Archived

This topic is now archived and is closed to further replies.

Memory transfer optimizations

This topic is 6078 days old which is more than the 365 day threshold we allow for new replies. Please post a new topic.

If you intended to correct an error in the post then please contact us.

Recommended Posts

hey currently i''m using a movq loop to transfer memory (sysmem-sysmem and sysmem-vidmem) I use ddraw. My question is: 1:is there a faster way to copy memory and/or can someone give me *the best* optimized code for it? i have also heard about prefetch instruction (or so..) can i use it under vc++?(or do i need masm/whatever) how does it exactly work? anyone with an example? and the second part is about sysmem-vidmem copy. my sysmem-sysmem copy routine is almost twice the speed of sysmem-vidmem , is there a way to optimize it? any tricks? or maybe using agp in 2d graphic? thnx for all answers/help.

Share this post


Link to post
Share on other sites