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: Paul Engel who wrote (39263)10/14/1998 9:30:00 AM
From: Steve Porter  Read Replies (1) of 1573696
 
Paul,

There doesn't seem to be a clear picture that the K7 has any major advantages
beyond a 128 K L1 cache.

All the extra execution pipes may go starved for data if the cache misses become a
bottleneck.

Well that's what I find really curious. I mean they have 3 integer and 3 fp units in there. Sure if you can keep them full the performance will be great, but (and it's a big but), there is a high-latency (by comparison) L1 cache and the options for L2 cache do not seem to include synchronos cache at this time.

IMHO, it would have been better to do something along the lines of Sharptooth for the K7. That is put 64K of low latency (1 cycle) L1 cache on the CPU, 128K or 256K of on chip L2 and have a BSB for L3 cache that would start at 2MB and go from there.

Steve
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