I am fairly proficient in C/C++ but I'm pretty sure that's not enough to get into Mesa development
In general I always found it difficult to get into opensource development, but Mesa probably is even more difficult because of its low level nature.
I am fairly proficient in C/C++ but I'm pretty sure that's not enough to get into Mesa development
In general I always found it difficult to get into opensource development, but Mesa probably is even more difficult because of its low level nature.
Sorry to burst your bubble, but ET:QW is running for me fine on the stock Mesa and R600 Gallium3D drivers currently available for Fedora 16. Just make sure that r_useIndexBuffers is set to 1 in .etqwctl/base/etqwconfig.cfg and you should be golden.
EDIT: Just saw that you only meant that you had to build a separate 32bit driver for playing certain 32bit games. My bad.
Last edited by Hamish Wilson; 08-07-2012 at 09:33 PM.
Its really sad that patches getting tested but are never comitted (e.g. GL_SELECT Patches for radeon) :/ Since over one year the code is there ( and from many ppl tested ) but no dev what to look at this :/
Interesting, on my hardware (Pentium E2160@2,7GHz, 2 Go RAM, HD4850) I remember that ETQW worked on F16 but was too slow to be playable. With F17 that was much better, but the current upstream is for me the first "version" really playable. ETQW runs at 30~60 FPS in 1680*1050 (low details), without bugs or crash obviously (except with Linux 3.5).
Yes it happens that devs ignore stuff, but that's not because they don't like it but more that no one has had the time to finish this up for inclusion. To get something into mesa you need someone actively driving stuff forward. In the GL_SELECT case there was a pretty good implementation that was even pushed to a branch upstream, but after that there was no one who did the work to cut the last rough edges.
If this is really a thing that bothers you, why don't you go ahead and try to get it mainline? It shouldn't be too hard, as it was nearly finished. Just get on the Mailing List and ask what other people think, what has to be done to get this integrated. Then if you fix up the last bits I think there will be no one that blocks this from inclusion. It's just not that devs go there and say: "oh, there's this code to enable a feature I never used and never tested, but ok I'll commit this". You need people to drive things forward, even if it's sometimes just a minor push that's needed to get things going.
So please don't just talk about "the devs", this is not a proprietary project, this is open source, so you can easily be a dev, too.