and print audio. I then mix the audio (as one would in PT) before bouncing to a stereo mix (or stems).

I'm considering altering my template significantly by using AUX tracks as my "returns" and having them route to Group Busses (followed by Audio "STEM" tracks). That way I can just have say 6 audio tracks and my "mixing" would be done using just MIDI and the AUX returns. (I'd also need to setup dedicated Reverb feeds for each stem group).

Any opinions on this?

I do like to have the option of mass-recording audio tracks and mixing audio, I just fear that it might be taxing my system and slowing down my workflow too much...

Thanks.

[bayswater]

The last time this was discussed here, the argument was the opposite — that aux tracks eat a lot of CPU.

I don’t know about the most recent OS and CPU versions, but my 2018 Mini can do well over 500 audio tracks with a load of basic plugins and well of 100 VI tracks. Its the Aux tracks with input and output busses that seem to bring it to it’s knees. I don’t know why, but my assumption is that when you route more than one bus into an Aux, a mixer operating in backgound is invoked.

There have been threads here from those with high end new Macs with up to date OS versions surprised when they see similar symptoms. I hope they look at the Getting Started manual on optimizing the settings that affect performance, keep the number of MIDI tracks per VI instance low, and other similar things we are advised to do.

[James Steele]

I think the issue with Aux tracks is that any plug-in inserted on an Aux track must run in real time. It cannot use pregen.

[Chris T]

Thanks Bayswater. Wow I didn't realize AUX tracks were more taxing than audio, but your theory about a "background mixer" makes sense. Do you happen to have a link to the previous discussion on this topic? (I tried searching...).

With this in mind, if I had say 500 stereo audio tracks which receive audio from the multi-outputs of my various Kontakt instruments and synths, you think that would tax my system LESS than 500 AUX tracks receiving input from the same VIs?

Both Audio or Aux tracks would be outputting to a set of Aux stem group tracks, which in turn would be routed to the final print stem (audio) tracks.

[HCMarkus]

One could bypass the "Convert all VIs to Audio" step prior to mixing and just go right to stem busses which feed audio tracks(for stem recording) and also fold down to the stereo audio mix...

That's what I've been doing. But my template is not nearly as expansive as Chris T's.

[James Steele]

With this in mind, if I had say 500 stereo audio tracks which receive audio from the multi-outputs of my various Kontakt instruments and synths, you think that would tax my system LESS than 500 AUX tracks receiving input from the same VIs?

I need to test this, and it would likely be easy to do in the Effects window where it shows use percent of CPU used by plug-in and whether they are running realtime or pregen, but it might also be that VI feeding an Aux Track is going to run in realtime instead of pregen? I need to check this.

[bayswater]

Thanks Bayswater. Wow I didn't realize AUX tracks were more taxing than audio, but your theory about a "background mixer" makes sense. Do you happen to have a link to the previous discussion on this topic? (I tried searching...).

With this in mind, if I had say 500 stereo audio tracks which receive audio from the multi-outputs of my various Kontakt instruments and synths, you think that would tax my system LESS than 500 AUX tracks receiving input from the same VIs?

Both Audio or Aux tracks would be outputting to a set of Aux stem group tracks, which in turn would be routed to the final print stem (audio) tracks.

I'll have a look and post the link if I find it. Whether you'll get the same result or not, I can't say, but it's easy to test. Just do a few tracks then duplicate them a couple of times. If there is a difference with 20 tracks there will be a difference with 500 tracks. I might try it tomorrow, but I'm on Intel with an old OS version.

One thing that you can do to keep the template using only the CPU you need, is to use a lot of instances of Kontakt or any VI, and put them in V-Racks. It would be unlikely you'd need them all for any one project, so you can just turn them off in the VRack in the template until you actually need them. And if you're satisfied with the sound from a lot of VI instances, you can print them, and turn off all the VIs and effects they use.

In my templates, I always return VI outputs to audio tracks via busses (no Auxes involved), and leave the Audio track Input Monitor buttons on (sometimes a blue button in the channel strip) so you can hear them without these tracks being in record mode. When I'm happy with a MIDI>VI>Audio track, I can put the Audio track in record mode and do a pass, then turn off (not just mute) the MIDI track and VI (and effects on earlier parts of the signal path).

[primeevolutionary]

Hello, Bayswater,

What’s the difference between putting the VI in V-Racks instead of in the Chunk (other than the ability to use the VI for multiple Chunks)?

This may be something I used to know and forgot. I only use V-Racks for a couple of things these days.

Joe

[bayswater]

Hello, Bayswater,

What’s the difference between putting the VI in V-Racks instead of in the Chunk (other than the ability to use the VI for multiple Chunks)?

This may be something I used to know and forgot. I only use V-Racks for a couple of things these days.

Joe

There are some differences in the way you apply automation, but the main thing is you can click the “power LED” in the chunks window to turn the VI off when you don’t use it, so it’s there, but uses no CPU. When you power it up again, it will take a few seconds to come up, but will have all the settings it had earlier. So, one of my templates has at least one instance of every VI I have, with a few patches I’m likely to use preloaded but “powered off”.

[primeevolutionary]

Do you use a separate V-Rack for each VI?

If the VI is still in the Chunk, when you click the Enable button, does that not achieve the same thing? I'm not challenging you here, this is just the way I've done it in the past. I thought enabling the VI turned it off. I'm eager to learn.

I have a super template that I used to use with VEP that had a MIDI channel for each VI and each articulation (EastWest) and each instance might use up to 16 MIDI channels per instance. How would you handle something like that using V-Racks?

I apologize if I've drifted off topic, but, the audio track vs aux track is leading me to rethink a new super template.

[HCMarkus]

Every Audio/VI track has a similar "Enable" button. V-Racks can include more than a single track and can be utilized across multiple Chunks within a Project but, unlike V-Racks, Audio/VI tracks can be automated.

In my template, I have each VI hosted in a Track; all Tracks start out Disabled, and I Enable as I need them. Like V-Racks, Disabled Tracks use little or no CPU.

"You pays your money and you takes your choice"

[primeevolutionary]

Thank you, HCMarkus!