nickludlam

I'm super late to this, but I'm looking through the code, being inspired to figure out how it works, and one thing you mention is that synth.rs is a port of synth.py. Is the python code anywhere I would LOVE to get hold of it, rather than make a dirty port from scratch if possible.

This and (the not quite yet open) Kimi K2.8 are my main coding engines, and I find them incredibly cost efficient. I find Deepseek in particular is really good with Svelte work.

Right now I use each of them in their own respective coding harnesses, deepseek-harness and kimi-code, but I do wonder what they are like inside something like opencode.

I've got visions of The Truman Show. In the middle of my question about vacation ideas ... GPT: "Why don't you let me fix you some of this new Mococoa Drink? All-natural cocoa beans from the upper slopes of Mount Nicaragua. No artificial sweeteners!"

Yes, I second Nemotron. I'm using Ultra remotely and Super locally, and I find them very useful for RAG-like problems. I wouldn't really use them for coding.

This looks great! What are the resolution limits? From the GitHub readme I saw mention of 1024x1024, but I was able to process an image 1920x2880 just fine. I didn't see any really obvious scaling artefacts, so is there something clever happening behind the scenes?

It's nearly double the size of Nemotron 3 Ultra, so I'd expect it to be considerably better, although the active parameter count seems to be a touch lower at 41B vs 55B

So its actually a very basic setup. I used pygame to output to the framebuffer device on the Banana Pi, and created a very simple controller class to run the drawing thread.

Its about 300 lines for the projector management class, with some internal state machine logic, and a number of draw calls using pygame.draw.circle(). Calibration offsets are stored in a config file to allow for variations between each physical device made up.

Using a python thread to make the draw calls can occasionally go wrong and the Banana Pi gets stuck, but that's sufficiently rare enough that you can just reboot to solve the issue in the factory.

Yep, we're opening up the relevant repos, and writing up documentation about how to deploy them, and I'm also arranging for a patch to the Bridge unit which allows you to repoint it at a different control URL.

Also, if I have time, I'm going to make this a shink-wrapped AMI to make deployment a little easier. There's sadly not a great deal of time left for this, though, so we'll see what I can get done!

This is great stuff. Is anyone taking videos of the talks as well?