> Yes the OpenCL side uses a totally different main loop so it basically > doesn't track the number of rays fired. We use a different main loop because > the pipeline is setup differently, in stages, to reduce thread divergence.
*nod* Good to know, and thought I recalled you saying as much when it was first implemented. What I didn't know is whether the timing is accurate -- presumably yes because the timers are high-level. There's no reason the main loop can't be changed. That's been a todo forever. :) > I usually just take the wallclock time into account for the GPU because there > is more to it than just the time spent to compute in the device and sometimes > bus transfers can get quite significant. But yes these statistics could be > improved. We just want to be really careful that we're not counting front-end app constructs like time spent blitting into a window (ie motivation for -o /dev/null) and can distinguish prep from post-prep time. We obviously want the whole user experience to be improved and tracked, but there is a significant mode of operation where prep and display times essentially don't matter or exist respectively -- namely librt's primary mode of operation embedded into third party simulation codes that just fire tons of rays and need partitions as fast as possible. Cheers! Sean ------------------------------------------------------------------------------ Check out the vibrant tech community on one of the world's most engaging tech sites, Slashdot.org! http://sdm.link/slashdot _______________________________________________ BRL-CAD Developer mailing list brlcad-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/brlcad-devel
