[samus]

CPUs and DSPs are not primarily designed for graphics work, therefore they don't count as GPUs. CPU are general-purpose, DSPs might be abused for graphics work.

The "G" in GPU doesn't imply that they have to render directly to a screen. In fact, professional graphics cards are commonly used for bulk rendering for animating videos.

Datacenter GPUs are mostly used for AI these days, but they can nevertheless do graphics work very well, and if they are used for generative AI or if their built-in super sampling capability is used, the distinction becomes rather blurry.

[Narishma]

But this particular one isn't designed for graphics work either, so it shouldn't be called GPU.

[samus]

It's in the very name: "Tiny-GPU". Since it's a demonstration project by a hobbyist, the author probably didn't want to implement the whole optimized rendering stack yet.

On the other hand, they also left out some features that you'd expect to find on a general-purpose compute accelerator.

For example, they focus on tensor math. No support for bit wrangling and other integer math. No exotic floating point formats. Minimal branching capabilities.

[Narishma]

The name is what I'm contesting. It's called Tiny-GPU but there's no mention of graphics functionality anywhere in the project.

[samus]

Graphics pretty much boils down to matrix multiplication, and that's exactly what this thing accelerates. If it were a generalized accellerator, it would have to support other kinds of arithmetic as well.

[Narishma]

Agree to disagree. I'll stop here because we're just wasting time running in circles.