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.
Technology Stocks : Advanced Micro Devices - Moderated (AMD)
AMD 222.67+3.1%10:12 AM EST

 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext  
To: Bill Jackson who wrote (19803)11/20/2000 3:12:34 PM
From: PetzRead Replies (1) of 275872
 
Bill, I believe the performance gain realizable for SSE2 will be very limited compared to SSE performance gain, which is already in most applications.

There are a few apps that really love bandwidth, either to cache (Quake 3) or to main memory (Linpack with large data sizes, SPECfp, and certain streaming encoders that don't require much floating point.)

These apps are optimized for SSE already with pre-fetching to maximize the bandwidth.

SSE2 only really adds double precision floating point to SSE. The problem is that the P4 FPU is too weak to take advantage of multiple operations on each clock cycle, so SSE2 doesn't buy your much.

Because of the limited improvement in SSE2 vs. SSE, and the fact that even the Athlon will outsell the P4 until 0.13 micron, I don't see developers embracing SSE2.

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