Hi Mario, Unfortunately Mac OS 10.9 (Mavericks) has a graphics driver bug with Nvidia graphics cards that causes a crash in the graphics driver when Chimera uses a feature called instancing (OpenGL routine glDrawElementsInstanced()). This advanced feature makes display atoms as spheres and bonds as cylinders about 20 times faster. You can see this in the Chimra graphics benchmark tables http://plato.cgl.ucsf.edu/trac/chimera/wiki/benchmarks where some graphics cards render Ball and Stick at 200 frames/second (for 30,000 atoms, pdb 1f4h) and some at 10 frames/second and some at 2 frames/sec. All of the slow rendering is for graphics or old Chimera versions that do not use instancing, and all of the fast ones use instancing. Chimera 1.8.1 and the daily build disabled instancing for Mac OS 10.9 with Nvidia because it crashes every time. There is nothing we can do about this except wait for Apple to fix their graphics driver. I actually don't expect it to be fixed in Mac OS 10.9.1 or anytime soon because Mac is not a video gaming platform. But it is hard to tell because Apple is so secret you can't even find out how mandy bug reports they have for this issue or if they are working on it. So the only way to get better performance is to go back to Mac 10.8 or use a different computer. Tom On Nov 14, 2013, at 2:08 AM, Mario Dejung wrote:
I found Tom Goddard's post about the Mavericks bug and the change on the daily build with the disabled advanced graphics. So I am not sure, if my problem is related to the crashing of chimera.
My problem is the speed of chimera under Mavericks. I work with molecules of about 20000 Atoms and rotating a PDB of this size is very painful. I didn’t had problems with Lion.
I am wondering now, if I am the only person with speed problems under Mavericks, or if other people have similar issues.
Is there a way to improve the speed of chimera under Mavericks?
Kind regards Mario _______________________________________________ Chimera-users mailing list Chimera-users@cgl.ucsf.edu http://plato.cgl.ucsf.edu/mailman/listinfo/chimera-users