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: Charles R who wrote (74906)10/11/1999 9:22:00 AM
From: vince doran  Read Replies (2) of 1575954
 
Chuck: Re CUmine benchmarks: I didn't make myself clear, in fact I find the numbers that one can project based on the experience of 'Emil' in optimizing his Athlon code to be very encouraging. I think you are correct that the bulk of the projected gains for Athlon (and the measured gains for CUmine) have come from code optimization, and since I had not seen those numbers pointed out on the thread I thought I would mention them. I also think that many of the business-oriented benchs may be split-decisions, however, as long as Athlon can hold a real advantage in FP, it should be somewhat protected from a price war (which, based on the rumored CUmine prices, Intel may not wage anyway).
Do you really think the Christmas season builds are already committed? I still think enough 820 boards will be delivered to carry Intel's standard into the battle.

On the AMD execution front, what's your guess on possible CC sandbagging re Dresden timeframe? MHz ramp on those .18s coming in november?

Vince

P.S. I noted in an AMD interview that the exact prefetch opcodes used in SSE are supported in 3DNow!, so Haim Barad's speculation that any such optimizations done for Intel should also work for Athlon chips is probably correct.
Report TOU ViolationShare This Post
 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext