Looks heavy on the NVidia specifics but it looks like a very comprehensive view of the important concepts in a GPU.

We take text rendering for granted. That said, it's easier said than done, especially on the GPU.

An interesting way to approach the topic of GPU programming nowadays. It might indeed make more sense nowadays than reaching for putting pixels on screen as a first objective.

Long but interesting chapter which shows how GPUs architecture works and the differences with TPUs. This is unsurprisingly written in the context of large models training.

A good reminder that "push it to the GPU and it'll be faster" isn't true. If you move a workload to the GPU you likely have to rethink quite a bit how it's done.

The memory models for GPU programming are complex. This isn't easy to squeeze more performance without introducing subtle bugs.

Somehow this is funny that it works at all. With the advent of SPIR-V we're clearly seeing more experiments in the shading languages space.

Nice in depth article on how to run particle simulations using WebGPU. This one has a very life like rendering.

Wow, very smart approach to solve discontinuity issues when quads are turned into triangles.

Nice set of tricks for post-processing effects all centering around pixelated patterns. Really neat.

Looks like an interesting DSL to write high performance Python code.

This is indeed something easy to get wrong. Also this misconception is very widespread, so it's good to debunk it.

And now we got all the pieces to run CUDA code in the browser. How will you like your cryptominer? Joke aside this opens interesting use cases.

Interesting research about feasibility of making compilers parallelized on the GPU. I wonder how far this will go.

Really nice in depth post. Everything you ever wanted to know about antialiasing but didn't dare asking.

Definitely a funny hack. Not usable for compute workloads though.

Another good tutorial about global illumination. Make sure to read part 2 as well.

Looks like an interesting library to build portable GPU compute workloads. Cleverly tries to leverage WebGPU.

Interesting how much extra performance you can shave off the GPU by going back to how the hardware works.

Interesting history behind the company which was instrumental in pushing computer graphics forward during its time.