randy's Recent Posts
I have definitely gotten soundplane + sumu working over MPE. I know the soundplane client is in need of a refresh. It might need a certain startup sequence but should work. If you want to email me at support@madronalabs.com with the specific things you are trying, I can try to reproduce.
Yes, for just over a year now (Sumu v1.1, August 2025)
I have just finished a new menu UI for the presets and scales, it's coming out with Aalto and Aaltoverb very soon. Then it will make its way into Kaivo and Virta updates. Thanks for your interest!
I'll definitely be working on some DAW integration things like this. When synced to host and playback is stopped it makes sense that the sequencer position should always be at 0 (+ offset). Thanks for the feedback.
Hey, are you planning to add a black skin option for Aalto 2?
If not in 2.0.0, then soon.
thanks @ignatius! it's not too bad up here. stay cool.
@kunoichi yes, quadraphonic as in four outputs.
Will there be Linux support on day 1? Thanks
I don't know yet.
Getting close to beta!
That's fine, there's a lot to take in!
WAVs are sent to the engine in the highest resolution possible, I believe the files can contain floating-point (~24 bit resolution) as well as 16-bit and 24-bit wav. The wav files can have up to 4 channels.
Indeed, storing a multisample as an exciter at varying velocities is one of the main reasons I wanted multiple channels in the granulator. There are a lot of ways to use it.
Release and the other time dials on the envelopes are logarithmic. look at the numbers when you are dragging the dial- 0.01= bottom, 0.5 = half way, 20.0 = top. so, what you are seeing makes sense in this light: 1 is vertical and 1 + 1 = 2 which is about one o' clock.
Glad you're enjoying Kaivo!
I'm not sure what you are describing, can you send me a screenshot?
best would be to email it to support@madronalabs.com.
yes, we'll be beta testing Aalto 2 soon. Hopefully before the summer is over. But this is only a rough estimate- as you can see we're always wrong.
I haven't had a chance to dig into this, and I want to double-check before saying anything for certain. But what I think is happening is that the model itself in some modes has a discontinuity that makes a small click at certain pitch transitions. You could probably verify this by connecting the LFO via the patcher in place of the modulation sources you are wondering about. This will rule out MIDI things and provide a smooth source of modulation. So if you're still hearing clicks it's definitely in the model.
No worries! I appreciate the feedback.
Thanks for the feedback— all the modes have somewhat different settings to compensate for the number of bands. It might depend on the particular signal you are sending in. It's been a while since I designed these and I'm always learning things so I'll put it on my list to give this some fresh attention.
I guess you mean Virta not Kaivo in your other question? If you send AUDIO/env out to OSC/level in, the env is the same for all the bands, basically you're making an envelope follower that turns up that oscillator level. All the vocoding in the plugin happens in the FORMANTS module. "vocoder, nice 4 voice" is probably the most "classic vocoder" patch that comes with Virta and a good starting place. It shows a path with AUDIO/comp out to program, and OSC 1 out to carrier.
The trouble is there's no "?" parameter really, you can think of it as a button internal to the synth, that sets random values for all the sequencer parameters. I'll think about whether there's some way to enable your (very brave!) use case with scripting or something.
We're adding more control over the reverb / output section. I'm not quite understanding you though - in Aalto you mean reverb at 0.01 is too loud? I can barely hear it at that setting (something like -40 dB). You can also set a lower value than 0.01 with the fine adjust and see it in the DAW's generic plugin interface if you want to.
Thanks for the report. What I'm hearing is that during the recording the problem happened 2 or 3 times and each time it's the last note in the sequence of 4 that does not play.
We'll be releasing Aalto 2 soon which has some internal code changes to the sequencer. Unlike in 2010 we are making automated tests for correctness and I'll add one to cover this situation.
In the new version, there is better checking to make sure the number of partials in the import does not exceed what Sumu can play (64). The old version would fail silently. So, Sumu is helping you (I hope).
When Sumu imports partials, it outputs a manifest file to .../Madrona Labs/Sumu/Partials/import_log.txt. This file looks like:
cellos/new-cello.utu, OK, written to Users/randy/Music/Madrona Labs/Sumu/Partials/cellos/new-cello.sumu
randy/huygens.utu, OK, written to Users/randy/Music/Madrona Labs/Sumu/Partials/randy/huygens.sumu
temp/trolley-service.utu, failed, partial limit exceeded
temp/xxx.utu, OK, written to Users/randy/Music/Madrona Labs/Sumu/Partials/temp/xxx.sumu
so you can see where the .sumu files were written to, and which files did not work and why.
Yes Aalto 2 will be preset compatible. It will not replace Aalto in projects — it will be a different plugin "Aalto 2" that loads the same patches. Otherwise people will install it and be surprised that their old documents don't render exactly the same, etc.
We're hard at work on Aalto2 and will release as soon as it's ready. We won't ask for an upgrade fee if you've recently bought Aalto 1.x. I'm shy about saying release dates before we're at beta but it shouldn't be too long—please stay tuned.
It's a commonly asked question so I'm copying this info to a new thread in hopes it will be found more easily.
The Vutu app makes .utu files. These are a plain-text, editable, JSON format containing partials maps.
In Sumu, you can import the .utu files. The import process creates .sumu files, which are compressed to save space. The import also does some more analysis and saves it in the compressed file.
To import .utu files in Sumu, click the [...] in the Partials module. There's just one choice in this Popup menu: import partials. Then select a folder to import.
When you import a partials folder, it imports the whole directory tree underneath the one you pick, including any folders that contain .utu files.
So if I have on my disk
~/VutuFiles/Strings/cello.utu
~/VutuFiles/Strings/viol.utu
~/VutuFiles/Noises/tinkle.utu
~/VutuFiles/Noises/harsh/blender.utu
and then bring up the import dialog (partials/...) and select the folder ~/VutuFiles to import,
the files
~/Music/Madrona Labs/Sumu/Partials/Strings/cello.sumu
~/Music/Madrona Labs/Sumu/Partials/Strings/cello.sumu
~/Music/Madrona Labs/Sumu/Partials/Noises/tinkle.sumu
~/Music/Madrona Labs/Sumu/Partials/Noises/harsh/blender.sumu
will be created, along with the directories on the way.
Importing again will (for now) overwrite these files.
The intention is to "sync" your entire partials development folder at once. So if you make a folder somewhere called "VutuPartialsForSumu" or something, with everything you want to import into Sumu, and always select it when you import, you will keep your partials organized based on that folder's structure.
-Randy
I've gotten a few requests for standalone versions. I don't have anything against it but it's not super high on the list to be honest. From my end it would be one more entire "product" to deploy and support that not too many people are using. I really have to focus, in these aspects.
For the lightest possible use I recommend a simple free VST host. There are some for both Mac and Windows that basically just give you a window where you can hook plugins together with MIDI and audio I/O. I use JUCE Plugin Host on Mac (Claude can probably help you compile it if you can't find a downloadable version) and VSTHost for Windows—there are some other choices as well and probably more recent options.
Thanks for the nice feedback!
If you want to contribute some feedback about errors or whatever that's always appreciated. Best for us is to send an email to support@madronalabs.com. Thanks!
There's no way to do this now. I'll add this as a feature request then. :-)
If you feed Vutu stereo samples does it collapse them to mono?
Yes. Sumu partials maps are all mono.
Is there a difference that affects translation regarding sample rate or bit depth into Vutu (or is it open to just whatever you throw at it)?
Above a reasonable sample rate like 16 bit / 44100Hz here will be no difference. I just took a look at the code and it looks like Vutu resamples to the current system sample rate for playback, then analyzes the audio at that sample rate. That's a bit weird but again, there should be no real difference since we are converting to a rate-independent format.
Is there a limit to the shortest time or longest time Vutu can process a sample (or is it open to just whatever you throw at it)??
I think it gives you up to 60 seconds?
Do samples benefit from being normalised prior to import (does this drive Vutu’s engine more optimally)?
the partials maps get normalized on output, so there should be no need to normalize your sample first.
Thanks for going over the docs first. These are good questions probably not covered anywhere else. I'll get this info into the Vutu docs!
In the previous newsletter I sent out a customer survey for the first time ever.
Thanks to everyone who took the time to respond. I was truly overwhelmed with the amount and the positivity of your feedback. This work can be isolating sometimes when I'm typing away here in my half-basement and thinking about new sounds and new UIs, so let me get a little mushy on you and say from deep in my feels: your kind messages gave me more certainty that I'm doing the right stuff, and more energy to continue on the path.
I have some tangible results to share that might be of interest to you too, so let's dive in!
Aalto being our most popular synth jibes with what I've been seeing over the years. It's something special and I'll work to keep it that way. Kaivo and then the new Sumu come after that, and Virta and Aaltoverb a ways down. again, no surprises.
This is cool because along with the first question it gives us a (bought / used it recently) ratio, which is the best quantative measure I can think of for “did I make something useful?” Aalto here is a perennial favorite, and Sumu leads the pack—though that's skewed of course because it has not been out for that long. Virta is looking the saddest and that gives me some ideas for how to make it more accessible. Honestly I think the “it's a synth, it's an effect too!” approach makes it a bit confusing. That’s something I can work on.
Let the OS wars begin! (seriously tho, ack, no no no let's not.) Um, I work hard to make good interfaces that are sometimes opinionated and design-forward, and it makes sense this would connect with a Mac crowd. Windows is well-represented though, and definitely not a second-class citizen here. Honestly given the missteps made by both Apple and Microsoft in software over the last year (UI eye candy nobody wants, forcing adware and AI slop on us, ignoring fundamental usability problems), both companies are on my shit-list right now. So to that 4% of Linux folks, I completely get it and we're making slow but real moves in your direction.
Ableton Live? Never heard of it, I'll have to try that one someday. Just kidding, it's my daily driver and it's no surprise that's true for a lot of you as well. Bitwig beating out Logic for second place is a bit of a surprise. I'm happy to see our affordable weird cousin Reaper rounding out the top 4.
This was interesting. Combined interest in new effect plugins is about twice that of interest in a new app. Now that doesn't mean we will take one course and not the other. (¿Por qué no los dos?) There are good reasons for making an app even if it's only the top pick of 25% of our customers. But honestly I guessed these numbers would be the other way around. Amazing what you learn when you ask people!
As far as the quantitative data go, this one makes me feel good about letting our freak flag fly. Sorry about the formatting here, but you can get the idea. Novel, experimental effects are in the lead, followed closely by other kinds of tools that could frankly also be in the novel/experimental category. What there's not much interest in is "Established effects with that Madrona Labs sauce." I love all the software bebs, including Aaltoverb, but my understanding now is that we should keep fishing for ideas deeper down in the ocean of new and different sounds. It's kind of what we do, and I'm happy to see it's why you're here.
Thanks for the Modartt mention, I've been thinking along similar lines. The coding part should not be difficult.
Thanks for the note. People keep asking for iOS versions of Aalto etc. over the years. Given the pricing model there I've been slow to figure out how we can make it work but I'd like to figure something out.
Klauss! <3
Happy to hear you got it working.