Lets make an instrument

Sound Design, Tech Development, DIY
81

that’s not true… EigenLabs did the same, as did Electra One.
and that’s just two products that Ive personally had some involvement with, there have been others :wink:

i’ts not uncommon, but we cannot ‘expect it’… developers, rightly, choose if they want to make their code open source or not, based on many factors… sometimes that decision will change over time, other times not.

for sure, Id like to see more devs at least open source code once the product life cycle is ended, but as anyone in dev will tell you… the code used for one product often still exists in some form or another into the new product … so hence the ‘fear’ of IP remains.

anyway, this is why I talk/work with devs in products im interested in, and talk to them about open source possibilities, and listen to their concerns, and see if there are viable options… its an important dialog, one that has to be done with understanding on both sides.

zrna akso

I’ve not been following closely, but I believe this product is pretty much also dead, or at least finished (it may always have been a limited run?)
perhaps @mtyas has been following it?

I will say there are some others trying to create a new board…but Ive not been following that.

the axolotl forum is pretty unreliable these days (due to issues I, as a mod, cannot resolve) , so it looks like the active discussion has move to lines

can any of the new variant succeed?
perhaps… perhaps not.
the issue is quite simply, its a large project to undertake.

its one thing to design/build a new hardware board, and to create the firmware for it.
but understanding the UI code, and its interaction with the hardware is also pretty complex.
also the current state of the code base is a bit difficult, do you go back to the stable version or try to pick up and fix the ‘development’ version - that is hard to judge if you were not involved in the project.

… and this comes from someone (me) who was involved deeply with the project.
Id love it to succeed, to move forward, but its a big ask.
especially, when its going to take a lot of effort, and there is no money in it to help fund it (you just won’t well enough new boards)

sidenote: axoloti creation was partially funded by an arts program, and it still probably never made Johannes any money… rather more likely it broke even (assuming you ignore alot of the time spent on development!) .

don’t get me wrong, I love axoloti, but I just can’t see without some commercial footing how it can thrive… its a shame, as it did have quite a niche.
but as per my video, there are some alternatives, which are getting on going support.

2 Likes
82

thanks for the updates and perspective!

1 Like
83

That is why I said as far as I know, to stay accurate to my experience. Whether or not there are actually more or less outside of my experience is irrelevant, because I cannot produce baseless claims on speculation or conjecture.

84

sorry, but with a lack of experience, your is a baseless claim … its neither uncommon nor would it be huge news… we cannot just make broad sweeping statements and caveat it with ‘as far as I know’.

sure, raise the Deluge as one example of where it has happened, as thats something you know… but don’t extend it to … its the only time it ever happened :laughing:

as for the case in point,
Elk as a platform does have some motivations to make itself more open, because as a platform you often need to do this to entice developers… as some devs dont like working on close source platforms (though we do all the time, be it OS’s or DAWs etc via plugins/extensions etc), and generally its quite difficult for new platforms to gain traction anyway.

Nina, I do not see much opportunity… (it is fantastic I played with it at superbooth).
sure it might use ElkOS, but its an analog synth first and foremost, it just use Elk as an implementation platform.

It’s this kind of project where Elk makes its money licensing…allow prototyping to be done free/cheaply on their platform, then license for production. its kind of a win win for everyone, and the dominant model…
remember juce does similar with it dual licensing model.

1 Like
85

Well one of its oscillators is digital, and to quote from the website when it comes to the fx, “Upgradable VST3 effects algorithms built on an Open Source architecture.”. And the analog parts are digitally controlled. So I think there is still potential there. However the first thing I would have done with it being hackable was to try to add MPE, but they’ve already done that themselves so its not like I am currently bursting with ideas of what else to add.

1 Like
86

exactly, we need to have the expectation that companies will add features that their user base want… not just make things open source to get 3rd parties to do it. it’s a slippery slope doing this, as companies could use it to shirk responsibilities… “oh we’ve made it open source so the community can do this”

as for the digital oscillator/fx, I think I’d prefer the Kong ‘Logue SDK’ approach, basically its a plugin architecture - this is much less involved for a developer to create for, than just throwing the whole OS at them.
sure, less flexibility (you cannot change everything), but making it easier, means its more likely to happen.

again… comes down to where to draw the line between manufacture and community responsibility / involvement, and which serves the majority of the user base the best (who are not devs :wink: )
as well ,of course, as to what the manufacturer is also happy with, in terms of their IP/investment…
that’s why its not a ‘one size fits all’ scenario (imo)

2 Likes
87

I never stated anything about the occurance being uncommon or not. I understand my experience is limited so I have proactively reflected that in my statement prior to creating my post. I already had first-hand experience with the Synthstrom Audible Deluge (OLED) and its transition from its proprietary roots to a gratis and libre open-source license (GPLv3), so dismissing my experience as baseless does not make any sense.

As for whether or not it would be huge news, that is from my perspective. Whether or not others also share my perspective and values on the matter are irrelevant. It would make no sense to make a claim on the basis of others’ perspectives and values, since I am not familiar with them.

There is nothing broad and sweeping about my statement about the Synthstrom Audible Deluge’s transition. It is one explicit and clear example, but if you want a detailed timeline, I would be happy to provide that down to the date, as I have already experienced it, or whatever it is you want to know about it regarding my post.

I never made any extended claims that it is the only time it has ever happened in the music industry. I kept my statement very narrowly focused on my experiences only, not on any historical viewpoint past that. I am not a music historian nor do I have any interest in becoming one.

88

it looks like the active discussion has moved to lines

I noticed the slightly argumentative topic on lines the other day, so I was interested in your comments on the difficulties of this attempt.

I’ve only just realised the supremacy of the STM32. I spent a while hacking around with the Artiphon Orba which uses it. I have a Daisy Pod and an Axoloti Core which also use it. It’s clearly time I tried to understand it better, and curently Daisy is probably the most accessible way in.

The furthest I previously got with Daisy was someone’s attempt to port OscPocket. I see this has moved on a bit since so I’ll start by checking that out.

1 Like
89

Im not following it… but yeah its a tricky project.

you can really consider axoloti as FOUR different projects (albeit interrelated)

  • hardware
  • firmware for hardware
  • dsp objects
  • UI

so the scope is vast…
but what makes it worst, is all four projects require different skills… and many, only have skills, in some areas, not others.

Johannes, was good at the top 3, but was less focused on the last one (UI/Java).
which is why I jumped in, I have more experience in the bottom 3, but not so much on the hardware… so before full release, I tried to help him out on the UI/Java side.

so yeah… what tends to happen is, someone gets excited and starts messing with one bit of the project e.g. a new bit of hardware. but the deeper they get, the more they realise how much more there is to understand, and skills required… and so they loose momentum

really it needs a few people to pick it up and work as a team… but thats hard to organise without a lead/focus… which is hard to do in a non-commercial context, esp. as there is, like any project, stuff to do that, frankly, is not fun/exciting to do.

so I won’t say it won’t happen, but I do fear we will see a few attempt and give up :frowning:

STM32 , oh yeah, STM are huge !

music tech is a minuscule drop in the ocean for STM, we are living off the back of other industries investments here :slight_smile:

though, its what caused the (music tech) chip crisis…
the STM32 F3/F4 was a staple for many industries, so when there was a shortage, the car companies bought up the entire stock ! …they weren’t going to allow a sub $10 chip, stop them making $50+ k cars, so had no worries buying them all up (for millions of $)

as you can imagine, smaller music tech companies, just couldn’t compete, so they were forced to change, or buy at exorbitant prices.
but, its interesting talking to music tech hardware manufactures about how its changed their design process, and trying to be a bit more flexible on what components they use.

fortunately, situation has improved now… but its had a bit of a lasting effect.

2 Likes
90

I’m currently thinking I should focus on Daisy as the easiest way into STM32, as the stuff I’ve heard based on Axoloti and Daisy (eg the horrible-looking Chompi Club) generally sounds better than the demos I’ve heard using synths based on Teensy. I’m still lingering slightly as I have a little experience with Teensy and none with Daisy so the learning curve will be steeper, but I just haven’t been very excited by the sound of Teensy-based projects I’ve seen, although Jeannie isn’t bad. I’m curious as to whether…

  1. I’m mistaken, and Teensy-based synths are just as good
  2. It’s a hardware thing
  3. It’s a coding thing

(I noticed in the Lines discussion that Striso includes some of Johannes’ code. Does that mean it’s also an STM32 device? The current focus on Lines is Ksoloti.)

1 Like
91

teensy vs others…

It’ll be a bit of a mix of hardware and software,
on the hardware side the codec (dac, and surround components) make a difference, but also software different algos, and of course if its a more powerful processor you might be able to to use more ‘expensive’ algos.
so to know for sure, you’d have to look at each in detail.

striso, looking at GitHub, etc , indeeds seems based on stm32, and indeed uses a bit of axoloti code (which is all good, as axoloti is open source, meant to be reused :slight_smile: )

1 Like
92

OK! Thanks for the insights…

1 Like
93

This might be worth a look. Pico ADK.

https://www.thonk.co.uk/shop/datanoise-pico-adk/

2 Likes
94

…just a signpost; the new official residence of the Ksoloti project, currently the main Axoloti-like.

https://modwiggler.com/forum/viewtopic.php?t=277847

1 Like
95

…and the “Screech Owl” is finished. :grin:

image
image1280×830 147 KB

image
image1280×888 206 KB

image
image1824×1638 177 KB

4 Likes
96

one particularly nice thing about this thread is it does introduce one to some fabulously innovative instruments.
Pretty sure ive come up with one based one motion tech and the ekg to midi stuff integrating ai - think ive figured out a way to train it to read the ekg waves and translate them into articulations based on allowing it to learn from the patterns of when you are playing a real instrument - thus once trained one could create instruments in ar, or simply use their imagination - ar would ultimately give one a better medium of with to work with though - still working on the crash course in coding, which isn’t exactly easy blah

1 Like
97

I’m currently concentrating on a project called the “Midihex”, which is an Axis-49 clone. Pleased with progress so far… :slightly_smiling_face:

image
image1920×1275 260 KB

5 Likes