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 : Intel Corporation (INTC)
INTC 39.88-1.7%9:56 AM EST

 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext  
To: Process Boy who wrote (89618)10/7/1999 1:18:00 AM
From: Charles R  Read Replies (2) of 186894
 
PB,

<- <Athlon Pro 700: Specint - 37; Specfp - 42>
Please clarify what configuration and what bus and when do you expect AMD to post these mumbers.>

We are comparing two pre-release products/platforms here. So, the exact data for either systems is unavailable. Nonetheless, the lead that Athlon has over CuMine is clearly not something to ignore. As I have noted on the AMD thread, the original Spec numbers are based on running spec on Intel binaries and grossly understated Athlon performance. This was well known for a long time. If you have any doubts I suggest you look up the AMD thread around middle of June about Athlon performance comments immediately after Dirk Mayer's presentation.

<<The miss penalty in CuMine will increase as core speeds ramp because of the lower hit rates due to smaller cache. So, expect the comparisons to get more in favor of Athlon as speeds ramp up.>

??>

As processor speeds ramp, the memory/CPU speed disconnect increases. Every time there is a miss in the L2 cache, the CPU has to go to memeory and that delay will get worse when measured in CPU clock cycles. A system with a smaller cache will have a higher miss rate and will see higher penalty on performance. Makes sense now?

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