randy's Recent Posts

No worries

Thanks for letting me know. I too have seen the older, JUCE-based plugins (that's Kaivo, Aalto and Virta) get slower over time. Meanwhile of course I've been working on Sumu (and Aaltoverb) and a whole new, GPU-based graphics framework.

I'll see if there is a quick fix for the behavior of the older plugins on newer DAWs. If there's no such quick fix, I'll update them to use the new graphics code. Either way I'll have a fix soon.

Here's a temporary link to the Beta discord (expires Aug 13, 2025):
https://discord.gg/TSzkZ2Ry

Please read #read-me-first and take some time to catch up on the recent conversation when you arrive!

Hi Dylan, thanks for the clear report and your update. I've chased similar issues with the body module and thought they were fixed. It could just be very rare now. If you find it's a specific body type that makes it stop, or anything else that makes it do that again, please let me know.

Thanks for the support! Humorous and otherwise. :-)

I have a release candidate up now, and I'm putting the wheels in motion. Please subscribe to the newsletter for the freshest news!

I understand it seems like a long time for this update, it's been a real battle for me this time. I've had to deal with low-level Windows API stuff, which is not my happy place, and honestly with unrelated life stuff, which I'm not going to vent about here except to say it impacts a one-programmer company a lot more than a big one. I hope we can both enjoy a more regular release schedule soon.

I just posted a release candidate to the beta list, so fingers x'd we will have 1.1 next week.

Thanks for writing in about this. This has been a problem off and on with some versions of Live, and I'm eager to fix it in all the plugins it effects, either by releasing the new versions I have in the works or interim updates to the v.1s. Meanwhile I apologize for the inconvenience.

There's at least one tutorial online— does this help?
https://www.youtube.com/watch?v=sUOsGtB0MDY

It sounds like you are doing things right. When Aalto is set to MPE mode, the channel aftertouch should come out of the "after" input to the patcher. Through the patcher, each voice gets its own aftertouch data depending on the channel.

You do need to set "MPE MIDI" explicitly in Aalto using the "Input Protocol" choice in the settings (gear) menu.

Been a long road this time but it's in beta finally!

Sorry I missed this- I've been cleaning up so much spam every day and we're working on a solution for that.

So there are other posts on the forums about how the import is done—I'll be working to make it more obvious in the future. I think you figured it out.

You can clean up and move the files around however you want. On Mac OS, partials are stored in (Your home directory)/Music /Madrona Labs/Sumu/Partials. On Windows, in C:/AppData/Roaming/Madrona Labs/Sumu/Partials. 

Hi Damo,

It's OK to post this here. But, this forum is not very busy. You may have better luck on for example KVR Audio, or on a Reddit synth forum where selling is allowed.

best,
Randy

Hi Chris, thanks for your patience, I didn't see this until now. Kaivo 2.0 will be a paid upgrade, unless you've bought the v1 recently, in which case it will be free. I haven't figured out the details yet.

We have updated all of our software instruments—Aalto, Kaivo, and Virta— to version 1.9.5, bringing native Apple Silicon support for M1 and M2 Macs. The new versions are Universal Binaries, which support both Apple Silicon and Intel processors. Users with Apple Silicon computers should be able to run 30% more voices or more, as compared with the previous versions in Rosetta 2 emulation.

This update is free. Installers contain Universal Binaries for both VST2 and Audio Units V2 versions.

Windows versions are unaffected by this update. Aaltoverb, previously released with Apple Silicon support, is also unaffected.

Adding popups to UI elements that move is going to be a little tricky, but it's not rocket science and I will definitely do it at some point. The popup menu can stay put while the "tail" follows the loop bracket.

I hear that the slow pace of bug fixes can be frustrating. I'm working to improve that this year. One way is by making it easier to deploy my software so that doing so doesn't take me away from other work for too long. (if you are a software person, think CI / CD)

I believe I am very close to shipping a 1.1.0 version of Sumu right now. Please stay tuned, and thanks for your patience.

Are your plugins registered? If not, that's why you hear the ocean sound. The demo versions play the sound until you register them.

Now that I think about it I should add a popup message when the demo sounds plays also, to make this clearer.

Thanks for writing.

Hi, welcome. Thanks for your support, and the nice words about Sumu.

I definitely hear you on this. There are a couple of way Sumu could be changed to work better for you. One is an MTS-ESP mode, which would take over and change the scale only under MTS-ESP control. I looked into this a couple of years ago and ran into some technical issues with MTS-ESP that I think have been resolved by now.

Another way would be a parameter lock. A number of people have asked to be able to lock various things when a patch changes and this could include the scale. I'll definitely make this possible at some point.

Thanks for writing with your feedback. I have good news: you can scale Kaivo by grabbing on the little triangle in the bottom right corner, then dragging the window to the size you want.

I am working on a dark mode that will provide more contrast and I'll include it in all the plugins when I update them.

You should not need to move or rename any files outside of the Vutu app or Sumu plugin. Renaming a file will definitely break things. (It shouldn't crash, though—I'll fix that.)

I'll go over the steps to turn sounds into .utu files and import them into Sumu.

We start with a sound in wav or aiff format.

• open Vutu and click "open" in the source row to open the sound.
• if desired, move the endpoints in the waveform display to select the start and end.
• if desired, click "play" in the source row to preview the sound.
• click "analyze" to analyze the sound.
• click "export .utu" to write an utu file to disk wherever you keep your partials files. Let's say it's in MyPartials/acoustic/violin1.utu.

Now we have an .utu file which is a JSON text format, you can open it with a text editor to look at the partials data.

Now open Sumu.

• click "..." in the PARTIALS module and then click "import folder..." in the menu.
• select the folder containing your partials files. In this case it would be MyPartials.
• the MyPartials directory and all subdirectories will be scanned for .utu files. All the .utu files will be converted into .sumu files in .../Madrona Labs/Sumu/Partials. So if you did not prebiously have a .../Madrona Labs/Sumu/Partials/acoustic directory, that directory will be created and the file violin1.sumu will be created in it.

The .sumu files are partials in a compressed binary format that Sumu can read.

So:
audio -> export to .utu using Vutu -> batch import into .sumu in Sumu.

Yeah I'm hard at work on Sumu using my primate brain and I didn't really know what to say about this. AI in the form of smart code completion like Copilot is something I'm already using and find helpful. More ambitious tasks like optimization and maintenance of this type of code are not things AI is going to be helpful with anytime soon.

I'm finishing up the beta of MPE support right now, and am also testing Channel aftertouch in both MPE and MIDI modes. I'll send out the newsletter with a beta link as soon as I post it.

Aaargh, you probably guessed but I meant to type 256 not 25.

Thanks for writing. Aalto and Kaivo 2.0 will be coming later this year. Definitely with more voices and VST3 support. I'm not 100% sure about CLAP support, but I would like to add it.

You did not miss it. I was optimistic as usual and it's taken longer than I hoped. I'm making consistent progress here. Excited to send out the newsletter when it's ready.

This was 48kHz, 25 sample buffers, in "Juce AudioPluginHost" which I use a lot for testing. Results will be vary a lot between different DAWs, so take it only as a very rough guide.

Ripping The Script: clipping.'s Jonathan Snipes Crossfades Between Film Composing And Hip-hop Innovation
by Dave Segal

Along with RZA and DJ Muggs, Jonathan Snipes is the rare music producer who’s straddled the worlds of film composition and hip-hop. Unlike Robert Diggs and Lawrence Muggerud, though, Snipes began his musical career in soundtracks and sound design for theater. (Snipes also did stints with rave unit Captain Ahab and noise group Unnecessary Surgery.) In 2009, Snipes, fellow producer William Hutson, and rapper Daveed Diggs formed clipping., hip-hop’s most interesting iconoclasts. 

Clipping. are poised to release their sixth studio album, Dead Channel Sky, on longtime home Sub Pop Records, and it’s another mind-bending mash-up of various electronic styles and lyrical flows. But before we delve into the splendors of clipping., let’s dig into Snipes’ unlikely path to becoming a prolific film scorer and hip-hop form-breaker.

Snipes’ father passed away when his son was 8, leaving behind a large record collection that was heavy on classical music. That music provided the bulk of his childhood’s soundtrack. Because Snipes’ mother shielded him from television and the radio, he would only hear tantalizing bits of non-classical music outside of his house.

As he progressed through his teens, though, Snipes became fascinated with outliers in his dad’s collection. Philip Glass’ 1985 soundtrack for Mishima especially captured his imagination. “That Mishima soundtrack cover—the only Philip Glass [record my father] had—was crazy-looking to me,” Snipes said in a phone interview. “And I’d never heard music that sounded like that, that was that repetitive, that had electric guitars and drum sets mixed in with string quartets and synthesizers.”

Snipes was 11 when he first heard Mishima, and it “was the first in a long litany of records that broke my brain.”

“I had to listen to this over and over again, until I had assimilated it into my language.”

As he entered high school, Snipes had greater freedom to explore music, thanks to his mother loosening control over his listening habits. Nirvana, Nine Inch Nails, and Aphex Twin opened his mind in those days, but it was the GaijinCD4 album by Disc—a collab between Matmos, Lesser, and Kid606—that blew it wide open. 

“I bought it because one track was made out of a Michael Nyman piece [“Call It In The Air”].”: GaijinCD4 taught Snipes that

“anything can be music and you can do crazy stuff and it’s really compelling to listen to. You can find musicality in the broken technology around you and the sounds that people don’t want. 

“Around that same time, I found MP3.com, rest in peace. That was my first exposure to just anybody making music and uploading it.”

Thus inspired, Snipes began to create his own music on a computer, albeit with the same nonchalance he used with MS Paint or playing Doom. “I never thought I’d be a musician because I lacked the discipline to practice. And the only window into being a real musician that I knew was the Western conservatory. You have to play the same two-minute Mozart piece 12 hours a day if you want to be even passable. I hated that. I love a lot of that music, but I was so bored by practicing. I still am. It’s a huge problem for me.”

At UCLA, Snipes studied theater sound design, and he’s been teaching that discipline at the same university since 2008. He’s worked on over 40 film and TV soundtracks, including Room 237 (2013, with Hutson), Excess Flesh (2016), A Glitch In The Matrix (2021), and Batman Unburied (2024). Some highlights: “Universal Weak Male” flexes muscular, spine-tingling, Patrick Cowley vibes; “Excess Flesh” summons momentous battle action with steely determination; “I’ve Been In Your Head” conjures the foreboding beat science of Scorn and Meat Beat Manifesto; The Nightmare (another Hutson collab) immerses you in dank, pitiless desolation. “Inside The Matrix”—my favorite piece by Snipes—generates 12 minutes of insanity-inducing, Steve Reich-ian hypnosis.

While most of his soundtrack work has come from low-profile horror movies, Snipes can easily shift modes when necessary (see the lush electronic score with soaring choirs and twinkling synths from 2015’s Starry Eyes). “In film, there’s not a lot of projects I turn down,” Snipes says. “I will turn down projects that don’t have enough budget for me to even do the thing they’re asking for. But I find that there’s usually a way to make any project interesting to me. 

“I’m not getting a lot of crass, commercial work that isn’t fun. I’ve done a few of those, but I’m getting approached by interesting people pretty much all the time. I have turned down work that would have been really good money, but I saw nothing of interest.”

Did Daniel Lopatin get some of those projects? Snipes laughs and says, “No. He’s done some stuff that I would love to work on. Certainly, I would love to get his budgets. But I’m not there. He has access to a tier of projects and budgets that I don’t seem to be able to crack into. I’d so much rather do one Uncut Gems every two years than three micro-budget horror movies or documentaries a year. Not that I don’t like the projects I work on, but it’s exhausting to be constantly grinding without having quite enough resources.”

His discography clearly shows that Snipes is drawn to horror films. Is it the moods that attract him?

“The reality is, almost everybody starts in horror, because those are the low-budget movies that need original music.”

“And that music can be made pretty cheaply. It also happens to be my taste; I love horror movies. I love horror-movie soundtracks. I understand the world and I do okay at that, so I keep getting [the projects]. 

“I’ve never scored a Hallmark or Lifetime movie, which is the other place where people start. Except I sort of just did. Because I did Adult Swim’s Yule Log 2, which is like a fake Hallmark movie. It was genuinely really fun. 

“That’s something that will always get me to say yes to a project: Oh, I’ve never done that before, but I bet I could. I like assignments and briefs. Which the band clipping. indicates, too. That has rigid constraints and all three of us are really good at following a brief. That’s also what I like about film music.”

Do you find that the methods you use for soundtracks apply to the work you do with clipping., or are you using totally different muscle groups in the band situation?

“A little of both. There is a fair number of clipping. songs that started as score cues. Often what will happen is, I’ll make something that doesn’t quite work for the movie, but it’s an idea that I like. If it doesn’t work because it’s too weird or too harsh or sounds too much like sound design, then I set it aside and play it for Bill and Daveed. ‘Hey, we could loop this differently or put it in a different structure.’ There are a handful of things that started that way. Even with just one sound. More of that happened before, maybe, but rejected score cues get added to the folder of clipping. ideas a lot. 

“Clipping. was the project that taught me my interests were compatible and informed each other. Clipping. was the one where if I take all the acoustic ideas that I’ve learned from theater sound design and I apply those conceptual things to making a rap song, that’s interesting. Because even though it’s an obvious connection to me, it’s not a thing that people maybe have heard before would have done. 

“Clipping. made a song called ‘Run For Your Life’ where Daveed is rapping over beats in cars that drive by. So he’s rapping over nothing on the street, then a car drives by and he’s perfectly in time with it. I wouldn’t have thought of that if I wasn’t a theater sound designer. Delays and narratives, the way acoustics work and things like that. It’s a very theater idea.”
 
What distinguishes clipping. is how unusual and shocking the music is compared to that of most hip-hop artists. They have an unusual configuration of two producers and one MC, so it would be illuminating to know how clipping.’s tracks manifest. 

“We’re all new dads now, so we’re figuring out how the hell we can make music,” Snipes says. “We also don’t live in the same city. It used to be easy when we all lived in the same neighborhood. We would hang out and music would just kind of happen. But now we have to plan things and that doesn’t seem fun.” 

Clipping.’s members have known each other for decades and their musical tastes have evolved communally, enabling them to work with what Snipes calls “a shared language.” During their undergrad days at UCLA, Snipes and Hutson bonded when the former overheard the latter chattering about a Chris Cunningham-directed Squarepusher video. They’ve been inseparable collaborators since.

“Daveed was Bill’s childhood friend from the Bay Area. Daveed was at Brown while we were at UCLA. He’d get out of school earlier and would crash on our couch for a few weeks. Daveed wasn’t a staple in my life till he moved to LA in 2009, 2010. By that point, Bill and I had started doing songs as clipping. that were remixes. Daveed said, ‘What if we did an original song like this?’ And the band came together. 

“Before starting anything, Bill and I have a very long conversation. If Bill has an idea, it’s ‘here’s an idea in experimental music. Here’s a kind of rap song. What if we made that rap song with this very experimental-music technique?’ If Daveed has an idea, it’s usually a cadence, a rhythmic idea for the way that the vocals will work. And if I have the idea first, it’s usually aesthetic. I made this sound and I’ve not heard anything like it. Or I made this field recording that I think has an interesting character to it. Usually, my ideas come from the acousmatic/sonic side. Bill’s are usually more like a critical-theory essay about music in general. And Daveed’s are poetic ideas.

“Clipping. songs are always a confluence of a bunch of different ideas. They all sort of coalesce in a place and the synchronicities between them are varying degrees of intentional.

“’Say The Name’ [from 2020’s Visions Of Bodies Being Burned] uses a quote from a Geto Boys song. I always have a list of ideas for a record. Bill wants to make a song that uses that Geto Boys song as a hook, because that’s the scariest line from that song and we’re making a horror record. I want to make a Chicago, Dance Mania-inspired house track about Candyman. 

“I was listening to a Wax Master track where he was shouting out different projects near his in Chicago and he shouted out Cabrini Green, which is where the Candyman movie takes place. I thought there’s gotta be a whole litany of Chicago house tracks that sampled Candyman. It’s a horror movie, probably my favorite movie, and again, a Philip Glass score. 

“I also have a great love for electro and breakbeat records from the ‘80s that sampled RoboCop. There’s like a million of them.”

“What if there’s this whole Chicago house/Candyman scene like that that I’ve just been unaware of? ”

“But I couldn’t find any, so I said, ‘We need to make THAT. We need to make a Dance Mania track that samples Candyman.’ We didn’t sample Candyman, but the song’s clearly about that movie and the Clive Barker story.”

Madrona Labs products figure heavily in Snipes’ studio. He has been a champion of the Seattle company’s products since Aalto emerged in 2009. “I buy everything the moment it’s announced, because they make really interesting sounds that nobody else can do. I know I used Aalto in a demo of our new song ‘Run It,’ but I replaced it with my modular later. I use Aaltoverb live with clipping. because it’s so tweakable. It’s a very playable reverb, in a really fun way.

“I’ve been using Sumu quite a bit in the film work I’ve done since it came out. I don’t think I used it in Yule Log, which would have been the first big project. I used it in another movie I got fired from. But I have been using it a lot in this short film I’ve been scoring. It’s pretty amazing. Sumu truly does things nothing else does in a way that’s really exciting. I’m sure I’ve used Virta; it’s one of my favorite vocoders.” 

What is it about Sumu that sets it apart from other gear you’ve used?

“It’s this special combination of an incredibly technical and precise tool that, also, I have no idea what half of it does. Some of it is incredibly intentional, but some of the knobs I just tweak until I get an incredible sound. It represents the way I normally work, which is in an extremely precise and technical way, and the way that I’m trying to allow myself to work, which is free play and fun. 

“In undergrad, I did a project for a media class where I used a terrarium filled with sand to make an FFT [Fast Fourier Transform] curve… Like a camera on a terrarium on the sand as a way to literally sculpt sound, like, dig the sound, pile up frequencies. It’s like an EQ curve you physically made out of sand. It didn’t sound good and it wasn’t interesting. That was me learning that any sound could be represented with additive synths, if you had enough partials. 

“A sampler that moves fluidly between the wave form and the spectral domain is appealing to me, because of the kind of manipulation you can do. Treating samples as oscillators is cool and unique and not a lot of people have done that. And nobody’s done it in the way that Sumu’s done it.”

Snipes hasn’t used Sumu with clipping. yet, but he plans to. “The funny thing is, I’ve been using it a lot while scoring a UCLA student film. They wanted a section of it to sound hyper-poppy. 100 gecs was one of their references. I was using Sumu to make the more weird, broken, esoteric sounds until I figured out it’s actually an awesome Trance Gate sequencer. You can actually make really clubby stuff in Sumu. So I used it for that stuff. 

“When Aalto came out, they said, ‘When you input your serial number, it recompiles the plug-ins so that yours sounds unique.’ I thought, ‘That’s a company I clearly need to be supporting.’” 

Snipes’ immediate concern now is preparing for clipping. tours in support of Dead Channel Sky (out March 14), the trio’s most diverse full-length. Inspired by William Gibson and other cyberpunk authors’ dystopian visions, Dead Channel Sky pits Diggs’ amphetamine-fueled bars against a panoply of electronic-music styles—technoise, Cubist house, acidic electro, boom-bap, and on the Nels Cline-aided “Malleus,” a fusion of mad, Squarepusher-esque d&b and guitar antihero Buckethead. The album represents the zenith of the group’s mercurial noise sculptures and disorienting beat programming. 

Credit, too, must go to Sub Pop, which gives clipping. complete creative control. “Yeah,” Snipes says,

“we can do whatever we wanna do, as long as it’s not illegal.”

///////////////////////////////////////////////////////

Clipping's sixth album, Dead Channel Sky, is out March 14, 2025 on Sub Pop Records.

The whole Apple Silicon architecture was such a big step up that I continue to recommend any Apple Silicon machine from the M1 forward.

I just pulled up Sumu on my M1 Macbook Air and four voices run at 40–50% CPU. An M4 should be roughly twice as fast. Personally I would choose the M4 Air, I think they are a greal deal and the fanless aspect makes them excellent for audio work.

Also, Sumu will only get easier on the CPU as time goes by and I optimize it more.

I hear you, Vutu is solid but can use a lot of "workflow" improvements.

Look for these releases in the near future:

• Sumu 1.0.1 : bug fixes
• Sumu 1.0.2 : optimizations
• Aalto 2 : VST3 support and many new features

Thanks for your continued interest and patience. Aalto 2 (and Kaivo 2 etc.) will be a big deal for me too.

Hi friends,

It's tough times. We hope you and your families are in good health. We know you know what's going on, and you have probably heard a lot of takes on it from a lot of companies so we're going to keep this to the point.

Just over a week ago, Minneapolis police killed George Floyd, the latest in a long history of unjust Black deaths at the hands of law enforcement. Since then, the outcry has been ringing out in protests all across our country. The violent suppression of these protests by police has resulted in more injuries and deaths. We stand in solidarity with the protesters and with the Black community.

How can a music software company contribute positively and constructively in this moment? For us, the answer is: by sending funds to those doing front line work. We're writing to ask for your support.

The Black Youth Project is an organization that both provides emergency funds to Black community members in need, and invests long-term in the future of Black leadership. They have chapters in many U.S. cities including Atlanta, Chicago, Dallas, DC and Detroit. From their website: "Our work is generally centered on ending systems of anti-Blackness and emphasizing the urgency of protecting folks living on the margins of the margins, including women, girls, femmes, and the gamut of LGBTQ folk." To find out more, please visit https://www.byp100.org/about.

From now until the end of Friday, June 5 (Pacific Time) we'll be donating 100% of our sales to BYP100. As a family business, our pockets are not deep enough for an impressive direct donation right now, but this way we can leverage the software we've made to direct some funds where they are sorely needed. We hope you'll help.

We and some folks we know have been contributing individually to community bail funds around the country and that is another way to help right now. If you are interested in contributing to these efforts, here’s a link to donate to many community bail funds at one time: https://secure.actblue.com/donate/bail_funds_george_floyd

There will be a time soon to get back to music, which is more important to us than ever as a force for change, as meaning and community and connection. But this week is not that time. If it were not for the inspiration of great Black music, Madrona Labs would not exist. We owe the Black community our help.

with all best wishes,
Randy Jones
Madrona Labs

photo: Nathalie Graham

working on it!