I am designing my own digital console
Hi all,
I wanted to share y'all a project I started over a year ago. I already talked about it in the comment of a post here months ago and now that I have more convicing results with my latest prototypes, I wanted to show it in its own post.
Soooo : I am making and designing my own digital mixing console from the ground up. Why? you may ask, well because I can. The goal of this project is to make a complete system in a distributed manner, in the modular fashion that we could have seen with studer/soundcraft, or Avid for instance (or Innovason for the elderly). Every component of the console is an independant system, IO Boards, DSP, Control surface, they all live on their own rackable board and they all communicate via a switched Ethernet network only. In addition to that, the software I am building is designed with horizontal scaling in mind, so technically you would be only limited by your network capabilities (even though that wouldn't be completely true, as always, reality is more complex).
Of course, I am still pretty far from my goals. However, I have promising results already. Hardware wise, I built an IO Board prototype with programmable preamps and 24 bits ADC @ 96kHz sampling rate. Still a lot of audible noise (My noise floor is around -40 dBFS), but good enough for a prototype, I am actively working to push it back to -60 and below, "fortunately" it is mostly due to a bad hardware design decision...
The \"RackedIO\" IO Board prototype
On the software side, I built a software defined DSP with a plugin interface, a basic ugly GUI control software, and an Ethernet (L2) protocol for audio transport and control streams between the sub-components of the system. There are a lot of features missing, like routing protocol (coming soon 👀) or show management and not everything is perfectly working as intended. But again, for prototyping and PoC, that's way more that I could've ever imagined when I started the project.
Old demo of the software using STEMs as input
My current goal is the get at least 8 clean physical inputs and 2 physical outputs with <= 10ms latency in-to-out processing included before making the console more real-world proof or exploring more complex architectures.
The software is open-source, if anyone wants to see what its like. However I must warn you that this is certainly not production ready, easy to setup nor to compile (And at some places in the codebase, it is not the most beautiful code I have written in my life). This is a pretty complex environment for which I haven't yet written any proper documentation (I will, it is on my task list, but I haven't had the time to). With that said, if you feel like you want to help or contribute in any way, feel free to reach out to me 😄. I'd be glad to discuss about it.
https://github.com/OpenSoundStack/OpenAudioLiveSystem
Here the two things I am also considering :
- Making a series of videos to document and show how the project progresses
- If I see enough interest in that project, I'll probably open a discord server for it