Is there a faster way for DSA ? (Test EXE)
While thinking about some stuff... i know there are posiblitis to put small and complete routines direct into the cache to get a lot of performance boost!! I dont know how to use nor code the X86 cache directly... i did this only for a very fast drawing routine on 68030+ amiga, years ago!
Would it be possible to have PureBasic commands saying the compiler (if possible), putting this loop or routine for faster access/operations direct into the cache? For example CacheStart() CacheEnd() ??? The compiler could alert an message if the user defined section (code) doesnt fit into the cache!? (i know cache has different sizes from cpu to cpu)...
using cache is just one possible way for me atm... isnt it?
Would it be possible to have PureBasic commands saying the compiler (if possible), putting this loop or routine for faster access/operations direct into the cache? For example CacheStart() CacheEnd() ??? The compiler could alert an message if the user defined section (code) doesnt fit into the cache!? (i know cache has different sizes from cpu to cpu)...
using cache is just one possible way for me atm... isnt it?
va!n aka Thorsten
Intel i7-980X Extreme Edition, 12 GB DDR3, Radeon 5870 2GB, Windows7 x64,
Intel i7-980X Extreme Edition, 12 GB DDR3, Radeon 5870 2GB, Windows7 x64,