What about creating a set of intrinsics for the SIMD operations, that way
SIMD could be used in the C(++) and hide the processor specifics until
compile time. Also that way you could use the operations transparently even
if the CPU doesn't have support for a specific instruction. Cross platform
SIMD optimization would become more accessible without delving into
processor specific code.
bye
From: "Alexander G. M. Smith" <agmsmith@xxxxxxxxxx> Reply-To: openbeos@xxxxxxxxxxxxx To: openbeos@xxxxxxxxxxxxx Subject: [openbeos] Re: app_server: MMX/SSE help wanted Date: Wed, 11 Aug 2004 08:32:13 -0400 EDT
Mat Hounsell wrote on Wed, 11 Aug 2004 10:27:33 +1000 (EST):
> By all means hand code assembler for MMX. But if Intel/AMD etc are going dual
> core then multithreading will give you better speed, than MMX could.
Which reminds me that it would be nice if it also worked on PPC processors, which means that having the C++ code around for the same algorithm would be useful.
- Alex
_________________________________________________________________ Express yourself with cool new emoticons http://www.msn.co.uk/specials/myemo