Same problem with Catalyst 5.8 . Has anyone noticed the same problem on non-firegl cards?

Sabuj Pattanayek wrote:

Oops, just realized Catalyst is now at 5.8 (just released today). I'll try that and let you guys know if the problem persists.

Thanks

Sabuj Pattanayek wrote:

...

Hi,

I had upgraded ATI's catalyst drivers from 5.4 (or 5.5) to 5.7 for my R9600SE video card under winxp sp2. In Chimera-1.2143 when I tried to create a surface representation or tried the surface benchmark the screen would go black (monitor led goes from green to orange) and the system would crash (no longer responds to pings).

Downgrading to catalyst 5.6 fixed the problem. I sent a trouble ticket to ATI. Has anyone had a similar problem with Radeon (non-firegl) cards?

Thanks, Sabuj Pattanayek

---

Chimera-users mailing list Chimera-users@cgl.ucsf.edu http://www.cgl.ucsf.edu/mailman/listinfo/chimera-users

---

Chimera-users mailing list Chimera-users@cgl.ucsf.edu http://www.cgl.ucsf.edu/mailman/listinfo/chimera-users

Hi,

No rush, but thanks for checking.

Also I have another question regarding the benchmark scores in linux vs WinXP on the same machine.

The surface benchmark under WinXP seems to be almost twice as high as that in linux, but everything else is higher in linux:

Chimera-1.2154, Linux-2.6.12.5, ATI Mobility Radeon 9600 (128MB), fglrx-8.16.20 drivers, ECS G736 laptop, 1G RAM, P4 3.2GHz

Surface rendering size rate 32 249.0 64 76.5 128 20.0 179 10.3 250 5.3 214 7.2 196 8.6 187 9.4 183 9.9 181 10.1 182 10.0 Surface rendering benchmark: 181 Mesh rendering size rate 32 249.5 64 74.1 128 19.7 179 10.2 250 5.3 214 7.2 196 8.5 187 9.4 183 9.8 181 10.0 182 9.9 Mesh rendering benchmark: 181 Contouring size rate 32 371.8 64 70.4 128 13.4 179 6.4 153 9.1 140 11.6 146 10.1 149 9.6 147 9.9 Contouring benchmark: 146 Solid rendering size rate 32 76.7 64 38.2 128 19.2 179 13.9 250 9.8 214 11.5 232 10.6 241 10.1 245 10.0 243 10.1 244 10.0 Solid rendering benchmark: 244 Recoloring size rate 32 145.6 64 43.9 128 8.1 96 17.2 112 11.6 120 9.7 116 10.6 118 10.1 119 9.9 Recoloring benchmark: 118 Benchmark scores surface 181 mesh 181 contour 146 solid 244 recolor 118

Chimera-1.2143, WinXP SP1, ATI Mobility Radeon 9600 (128MB), driver 6.14.10.6436, ECS G736 laptop, 1G RAM, P4 3.2GHz

Surface rendering size rate 32 256.0 64 130.0 128 27.5 179 39.0 250 29.8 350 26.5 489 1.6 419 1.3 384 1.7 367 1.9 358 1.9 354 2.0 352 1.8 351 2.0 Surface rendering benchmark: 350 Mesh rendering size rate 32 135.0 64 49.0 128 15.0 179 7.4 153 9.1 140 11.6 146 11.6 149 11.3 151 9.9 150 11.0 Mesh rendering benchmark: 150 Contouring size rate 32 169.0 64 35.0 128 6.7 96 15.8 112 9.0 104 12.0 108 11.3 110 13.0 111 8.0 Contouring benchmark: 110 Solid rendering size rate 32 59.7 64 55.1 128 14.5 179 32.0 250 7.5 214 8.6 196 29.3 205 30.5 209 8.7 207 29.6 208 30.0 Solid rendering benchmark: 208 Recoloring size rate 32 75.0 64 14.8 128 1.7 96 4.7 80 5.7 72 9.0 68 11.6 70 11.6 71 9.7 Recoloring benchmark: 70 Benchmark scores surface 350 mesh 150 contour 110 solid 208 recolor 70

If you add up the scores you get for linux = 870 and for WinXP = 888. The difference I notice in the surface benchmark is that under linux after a surface size of 250 is rendered it renders a size of 214, whereas in WinXP it attempts a higher size of 350. How does it decide to attempt a higher or lower size? Is this a bug or some sort of malloc error in linux? It seems to me that if the linux surface score is >= the WinXP score, the overall score under linux would be much greater.

Thanks for your help, Sabuj Pattanayek

Thomas Goddard wrote:

Hi Sabuj,

We use Chimera with an ATI Radeon 9800 Pro on Win XP. I'm not sure we have tried Chimera version 1.2143 (the machine is down now) -- could be 1.2129 was the last version we tried on that machine. Also the drivers on that machine are probably a year old. If you are interested I can get more definitive information about that machine when we get it running again.

 Tom

Hi,

On multiple tests of the contour benchmark under WinXP I get anywhere from 108 to 120 with the mode being 115 or 116.

Under linux I always get a score of 146 no matter how many times I run it consecutively.

..Sabuj

Thomas Goddard wrote:

Hi Sabuj,

I'm surprised how different your Chimera benchmark scores are running Windows versus Linux on the same machine.

...

Chimera-1.2143, WinXP SP1, ATI Mobility Radeon 9600 (128MB), driver 6.14.10.6436, ECS G736 laptop, 1G RAM, P4 3.2GHz

Linux: surface 181 mesh 181 contour 146 solid 244 recolor 118 Windows: surface 350 mesh 150 contour 110 solid 208 recolor 70

The graphics drivers can make a big difference in the Surface / Mesh scores. I believe this has to do with the rules the graphics driver uses to decide whether to put the surface geometry in the graphics card memory (fast), or shuttle it from the main computer memory for each drawing (slow).

The difference you see in contouring scores (146 vs 110) is hard to fathom. This is not using any graphics -- it is just a calculation done on the computer CPU. Do you get close to the same number if you run the benchmarks twice? It could be that Windows is much slower at allocating memory -- that is also done during the contour calculation.

I would definitely *not* add the scores to get an overall measure of performance. Some may be much more important to you than others. Surface and contour are probably most important for looking at density maps. Surface is probably most important for looking at large molecular models. Recolor is relatively unimportant (it concerns recoloring of solid volume renderings).

The benchmark code works by increasing the volume data set size until the drawing or contour calculation speed drops below 10 times per second (considered "interactive"). Then uses bisection to find exactly what size volume data set can be handled at 10 frames per second.

See the Chimera User's Guide page for more info about the benchmarks. It's in the Tools section under Benchmark.

http://www.cgl.ucsf.edu/chimera/docs/ContributedSoftware/benchmark/benchmark...

We are currently enhancing the benchmarking tool to time display of molecular models.

Tom