Hi Tom, Thank you for your insight, I appreciate it. This helps me with my decision! Good luck on your work! Best, Joe On Wed, Mar 4, 2020 at 1:08 PM Tom Goddard <goddard@sonic.net> wrote:
Hi Joe,
If you are dealing with modest size systems, e.g. molecules with less than 10,000 atoms or density maps less than 100 Mbytes in size just about any new desktop or laptop will do. If you deal with 500,000 atom structures or 4 Gbyte density maps then something nicer than Intel graphics would help render fast, an Nvidia or AMD GPU, like an Nvidia RTX 2070 is a high-end very high performance GPU for about $400, but a $100-200 GPU video gaming GPU would also be a big improvement. The large data will also benefit from having enough memory, 16 Gbyte, or 32 or 64 if you have really big data like lightsheet microscopy time series that are 30 Gbyte data sets. The CPU speed is also important for non-graphical steps, like computing hydrogen bonds or adding hydrogens to your 50,000 atom structure. ChimeraX rarely uses multiple cores for compute so fast single core speed would be better over high number of cores.
Just to reiterate, any modern desktop will do if you are looking at modest size data, and you only need to look at the higher end computer hardware if you work with large molecules and maps.
Tom
On Mar 4, 2020, at 10:07 AM, joe viterbo <joeviterbo@gmail.com> wrote:
Hi everyone,
I am about to purchase a PC desktop computer to run Chimera. Regarding the hardware, should I focus more on CPU speed & cores or should I focus more on a powerful GPU? Or should I try to find both?
Thanks for your time!
Best, Joe _______________________________________________ Chimera-users mailing list: Chimera-users@cgl.ucsf.edu Manage subscription: http://plato.cgl.ucsf.edu/mailman/listinfo/chimera-users