BBC BASIC for Windows
« GFXLIB2: ChangeBitmapRGBLevels »
Welcome Guest. Please Login or Register. Apr 6th, 2018, 12:28am
ATTENTION MEMBERS: Conforums will be closing it doors and discontinuing its service on April 15, 2018. Ad-Free has been deactivated. Outstanding Ad-Free credits will be reimbursed to respective payment methods.
If you require a dump of the post on your message board, please come to the support board and request it.
I tried writing an MMX version, but it was about three times slower! MMX is simply not suited to table look-up, despite the pextrw and pinsrw instructions which were added at the SSE stage to address that issue to a degree.
I tried writing an MMX version, but it was about three times slower! MMX is simply not suited to table look-up, despite the pextrw and pinsrw instructions which were added at the SSE stage to address that issue to a degree.
Thanks for trying anyway! Your attempt is appreciated.
Perhaps a re-ordering of the instructions in the current non-MMX code; somehow doing all the reads first - source pixel reads, then the table reads, then do all the writes might lead to faster execution. Registers aren't exactly in abundance, so it'll be tricky. I may still give it a crack.
Rgs, David.
« Last Edit: Oct 15th, 2011, 12:29pm by David Williams »
Logged
