Ksoloti for Professional and Commercial Use: Capabilities and Limitations

I have used the Ksoloti platform to carry out various projects, including some high-profile ones, such as Hans Zimmer’s giant modular synthesizer, in which the Ksoloti is used to create a digital oscillator.

I continued to explore ways to integrate Ksoloti into other projects, but several limitations led me to ultimately stop using Ksoloti for anything related to synthesizers.

However, there remains one area where this platform is still quite interesting: effects, particularly dedicated effects (with the help, notably, of SSS, a well-known contributor to the community). So I developed a board dedicated to effects that integrates the Ksoloti platform. This platform includes an adaptive buffer system that allows the board to be used in mono or stereo with Hi-Z or Low-Z inputs, so you can directly plug your electric guitar or a synth into it. You can also choose between a true bypass or buffered bypass configuration. And of course it works on 9V.

This board was designed to accommodate a whole range of daughter boards equipped with various controls (potentiometers, rotary switches, toggle switches, LEDs); each type of effect has its own dedicated daughter board.

The Ksoloti thus allows you to develop powerful digital effects, but again, by limiting yourself to one type of effect per board, a complex stereo phaser (input/output) switchable between 12 or 24 stages per channel can consume nearly 70% of the DSP load on its own.
For complex reverb effects, this goes even further.

I don’t know what the status is on the project to port the Ksoloti to H7-type microcontrollers, but I think this is a necessary step to ensure the long-term viability of the Ksoloti and all the work Sebastian has done.

By the way, Sebastian, feel free to reach out to me so we can discuss how I might support you in this project to make Ksoloti more powerful.

-S-

Hi Syntesla,

Thanks for sharing the pedal gen design. The all-in-one concept and adaptive input stage look very cool for prototyping. And of course Ksoloti being used in Hans Zimmer’s (and Jean-Michel Jarre’s) setup is an awesome milestone for the ecosystem!

The H7 port:

  1. Hardware: The H7 board hardware design is finished. It has 4-channel in, 6-channel out by default (2 of the outs are independent headphone outs). The pinout remains compatible with existing Ksoloti Cores (the additional IO header is located above the original 6-pin IO header). So if all goes well, you can easily upgrade existing devices.

  2. Firmware: This is the bottleneck. To make sure we don’t end up with an unusable platform or repeat the issues of projects like Akso, the H7 firmware needs to integrate perfectly with the existing Ksoloti Patcher and object libraries.

I work on the firmware and Patcher whenever my free time and energy permit. It involves a lot of learning from scratch, experimenting, and comparing against existing repos (like Akso), so progress happens step-by-step. So where support would be most valuable: If you or anyone in your network has experience with STM32H7 C/C++, firmware code contributions (or object code checkups) would be the single best way to accelerate the H7 release.

Platform sustainability and licensing
Since Ksoloti is a derivative of the original Axoloti project, we are “stuck with” the GPL v3 license. Strictly speaking, GPL v3 requires that derived hardware schematics and software changes are published open-source under the same license, that original copyright and links back to Ksoloti are preserved, and even that users have access to modify/reflash the firmware.

That being said, so far we have been relaxed and tolerant about how strictly this is enforced. I enjoy seeing people build and experiment. But for the platform to remain viable, reciprocity is important. Simply honouring the core principles of the license, like providing prominent attribution/links back to Ksoloti and, if feasible, sharing derived work back with the community, helps keep our effort unified rather than fragmented.

If you or anyone would like to jump in and help with the current H7 firmware code, let me know and we can coordinate directly!

Is the H7 stuff I did no good then? Or is it just all out of date too much now?

My Gills is running with the H7 board you gave me, but I guess it is far behind your recent changes now!

Your branch is essential! It is really just the branches diverging too much, so I was going step by step, first updating to ChibiOS 21 for all boards, then starting to figure out the memory handling on the H7 (as my branch has all that SRAM3 for code stuff going on now… now how do we figure this out for H7 without breaking the patcher), etc. etc.

Great!! thank you very much for the news.

If the Firmware need to integrate perfectly with the existing Patcher and object, does it mean that the audio resolution will be stucked to 48Khz?

I’m way behind you when it comes to development; you’ve been working on Ksoloti for years, and you have experience and knowledge that no one else has—I wouldn’t be of any help to you.

However, I’m more than willing to provide you with material or financial support.

SSS convinced me to try out AI tools for coding, and I have to admit that, even though I’m not a big fan of AI, for certain tasks—like testing cross-platform compatibility, compiling complex code, and testing compatibility between microcontrollers and codecs—it becomes a useful tool, provided you know exactly what to ask of it.

Another way I could help you would be to fund the development work—“STM32H7 C/C++, firmware code contributions”—through a platform dedicated to freelance pro software engineers…

Your thoughts?

Good point, there has always been the plan to up the samplerate to 96k. Worth it for the filters alone. It should be possible to have it as a separate firmware mode - because 96k would come at a cost, reducing the improvement in DSP power on H7.

Another interesting idea. Maybe some kickstarter thing could be set up, and this would also allow us to estimate how many people are willing to help to get the H7 port working. I have been working fully freelance for my day job which means I would be free to spend a full-time equivalent on Ksoloti development

That would be the ideal solution in the end. Could you tell me what the estimated budget would be to complete this development? As a user of the platform who wishes to use it for commercial purposes, I could sponsor this project and classify it as research and development. This would be fully compatible with making the resulting work publicly available. Of course, everything will depend on the project’s budget and the expected benefits in terms of power gain.
I’d also be very interested in alternative firmware that allows operation at 96 kHz when necessary.

Let me get back to you next week once my current project is finished!

michaelwhite, are you human?

elizabethsmith seems pretty similar as well.

I would guess neither of them are.

Animated Text Captcha Puzzle