Madrona Labs Sumu

[dayjob]

Does sumu only support spectrums made with the Madrona analysis tool? Can it be fed with wav files or fft Data generated elsewhere? I like everything about this synth as an idea. Also is the filter "polyphonic" ? Like one for each FM pair/ noise? Where in the signal path is everything summed?

you have to use Vutu but it's dead simple and offers enough flexibility in its interface to give options for getting good partials from files.

this is in the Sumu manual btw
https://madronalabs.com/media/sumu/SumuManual.pdf

worth watching Randy's Vutu walkthrough video too for how to make choices to get good results for whatever your goals are.

edit: also from the manual -

Each Sumu voice is made up of 64 independent pairs of oscillators with FM and noise. Every pair has its own lo-passable gate and its own motile position in sonic space. This gives you lots of sonic potential, but it may sound like a lot to wrangle.

OUTPUT
The OUTPUT module is where all of your audio comes together be- fore venturing out into the wilds of your DAW. It gives you control over the final output level, with a handy limiter to rein in roaring mo- ments. It also contains a handy oscilloscope, which gives you visual feedback about the sounds you’re making.

i suggest you join the madrona labs forum on the madrona labs site and ask there about super specific questions not answerable by the manual. there is no circuit diagram in the manual

[Ah_Dziz]

I haven't given it a go yet at all. It won't run on the system I'm using most of the time. Just curious. It's a neat concept like all the ML synths.

[foosnark]

Also is the filter "polyphonic" ? Like one for each FM pair/ noise? Where in the signal path is everything summed?

Spaces combines the 64 oscillator pairs into a stereo output. There's a single stereo filter per voice. (I was hoping for 64 bandpass filters per voice, but on reflection, that'd probably be an absolute murderer of CPU on top of everything else it's already doing).

[kcrosley]

Oh hai! It’s that KVR guy who doesn’t like Myth! But Sumu doesn’t suck! Check it out!

https://youtu.be/U4zVIqyCQKM?si=Sw25rZBdnO2C79I-

[kraster]

Does sumu only support spectrums made with the Madrona analysis tool? Can it be fed with wav files or fft Data generated elsewhere? I like everything about this synth as an idea. Also is the filter "polyphonic" ? Like one for each FM pair/ noise? Where in the signal path is everything summed?

The only way to get sounds into Sumu is via Vutu.

Vutu uses the Loris Time-frequency assignment Library by Kelly Fitz and Lippold Haken but Vutu produces the proprietary format for Sumu to use.

[kraster]

New Beta update for Sumu available now.

ui: fix bug if Scope was drawn before resize (crash on launch)
ui: fix collectionPicker bug not showing last few files
ui: fix excess drawing of Popup menus (caused slow UI)
ui: fix too-big window on creation
ui: remove mouse reposition on windows pending fix
pulses: allow turning bpm down to 0
demo noise: half as loud and half as often
core: fix recent mod osc sound bug
core: fix AM sound bug
core: fix beta expiration bug
patches: set all to 4 voices
patches: fix inaccurate Fargo quote

Also: "If you were trying to run Reaper, Cubase or Studio 1 on Windows, thanks for your patience—the problem affecting those DAWs is fixed."

[masterhiggins]

Does sumu only support spectrums made with the Madrona analysis tool? Can it be fed with wav files or fft Data generated elsewhere? I like everything about this synth as an idea. Also is the filter "polyphonic" ? Like one for each FM pair/ noise? Where in the signal path is everything summed?

The only way to get sounds into Sumu is via Vutu.

Vutu uses the Loris Time-frequency assignment Library by Kelly Fitz and Lippold Haken but Vutu produces the proprietary format for Sumu to use.

Shouldn't that functionality have been added to sumu?

[dayjob]

Does sumu only support spectrums made with the Madrona analysis tool? Can it be fed with wav files or fft Data generated elsewhere? I like everything about this synth as an idea. Also is the filter "polyphonic" ? Like one for each FM pair/ noise? Where in the signal path is everything summed?

The only way to get sounds into Sumu is via Vutu.

Vutu uses the Loris Time-frequency assignment Library by Kelly Fitz and Lippold Haken but Vutu produces the proprietary format for Sumu to use.

Shouldn't that functionality have been added to sumu?

it's a free app. i think there would be a licensing issue to get that library working inside Sumu. or randy would have to reinvent the wheel to make the same functionality and it seems that wasn't the way to go. shrug

so, the functionality is part of Sumu but requires a free app that does that one thing and offers a lot of flexibility. it's not a hassle to me. i kind of enjoy spending a bit of time and making a bunch of partials and then importing them to see what i can do with them. i think it helps me not get bogged down. also i think adding all of that into Sumu would have been a gui mess.

theres info on the loris library on github

https://github.com/madronalabs/vutu

also,
https://youtu.be/crvpsnbLtt0?si=jhdrtPsabK8wachZ&t=1

Edit: there’s a forum at madrona labs where randy participates. So, you can always go over there and ask him about whatever

[mcbpete]

Shouldn't that functionality have been added to sumu?

it's a free app. i think there would be a licensing issue to get that library working inside Sumu. or randy would have to reinvent the wheel to make the same functionality and it seems that wasn't the way to go. shrug

I reckon it'd be OK to bundle - A good president was set with PluginGuru's Unify which, although being a commercial product, is bundled with a whole load of free plugins. *crosses fingers*

[kraster]

Does sumu only support spectrums made with the Madrona analysis tool? Can it be fed with wav files or fft Data generated elsewhere? I like everything about this synth as an idea. Also is the filter "polyphonic" ? Like one for each FM pair/ noise? Where in the signal path is everything summed?

The only way to get sounds into Sumu is via Vutu.

Vutu uses the Loris Time-frequency assignment Library by Kelly Fitz and Lippold Haken but Vutu produces the proprietary format for Sumu to use.

Shouldn't that functionality have been added to sumu?

I think that would violate the GPL licensing agreement.

Vutu is just a standalone implementation of Loris with a GUI.

Even the partial files it produces (.utu) need to be translated to Sumu's native format (.sumu) via import.

There could be more integration by say dragging a wave file into Sumu and it launches Vutu and then passes the result back into Sumu but I don't see it as being part of sumu anytime soon.

One thing I would say is that each partial file it produces is a wealth of content in and of itself.

Each sound can be up to 64 voices so harmonically dense sounds produce a huge amount of possibilities.

I've literally spent hours with one partial file producing all manner of weird and wonderful sounds.

[foosnark]

Licensing questions aside, having Vutu as a separate application keeps Sumu leaner.

[Shabdahbriah]

Licensing questions aside, having Vutu as a separate application keeps Sumu leaner.

It's also just super fun to play with.

[dayjob]

Shouldn't that functionality have been added to sumu?

it's a free app. i think there would be a licensing issue to get that library working inside Sumu. or randy would have to reinvent the wheel to make the same functionality and it seems that wasn't the way to go. shrug

I reckon it'd be OK to bundle - A good president was set with PluginGuru's Unify which, although being a commercial product, is bundled with a whole load of free plugins. *crosses fingers*

Bundle? It’s a separate download available directly from madrona labs. Is it a problem to download the app or something?

[mcbpete]

Bundle? It’s a separate download available directly from madrona labs. Is it a problem to download the app or something?

Of course not but wouldn't it be great if you could drag a wav directly onto the partials module then click on the 'three dots' on the top right to enter an 'edit' mode (effectively Vutu but within the Sumu gui) and then go straight back to the main Sumu ui with those settings. (instead of loading up Vutu separately, opening and editing the partial from a wav, export a json, load sumu and import the json)

[dayjob]

Bundle? It’s a separate download available directly from madrona labs. Is it a problem to download the app or something?

Of course not but wouldn't it be great if you could drag a wav directly onto the partials module then click on the 'three dots' on the top right to enter an 'edit' mode (effectively Vutu but within the Sumu gui) and then go straight back to the main Sumu ui with those settings. (instead of loading up Vutu separately, opening and editing the partial from a wav, export a json, load sumu and import the json)

sure. but it is what it is and works just fine. i don't see it as an inconvenience and i enjoy using vutu in a partial making session. but i'm big on different 'modes' of production process and like to focus on different things at different times.

but you should go to madrona's forum and ask randy about it. i'm sure he has a lot to say about it. he doesn't seem like a guy who does stuff w/o careful consideration.