Show HN: High speed graphics rendering research with tinygrad/tinyJIT

17 days ago (github.com)

I saw a tweet that tinygrad is so good that you could make a graphics library that wraps tg. So I’ve been hacking on a gtinygrad, and honestly it convinced me it could be used for legit research.

The JIT + tensor model ends up being a really nice way to express light transport all in simple python, so I reimplemented some new research papers from SIGGRAPH like REstir PG and SZ and it just works. instead of complicated cpp its just a 200 LOC of python.

11 comments

quantbagel

Reply
nl  17 days ago

Why is this a fork of tinygrad and not just something that imports it?

  • sixtyj  17 days ago

    Because forking is new coding /s (What we see is natural entropy of systems. Wannabe codies fork a repo… and instead of contributing to original one they make their own copy. What will happen if you repeat this a few times? ;)

    • Keyframe  17 days ago

      That is a legit way of working on contribution. You fork, you work on the fork - if it's not junk then you issue a pull request. What's the deal with belittling and holier-than-thou moralizing?

      1 reply →

    • quantbagel  17 days ago

      Well I wanted to implement light transport papers without having to deal with cpp. I think tinygrad, and more specifically tinyJIT are super useful abstractions. This is def not available in ts

      1 reply →

sxp  17 days ago

Claude didn't follow your "Every line must earn its keep. Prefer readability over cleverness. We believe that if carefully designed, 10 lines can have the impact of 1000." from https://github.com/quantbagel/gtinygrad/blob/master/AGENTS.m... given how bloated this demo is.

https://blog.evjang.com/2019/11/jaxpt.html is a better demo of how to render the Cornell Box on a TPU using differentiable path tracing.