openAV-Ctlra VS cane

I have been talking to the author of Ctlra who has reverse engineered the screens. The screens work by sending a bunch of pixels to a vendor defined class USB bulk endpoint. Unlike some other controllers with screens which just take some metadata from the software and do their own rendering, Native Instruments controllers with screen rely on the software to do the rendering. Traktor does this by rendering QML to an offscreen buffer. However, according to the author of Ctlra it is more complicated than that to get 60 FPS framerate to the screens because simply sending all the pixels continuously saturates the USB bus. Fortunately that doesn't interfere with any other features of the controller because the other USB signals have higher priority (isochronous for the audio, interrupt for the HID). So to get a good framerate, which would matter for sending scrolling waveforms to the screens, my understanding is that Mixxx would have to compare every rendered frame to identify which rectangles of the screen to update in each USB packet.

I've been working on my language Cane (https://github.com/Jackojc/cane) a lot in the past few months and it's getting to a point where I feel like I should show it off.