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: Jim McMannis who wrote (73760)10/3/1999 1:27:00 PM
From: Steve Porter  Read Replies (1) of 1572717
 
Jim,

Athlon has 3 load store units, and trust me (I know from hacking code for these thing), it's no slouch in the 'moving memory around' category. If you stay inside that wonderfully large 128K L1 cache, the 'memory' performance of that thing is stunning..

Add to that the huge sceduler in Athlon, and deep pipelines and the 3 FPU(mmx/3dnow) units and you can do an awful lot of work in a very small amount of time.

My only gripe with Athlon is that load/store operations have a latency of 3 now (like the PII/PIII). This is a bit annoying, as it gets harder and ahrder to hide the latency of those (especially when you can have 3 instructions issued a clock.. you someone how have to find another 8 instructions before using the memory, or you 'waste' the cpu)... not easy at to do at all..

At least with only 2 pipe to fill on the PII you only had to find 5 insturctions to 'fill' the blank spots..

Regards,

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