Search:

Expected, yes, also because there aren't any fun non-casual games there. I guess not many gamers (the target audience ?) are going to play a couple of old shooters are they ... Or people like me who...

Same CPU though. So you may conclude that the driver code itself, while it may configure the GPU well, isn't very fast.

Thank you :)



The number of contributors is difficult to express in a meaningful measure because most of them (in this case, all but me and the guy hired by Red Hat) only very rarely find the...

Fine. But that approach is not practical, because then you'd not be able to use nouveau on modern cards at all; also, for example, one of the components extremely vital for performance, the shaders,...

That's not how it works. It's like saying you have to learn how to drive a horse carriage before you can learn to drive a modern car.

To see how pointless and hopelessly slow nouveau is, let's pick an example in its favour, say 9800 GTX vs Radeon 6950.
The Radeon card is about twice as powerful...

And I have time to post on phoronix forums ... something doesn't add up.

I have no / too little time to implement the tons of shader compiler optimizations I'd really like to implement to make things faster.
I have no time for tuning and tweaking performance for various...

For most cards there seem to be people with issues ... my GTX680 works perfectly fine, for Michael it seems not to. Should we disable acceleration for everything except, say, whatever works for him ?...

The subliminal message was more along the lines of "If it doesn't work, fix it !" I'm trying to get people to contribute but they're awfully resistant. Maybe that's not the right way. But then, if...

Docs can also be crappy and near useless (except for public opinion). If you tried to write a driver for r600 having just the released docs, you'd probably fail miserably. Unless you count existing...

That's not a fix, it's a workaround. If you had a fix, we might just care.
Myself I really don't care (well, I do, but not to the point where I'd do something about it), my ~2000 hours of wasted...

That will get done when someone who cares joins the circle of nouveau devs.

Except that you occasionally have to quarrel with Marek about shared interfaces ;)

Do you think we're making this up ? - https://github.com/pathscale/envytools/blob/master/rnndb/nv50_2d.xml (NV50 = G80, naming always uses chipset id where the class interface first appeared)

It...

Neither does NV's 2D engine. It can do solids (with ROP) and blits. But still, setting up the 3D engine for a single, known operation is much easier than dealing with all of OpenGL. The most...

Luckily the memory type (xDDRy) doesn't change that often, and the interfaces to it tend to only change with each new card generation (Fermi, Kepler, ...).
The trouble is, the register values that...

And why exactly are we supposed to cripple our perfectly good (ok, maybe not perfectly) 2D driver ? Going via OpenGL would add considerable overhead and deprive us of the opportunity to use the 2D...

Btw., I also added support (on Kepler) for a number of the shader processors' performance monitoring counters that the CUDA profiler exposes (like number of instructions issued/executed,divergent...

Can't say what the UVD firmware does (since it's closed source ...), but video decoding "firmware" on NV can do quite a bit. E.g. one of the stages of VP2 seems to be based on an Xtensa processor and...