🫡

And its still incorrect, cluster bombs are banned by the Convention on Cluster Muntions, which neither Israel or Iran are a party to.

Men really out here inventing entire JavaScript engines so they don’t have to learn Rust/C/C++/whatever

Thank you to all the admins and donators who kept lemm.ee going.

Want to give a shoutout to the tutorial that got me into game dev: ChiliTomatoNoodle - C++ 3D DirectX Programming, A very long tutorial but it really goes deep into the basics of graphics api’s and win32, Even though i now use vulkan and wayland the underlying concepts that are explained in the tutorial are still the same.

Currently I use Code OSS, which is less my favorite but it works.

Out of all the IDE’s I’ve tried (vscode, webstorm, Code OSS, Kate, KDevelop), regular old Visual Studio 2022 is still my all time favorite, using it is such a smooth experience.

Its biggest flaw and why i had to switch is no linux support :(

I’m trying to understand how this is used, my current selfhosted setup is quite simple:

frontend - backend - vllm

Using the openai compatible api vllm returns with tool calls that need to be handeled by the backend, an results are then send to the frontend by the backend, the frontend never communicates with vllm directly.

Form what i can gather AG-UI is a way for the frontend to directly communicate with in my case vllm in a secure way? letting the frontend handle the tool calls?

Well yes the peformance ceraintly hasn’t caught up yet to x86 but the strongest riscv cpu on the market as far as I know has 64 cores on 2ghz. More then enough to run a desktop.

What is stopping people from bringing RISC-V to the desktop now? Major distros already support it and you can run x86 programs with box64.

What is not fast enough then?

All the calculations could be done before hand and stored and then the only thing left in the delayed draw is to set the buffer.

I haven’t looked at the code yet so not sure how much if any it will save though.

Could also group pixels that are far away from eachother into a single call, while a compromise i think it will maintain the effect.

A random suggestion would be to draw to multiple canvases, and use a CSS animation for the delay.

Also not sure if you are already doing this but it might be more peformant to use the raw buffer instead of draw functions.

Alternatively you could look into webgpu, it is ment for these kind of things.