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: Elmer who wrote (26262)11/30/1997 1:23:00 AM
From: Ali Chen  Read Replies (1) of 1572629
 
Elmer, <AMD used these same techniques in designing the K6.>
You apparently have a vague understanding of the difference between a general RISC idea and particular implemenation.

AMD stresses everywhere that they have the shortest instruction latencies. When a design uses the "short tick" rule, the latencies become longer (by definition). Therefore, it is apparent that the gesign with longer latencies (P-II) has shorter "tick rule" than a design with shorter latency (K6). I am not expecting you to comprehend this and will not elaborate this issue further.

<This is why the Intel design is so effective is SMP servers> This is not entirely true. The Intel servers have better PERFORMANCE/PRICE ratio. This is not the same as "so effective" as compared to real servers. Try to understand this: in business of servers, people need absolute performance, not relative performance/price ratio. What would be a reason for a server that is VERY cheap but makes only a few transaction per minute (TPM)? The Intel's best TPM/$ is a common misconception and Intel's marketing hype.

<and why you don't see any socket7 SMP servers. > Wrong again, dear Elmer. The socket7 targeted the desktop/laptop market sector, and never was intended for servers. That's why you cannot see a thing that never intended to be as such. That simple.

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