On 09/03/2010 08:55, Ludwig Nussel wrote: > Dominic Fandrey wrote: >> On 09/03/2010 06:30, Michael Menegakis wrote: >>> On Sat, Mar 6, 2010 at 1:44 PM, Ludwig Nussel <ludwig.nus...@suse.de> wrote: >>>> So, has anyone tried the new build yet? >>>> http://www.ioquake3.org/files/angst/ioquake3-1.36_SVN1778-18.3.x86_64.exe >>>> http://www.ioquake3.org/files/angst/ioquake3-1.36_SVN1778-15.2.x86.exe >>> >>> This one appears to be working fine >>> http://www.ioquake3.org/files/angst/ioquake3-1.36_SVN1778-19.1.x86_64.exe > > Great, thanks for testing :-) > >>> performance appears to be relatively good too. >>> >>> I don't know why it's not _better_ than the x86 version on windows - >>> it is here when comparing the two on linux - but anyway; probably >>> related to the x64 thing not being as mature on windows yet. > > I don't see why it should be faster. It's not optimized at all while > the x86 vm tries to eliminate some "useless" instructions at least. > >> There is considerable overhead in the 64bit VM code. I'd assume that >> this is the reason. > > You mean the argument conversion? Or do you have anything else in > mind?
Yes, that's what I'm talking about. The way I understand it, this code is called for every frame. I remember I intended to try tackle that, but I'm swamped with work at the moment. Regards -- A: Because it fouls the order in which people normally read text. Q: Why is top-posting such a bad thing? A: Top-posting. Q: What is the most annoying thing on usenet and in e-mail? _______________________________________________ ioquake3 mailing list ioquake3@lists.ioquake.org http://lists.ioquake.org/listinfo.cgi/ioquake3-ioquake.org By sending this message I agree to love ioquake3 and libsdl.
