Kleber Garcia

54 Followers
34 Following
7 Posts
Staff Graphics Engineer @ Unity
Previously EA Frostbite

So much of the recent engine performance discussion is vibes based and it’s just bewildering.

Performance needs to be measured, not theorized about. Benchmarks + detailed breakdowns give data, that competent engineering teams can run with to make things better.

If you haven’t benchmarked a workload or something very similar, just assume you don’t know its performance characteristics instead of making incoherent claims.

It's been an incredible journey for me and the team from 2021 onward to build out what has become Unity Wētā Tools, launched this year at SIGGRAPH in LA.
Take a look at what we've built in our SIGGRAPH keynote: https://lnkd.in/g9KmHnHT
LinkedIn

This link will take you to a page that’s not on LinkedIn

A response to a student who was surprised that I still bother working on "old" (non-AI/ML) topics in graphics.
@aras honestly also good compute code lacks in the wild because is hard to write GPU compute without a giant infrastructure framework. Unity is not lean, unreal is crazy, frameworks are all c++ and CUDA shrinks your options.
This is why I've written coalpy, so people can write hlsl and tie it with simple python bindings. It includes also imgui integrated support, live editing, window management and can use a vulkan backend. The goal is to democratize real hlsl compute
https://coalpy.org
Home

Compute Abstraction Layer for Python.

CoalPy
@aras thank you Aras for the fixes. We encourage everyone to help us. If it wasn't for Natasha and Seb this work would be dead. So we shouldn't take it for granted, is a privilege to have it out and we fought current unity management to not kill these features
We have a full prototype of the hair rasterizer as well on a smaller GitHub project in python and hlsl
https://github.com/johnpars/LineSoftwareRasterizer
If you anyone buys me MacBook pro I'll port it metal :)
GitHub - johnpars/LineSoftwareRasterizer

Contribute to johnpars/LineSoftwareRasterizer development by creating an account on GitHub.

GitHub