28 Followers
44 Following
254 Posts

Kerfluffle expert

SuperTux dev. I try to contribute to other projects that interest me in my spare time like EFL, Gentoo (mostly Gnome), and some others... :)

I post rather infrequently

WARNING: THIS SWAG IS NOT A TOY.

The Git Hubhttps://github.com/swagtoy
I am too lazy toupdate my web site
irc, matrix, etcask! :)

Of all the things I've done in my career, this is probably the one I'm the most proud of.

When me and two others at Intel started working on the Vulkan driver, Mesa had a reputation for being behind on everything. The Intel drivers were still on OpenGL 3.3 (fp64 was a pain), OpenGL ES 3.1 or maybe even 3.0, and perf okay but kinda meh. I think there might have been a driver or two in Mesa exposing GL 4.x at that point but, as a project, we were still a ways from full OpenGL 4.5.

With Vulkan, we jumped the line and had Vulkan 1.0 conformance on Intel on launch day. It was a hell of a lot of work (I worked 80+ hours/week that last month or two) but we got there. The driver branch we dropped that day was pretty shaky and it was missing a lot of features but we were there. It took a year or two to get to where we had decent perf, working games, and feature parity with the hardware. But that was okay because there were only two titles that came out that first year and getting them working was the important bit.

Then Vulkan 1.1 came out and we were there with a day-0 driver again. This time, without missing any interesting features. Then 1.2 and 1.3 and now 1.4. With every new version, more drivers joined the train. When Vulkan 1.4 launched, there were 5 different Mesa drivers that landed MRs to enable Vulkan 1.4 on launch day.

This has totally changed the conversation about open source graphics. When I started, everyone scoffed at Mesa. Today, the speed at which we're able to implement features and launch new API versions is the envy of the graphics industry. We're still not totally caught up everywhere—NVK and PanVK still need work and etnaviv Vulkan doesn't exist—but we're going toe to toe with the proprietary driver teams across most of the industry. Also, the fact that Linux Vulkan drivers are being hammered by most of Valve's library via DXVK and VKD3D often means the Mesa drivers are often more stable and robust than their closed source or Windows counterparts.

It's a totally different world for 3D graphics now than it was a decade ago.

Hi #fediverse. We need to talk about something.

While talking to a colleague it came up that they have never sat on a chair. Like, not even once in their childhood.

Another colleague listening in admitted they also have never sat on a chair.

My hypothesis is that most people have at one point in their life sat on a chair.

💺 🪑

Have you sat on a chair?

Please boost for scientific accuracy.
No
Yes
Maybe
What
Poll ends at .
One of the ways LLMs are damaging society is by causing us to endlessly talk about LLMs when we could be doing something creative.

RE: https://social.treehouse.systems/@swags/116235865719541791

Today was the SuperTux 0.7.0 release (yay!), but some of us were alarmed by the "Claude contributed to this repo" message on GitHub (oh no!)

Well, I've learned a few things from the discussion about this topic.

A single PR "co-authored by Claude" has slipped through. The code was written by a human (according to the commit), then properly reviewed and eventually merged by a maintainer. The Claude contribution is attributed solely to the code comments added through a mid-PR-flow commit update.

And now the commit history will show Claude as a contributor. It might be impossible to rewrite it. Rewriting kind of goes against the idea of having the git history, too. Even if that was possible, it's an extra effort which not all maintainers can prioritise.

So: seeing "Claude was here" in a GitHub repo is a "red flag", but it shouldn't be "therefore guilty of slop and LLM proliferation by default"; please use your head/best judgement.

I guess it is a cautionary tale for the maintainers, too 

After many years of development, we are finally excited to release SuperTux v0.7! This version features a completely revamped art style, fresh new levels, new music and so, so, so much more!

Fetch the downloads here: https://github.com/SuperTux/supertux/releases/tag/v0.7.0

And seriously, thanks as always for sticking along; we really hope this release also reaches those who need it most during these times. Thanks a lot, and have a lot of fun :)

Release trailer: https://www.youtube.com/watch?v=PQedJ3SX8s0

Github actions is so very very very very bad and is in fact the entire reason why this release is taking so long
🫩

🚨 #GNOME 50 will be released on March 18!

😍 Star or/and boost this post if you're excited! With 5 more days to go we just asked our developers to drop their final GNOME 50 release of their projects.

#Linux #Opensource

Nevermind LOL