On Monday 23 January 2012 14:47:44 Michael Cronenworth wrote: > Marko Vojinovic wrote: > > That is, assuming one can > > actually trust the glxgears output numbers. > > Marko, please educate yourself. > > http://wiki.cchtml.com/index.php/Glxgears_is_not_a_Benchmark > > If you start typing a reply back with "yeah I've seen that" then you > need to read it again. This wasn't the point of my post. I understand perfectly well that glxgears is not a proper benchmark. It is not even supposed to be a benchmark, since it doesn't test all capabilities thoroughly enough. However, my "assumption" above is about the validity of the fps numbers, for whatever glxgears is *actually* *rendering* . Namely, if it reports 4000 fps, does that mean that the gears animation has been actually rendered 4000 times per second? Or not? If not, then the reported numbers are incorrect, and one should file a bug. If yes, then it is a major waste of resources, since the display can only show 60 out of those 4000 frames in one second. Either way, anything bigger than 60 fps in glxgears means that something somewhere is very wrong (either as a bug, or by design). That was my point. Best, :-) Marko -- users mailing list users@xxxxxxxxxxxxxxxxxxxxxxx To unsubscribe or change subscription options: https://admin.fedoraproject.org/mailman/listinfo/users Guidelines: http://fedoraproject.org/wiki/Mailing_list_guidelines Have a question? Ask away: http://ask.fedoraproject.org
