AMD's R300 Gallium3D Driver Is Looking Good For 2011

[marek]

Then how come ARB shaders work much faster using r300classic?

That's a real mystery to me.

Can't the developers use the assembly compiler of r300c in r300g?

Both the drivers have been sharing the same compiler backend since ever.

To Michael and the others: The article only benchmarks r300g and st/mesa. The changes made to the r300 compiler are not visible in the graphs, because all the new compiler optimizations are applied to both r300g and r300c. A benchmark of both the drivers with several different versions of Mesa would better show the overall improvement. Yes, r300c is getting faster as well through the compiler work.

[mattst88]

It's not about the lack of a threading library. It's that drivers don't make use of threading. It's nod odd, it's the norm.

[popper]

The open source drivers are not multi-threaded AFAIK, so single-thread CPU power probably makes a big difference in the performance results. It would be great if the same CPU could be used across a series of benchmarks so that the driver/hardware differences could be isolated.

"The open source drivers are not multi-threaded AFAIK" which is Very ODD Today if thats really the case, as NPTL(Native POSIX Thread Library)has been in since 2.6 started http://en.wikipedia.org/wiki/Native_...Thread_Library
"NPTL has been part of Red Hat Enterprise Linux since version 3, and in the Linux kernel since version 2.6. It is now a fully integrated part of the GNU C Library.[citation needed]
There exists a tracing tool for NPTL, called POSIX Thread Trace Tool (PTT). And an Open POSIX Test Suite (OPTS) was written for testing the NPTL library against the POSIX standard." ,not to mention there are several other optimized 3rd party threading libraries suitable for any such driver inclusion around too.

[glxextxexlg]

Because the current GLSL compiler in Mesa rocks and really produces optimized code. There's ongoing work to pass every shader (ARB assembly ones and fixed-function ones) through the GLSL compiler to optimize them a bit (and mainly to simplify things for hw drivers), but it's way harder to optimize low-level code than high-level one.

Then how come ARB shaders work much faster using r300classic? Can't the developers use the assembly compiler of r300c in r300g?

[GDJacobs]

Ba?l?klara yaz?lacak cevap

Gort! Klaatu barada nikto!

[marek]

Well r300's are, what, 9 years old?

I think the last r500 was released 4-5 years ago and r500s and Xpress IGPs were still being sold.

[pejakm]

I didn't know about this. Yes please file a new bug with the whole backtrace.

https://bugs.freedesktop.org/show_bug.cgi?id=31232

[V!NCENT]

Well r300's are, what, 9 years old? The fact that they can provide a rich desktop experience is a great gift in itself. r600 and up should be concerned with high def video and full hd 3D.

I, for one, do not realy care that much about 3D games orr accel hd playback, as long as piwer saving, compositing 3D and stability rock. OpenCL would be a nice adterthought as in the next step. However it's floating point on the GPU and that is patented >

[NomadDemon]

I have sacrificed a lot of time for this driver, unpaid. All I can say is "patches welcome".

come to poland.. i can drink with you full crate of beer, good polish beer, for drivers ^^ i know you code for free, thats why i want to give something from me to you.

[pingufunkybeat]

I'm not sure why people are complaining about video acceleration, I've never had a problem with it. My system spends most of its time at 800MHz while playing full screen video.

They want GPU decoding of HD h264 videos, which open drivers do not do at this time.

To me, it would be a nice bonus, but not the highest priority. Solid powersaving (getting there), stability and reliability are more important.