So I’ve got a JUCE plugin that is getting close to a 1.0 version. Let’s say I wanted to turn my JUCE plugin into a piece of hardware – either a Eurorack module or standalone gear – what are my options in 2021? I’m assuming I would want to figure out some sort of embedded hardware that could run some version of Linux that I could compile against. I’m positive that I could do this as a hacked together R&D project for myself, but has anyone gone down this path and actually brought a product to market?
1 Like
I don’t think there’s really any “supported” way to do this, but you’re right that compiling to Linux is your best guess.
For eurorack there’s lots of proto boards you can get which utilize Arduino teensys – but you can’t run JUCE on those, so you’d need to port just your DSP code over.
Others have posted hardware prototypes they’ve built using raspberry pis – but I don’t think anyone has shared the details of any products which have gone to market – at least not that I’ve seen on this forum
Never worked with it myself, but I heard that Elk should make all that quite easy as it supports JUCE out of the box
5 Likes
Yes, totally feasible. This is primarily what I focus on.
The whole conversion from plug-in to hardware product is fairly straight forward. The primary focus here is to conceptualise how the current product that resides in the digital domain behaviours in the analog domain (despite it being digital).
So… input / output behaviours (signal path), the tactical improvements? Additional feedback elements that provide a different experience to the user? Different formats will add an insight into the limitations placed on the final product. Although, limitations can be good in the hardware world. Having a product that does everything (is feasible), but then you have cost and space concerns, and ultimately who are you trying to target.
Once you’ve narrowed in on these, the next is going down the path of component resourcing to bring these ideas to life. What integrated circuits do you need, control parameters, feedback systems? This is where a hardware engineer helps out.
Without knowing anything of the product (or idea), it’s hard to give advice on which route to take moving forward. However, if you are thinking of a potential commercial product, I’d move away from the Arduino, Teensy, Rasberry Pi etc and integrate an ARM processor or other DSP chip onto a well designed PCB with all the required and auxillary components. It all depends on what your vision is and how far you want to take it.
Happy to further consult on this, so by all means you can send me a pm if you’d like.
3 Likes
@cackland I’m in a similar situation and I’d like to consult with you. Can’t PM yet probably because it’s my first post.
I just saw this, depending on how much cpu your plugin needs and how much i/o, i might work for you: [PRE-ORDER] EuroPi Panel + PCB | Allen Synthesis
2 Likes
Hi @cackland
I’m after the same thing (integrating JUCE to custom PCB / ARM / DSP chip), and would love to chat with you about this. I’ve just signed up to this forum, and my new account seems to be restricted from sending PMs.
If you could please get in touch (here, if it works?), or via andy@andyoneilmusic.com that would be much appreciated.
Thanks and hopefully chat soon - Andy
The Electrosmith Daisy Eurorack platform is worth a look for people interested in this who are already invested in the Eurorack ecosystem.
It seems plenty powerful and some respected Eurorack brands like Noise Engineering and Qu-Bit Electronix are using it in their products. I’ve been dabbling with my own Eurorack firmware on their patch.Init() module, which is nice for me because I know very little about electronics. They have other hardware options, including non-Eurorack and more low-level boards if you want to design your own interface panel and solder the electronics yourself. For the more custom route, the Synthux Academy Simple Designer prototyping board may be of interest. That is also powered by Daisy.
I haven’t ported a JUCE plugin to this because I am just learning JUCE now, but the C++ code for running on this platform is pretty straightforward (here’s an example of a simple reverb), so I imagine if you keep your core DSP algorithms fairly decoupled from the JUCE framework, you could run it on a Daisy pretty easily. Or even if you are using JUCE classes for DSP or whatever, maybe most non-GUI JUCE code will run on it. I should investigate that…
3 Likes