SI
SI
discoversearch

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor.  We ask that you disable ad blocking while on Silicon Investor in the best interests of our community.  If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.
Politics : Formerly About Advanced Micro Devices
 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext  
To: Process Boy who wrote (74517)10/7/1999 1:56:00 PM
From: Haim Barad  Read Replies (3) of 1572928
 
I'll give you an example. The Direct3D API (starting with DX6.1) has an optimized pipeline for PIII. It was almost completely written in C++ and does take advantage of SIMD. The 3D Winbench scores are based upon Direct3D performance. So the answer to your question is yes. (keep in mind that there also is an optimized AMD path written with 3DNOW instructions... although they had to do it in assembly)

HOWEVER!!! The C++ compiler DOES NOT AUTOMATICALLY use SIMD without the user declaring SIMD data structures (e.g. 4-wide Single Precision FP). Since the SPEC benchmark code (both INT and FP) can't be changed, then the "optimized" scores CAN'T BE USING SIMD.

Haim
Report TOU ViolationShare This Post
 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext  

HomeMailHotSubjectMarksPeopleMarksKeepersSettings
Terms Of UseContact UsCopyright/IP PolicyPrivacy PolicyAbout UsFAQAdvertise on SI
© 2025 Knight Sac Media.  Data provided by Twelve Data, Alpha Vantage, and CityFALCON News