AMD Open-Source S.I. Botched, Hope For The Future

[curaga]

FWIW, llvmpipe is greatly slower in compiling bigger shaders. As in from 200ms to 10s slower.

[bridgman]

We're now going into eight months since the AMD Radeon HD 7000 series "Southern Islands" graphics cards first launched. [/URL]

Probably worth mentioning that the corresponding time period (launch to glxgears) for the last major architectural change was a lot longer :

r600 - May 2007 to July 2009 (~26 months)

SI - Dec 2011 to July 2012 (~7 months)

http://www.phoronix.com/scan.php?pag...item&px=NzM4Mg

The times are even shorter if you measure relative to cards people are likely to actually purchase for use with the open source drivers (Feb for 77xx, Mar for 78xx so 4-5 months rather than 7) but that's starting to get downright boring.

[jrch2k8]

Try with a laptop, it will became warm in a second, the battery will just yell for the discharge rate. It is not only a 3D performance problem. I tried llvmpipe (by accident to be honest) on my brand new quad core ivy bridge. It is not bad, the first minute I didn't saw the difference, but then I started moving windows around.... KDE works only with xrender backend with llvmpipe and you can see the difference. Now this is with an high end CPU.... Really you can't pretend users are ok with this. Some user might be, but really you simply can't ship it as ready for the mass. I would not say it is botched, it is not. It is simply not ready. As John said the community didn't worked a lot on the driver -> you can't expect it will autocomplete. I'm not pointing the finger to someone here, nobody is doing a bad work here, I want this to be clear. But radeon still has major missing features for the masses.

mate is i were you ill check my mesa install cuz is either terrible out dated or is badly compiled cuz LLVM provides 3d support and support GL3 and GLSL 1.30 with KDE Kwin shaders 1/2 good enough for daily work in a semi decent dual core[in ivy must fly].

note that AMD is not recommending or letting the driver into the wild is just that distros instead of providing VESA they include semi upto date FOSS drivers [i agree with the fact distros should put a warning about it] and sometime nerfed or imcomplete but this is not AMD or INTEL or NOUVEAU fault since they can't gun point force the distros to either put a warning or remove the drivers

another truth is r600g is not an AMD driver is an community open driver with the only coincidence that AMD is willing [and they has] to provide technical documentation about the innards of the GPU and support for the community developers, so if you ask me they are doing exactly what they said they will but a bit more help is always welcome

so AMD didn't gun point forced ubuntu[for example] to put mesa in the distro but ubuntu decided to provide open drivers and give ya the choice of install the closed one and i think that is 50% good since you create a bigger userbase for testing the driver but 50% bad cuz if you are not a techie ppl get IANAL cuz the driver is not feature complete yet blaming AMD for something that is not their fault

[Prescience500]

If part of the problem is that the community isn't doing enough to develop the driver, maybe AMD could give the new GPU's to driver devs to so that they can work on their drivers and test them. That way, they aren't only working on the drivers of the GPU's they currently own as well as encourage new development by people who want free stuff. It may be cheaper for AMD than hiring new people.