On Dec 11, 2010, at 5:50 PM, Raistmer wrote: >> From Lunatics beta thread: > " > Quote from: cenit on Today at 00:35:57 > Put for example that I have a HD 4870 and a HD 5870 in the same host, with a > unified app it would be easy to get full performances (with both app, I would > have to use the non-HD5 for both!) > > Devices with different capabilities should have different plan classes first > of all. And if there are different plan classes I see no problem to use few > different app each fine tuned for particular device. > > " > > Could we expect such different plan classes for OpenCL apps for ATi HD4xxx > and HD5xxx that drastically differ in capabilities (no scratchpad local > memory on HD4xxx at all) ? >
Just check at run time. There aren't such huge differences necessitating an entirely different application. At most you probably have different kernels for the two architectures. You can then ship both, and then just use the ones corresponding to the detected GPU. Even if host side code also needs to be different it still isn't a problem. There's no real need for a special plan class in this particular situation. _______________________________________________ boinc_dev mailing list [email protected] http://lists.ssl.berkeley.edu/mailman/listinfo/boinc_dev To unsubscribe, visit the above URL and (near bottom of page) enter your email address.
