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 -- Ignore unavailable to you. Want to Upgrade?


To: Petz who wrote (43756)12/21/1998 8:06:00 PM
From: Tenchusatsu  Read Replies (3) | Respond to of 1572381
 
<Anand's benchmarks underestimate superiority of the K6-3 because he used only 64M RAM (definitely skimpy for a high-end machine, especially for NT!).>

Yeah, but he also has 2 MB of L3 cache on the motherboard in the AMD K6-3 system. I'm not sure whether that makes a difference or not in the benchmark scores, or whether actual system makers are going to use a full 2 MB of L3 cache on the motherboard. I do know that Price Watch lists 512K of pipelined-burst cache at almost $50, so 2 MB of L3 cache on the motherboard can approach $200 in additional costs.

<Finally, IMO the SLOWEST K6-3 when it is available in a few weeks will be the K6-3-450. From a marketing point of view, it doesn't make sense to have two chips at the same speed. So who cares if an underclocked K6-3 is still faster than Intel's finest?>

First, AMD still hasn't released a date. Intel, on the other hand, already released two release dates, January 4th for the Socket Celerons and notebook Pentium II Enhanced (a.k.a. Dixon), and March 1st for Katmai. How do you know that the K6-3 will be "available in a few weeks"?

And second, the "underclocked" K6-3 is a good indication of its potential performance in notebook computers. It's no doubt that AMD will be positioning the notebook K6-3 against Intel's mobile Dixon and mobile Coppermine processors.

Tenchusatsu