MM as an FX mixer

Hi MM users, would someone try a patch and let me know the CPU usage before I pull the trigger on this thing?
I would like to use it as a mixer for external synths : 4 mono synths and 1 stereo drum machine into the MM audio inputs, with stereo FX for each track (individually or as sends). Ideally with the pressed duck that I really like! (I would control everything with midi from Octatrack)
Can it run a patch like this, and if so, how much power left? can it run a plaits clone along with it? what would be the latency?
I know I’m a dreamer, this patch posted here from 4ms only runs at 24Khz apparently, but I’m still curious

Thanks!

First, we are on a Modular Level here—no line! For proper Gainstaging, you need a preamp in your mix. I think something like 1010 Bluebox is the better solution for your case of use.

cheers
Mike

You’re right, I have to check but most of the synths are semi modular MFB/Behringer maybe I wouldn’t need a preamp
Good suggestion for the bluebox, but I tried it and was not convinced by the FX. I could have added pedals but I like the idea of self contained FX/Mixer box that’s modular. I like also the variety of FX on the MM, if but again, if it can run a few with the pressed duck.

Leaving aside the gain-staging question: If you’re using the effects as sends I’d expect this to work well, since you’d only need to run a single instance of each effect, along with a bunch of audio mixing. Having multiple insert effects on all five input channels would might start getting expensive, although it would depend on which effects …

Note that line level signals can be boosted to modular level inside MetaModule using virtual modules such as Airwindows PurestGain.

2 Likes

I created a basic mixer / FX patch as per the OP specifications. (ignoring the Gain Staging - it works in any case)

FiveDuck1.vcv (1.9 KB)
FiveDuck1.yml (13.1 KB)

Works great with the following caveats:

Usage:
24K/256 55%
32K/256 77%
32K/128 77%
48K/256 overload

Going down to three Airwindows Galaxies allows 48K operation.

I set the 5 Airwindows Galaxy Instances to Monophonic Processing and Block Size 64 Lowest CPU.

Whether 32K works for you sonically is up to you but the Octatrack is a bit crunchy in any case.

@Dan, having created this patch I realise that I can’t load a different Airwindows preset (its greyed out) into
each instance on the Meta, meaning that I would have to delete,load,rewire a new instance into the Meta each time I wanted to which seems cumbersome. Is module preset loading via the spanner/screw driver icon something that would be easy to implement in future or is this a limitation of the Airwindows platform?

All using the latest v2.12.12 beta btw

Magnificent. I may have seen your FiveDuck1 and raised it to SixDuck1 (using expander) Given the dual-core nature, not surprisingly it sits at 79% 32K/256 and declined to do anything duck-like @48K. I did not touch your Galaxy settings
SixDuck1.yml (16.0 KB)
SixDuck1.vcv (2.3 KB)

Thank you all for the help! The community here is awesome it’s a very good point in favor of the MM :slight_smile:
One thing I’m quacking about this Fiveduck/Sixduck patches : Is sending actual audio to inputs raises the CPU usage?
I also tried some alternatives in VCVRack, based on the CPU test by module thread, the 4msCompany StMix is way less CPU hungry so I guess I could add more FX for sound coloring and maybe a simple compressor for sidechaining
I would probably also remove the stereo drum machine from the equation and process it in the octatrack, so that would be only 3 or 4 mono synths going in.
It seems feasible and it looks like I could add some internal sound generation at the same time like a plaits with a filter or a drum voice…
That would be a neat live setup imo if that’s possible so if anyone thinks it’s not please let me know!

There are two dated spreadsheets of module performance titled “CPU Test” which details performance under various loads which is helpful. Worthy of note is that between the two tests conducted a lot of modules have seen performance improvements.