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: kash johal who wrote (52881)3/18/1999 7:48:00 PM
From: Scumbria  Read Replies (1) of 1578895
 
Kash,

Let's assume that K7 is 30-40% faster than equivalently clocked Xeon.

I will be very surprised if K7 is more than 10% faster on integer code than Xeon. It may even be slightly slower than Xeon at the same clock speed, due to the longer latency L1 and L2. (This would be more serious except for the fact that it is mitigated by a larger L1 on K7.)

The AMD's will use interleaved memory as I understand it.

Does that mean for a server that one needs 2x this memory for K7.

Everyone uses interleaved memory these days. I believe that K7 is using a LOI (low order interleave) scheme. The bank select changes as you walk through sequential addresses. The choice of LOI vs. HOI makes no difference for the amount of DRAM required. It does affect page hit rates, resulting in differences in latency.

I'm not aware of anything about K7 which inherently requires higher system costs than Xeon.

Scumbria
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