There’s nothing preventing any of those: Schwung uses SysEx already for some internal messages, and OSC support exists via the schwung-jack driver used to drive the RNBO overlay. You probably have two options:
An Overtake module (Schwung/Move are running, and it runs “on top”, taking over the UI and control over all midi).
A standalone tool. There is support for standalone tools that kill Move and launch a process, like the native RNBO runner experience.
In both of those cases tho it’s less about “does Schwung feature these protocols” and more “there’s room for you to build whatever support you want”!
This is amazing @charlesv ! What is the recommended workflow to get Schwung tracks back out to a DAW? Would it be possible for Song Mode to output multiple audio tracks?
You are hijacking the thread, exact reason why Charles asked you to pop to Discord. Troubleshooting something over forum DM’s is not the same as troubleshooting something over Telegram / Discord or whichever real time chat app. It just isn’t the same and if developer asked you nicely to use that channel for troubleshooting bug why are you being annoying about it? You don’t have to use Discord, you don’t have to use Schwung, Charles doesn’t have to troubleshoot something over more clunky message system because you feel entitled to. Discord exist as web service, it’s matter of opening it in another tab, but I guess it is too much to ask from you, because…yeah.
use the quantized resampler and map to drum pads – those will come with you to live (My recommendation, but is very non-naive feeling)
use the Schwung -> Link setting to output Schwung slots over Link Audio to Live Beta
Just bring the midi clips in and recreate your instrument chain in Live
My preference: treat these like you would an external synth; mute the other tracks and track out the audio over whatever way you get audio to live, then bring it into your Set
As another old-skool forum head, I also loathed Discord when I first used it as it reminded me too much of IRC. So I can empathise about wanting to avoid it.
But - as someone who has vibe-coded a free software application in a completely other field that has very rapidly boomed and now has more than 100 support queries a day, Discord allows me to use deep hooks and automation to funnel and triage requests, and then overlay that with AI agentic support, automatic bug reporting, investigation and fix coding via my Claude Code instance and Github all whilst I’m able to sleep or get on with my day job. If I have to look in more places, and manually duplicate support queries over to another system that’s another layer of tedium and annoyance that I don’t need.
Personally - I don’t think it’s too much to ask users of free software to work with the community to centralise support, and Discord offers some very real advantages in this area.
Not wanting to stir up more ill feelings, but I think there’s a significant readership of this thread who receives your messages in a different tone than you seem to intend them / you think they’re being conveyed. I’m sure this reflects the challenges of written communication in an international, multicultural forum like this.
If you notice, I did try to help and said “for more support” to please join the discord. I see a forum as a place for discussion, answering questions (and yes maybe saying you had a problem), but diving deeper or one on one support not making sense in a thread. I don’t think it’s productive to continue to discuss this, either.
I’d appreciate if we can all keep things cool so this thread can stay open.
Large update to the Schwung Module Creator skill with more extensive knowledge and learning from building. Also incorporates karpathy llm wiki approach for improved memory per project and karpathys skill with behavioral guidelines.
@charlesv Curious to know how you port Ableton FXs into Schwung so I can update the Schwung Module Creator skill. Can you describe in detail how this is done?
This is a fork of another user’s module, i was just opening a PR to fix a json issue, so I don’t have the details on how they extracted the chord values beyond:
i (well, gemini) extracted the chords from expressive chords, and reimplented the controls
I’ve also already asked them to revise the description as this isn’t really a port, it’s more of a recreation based on observed values to get a similar effect.
In general, I’ve been able to get some things like built in fx values by doing memory observation while things are running on the Move itself, searching for biquads, etc. I don’t have any technique for porting native fx.
