Silicon Investor (SI) -- The First Internet Community

STOCKTALK

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: Elmer who wrote (62335)	6/18/1999 4:39:00 PM
From: kash johal	Read Replies (3) \| Respond to of 1572172

Elmer, Re: Mobile 380 K6III laptops. Well you can now go out and buy one for $1800 at compaq.com. For about 1/2 the price of a 400 Mhz dixon. Oh I know you like performance specs. Check out the performance numbers for a 380 K6III laptop vs the new mobile Dixons. WoW!!!! You can get them from amd's web site if you're interested. Regards, Kash

To: Elmer who wrote (62335)	6/18/1999 11:06:00 PM
From: Cirruslvr	Respond to of 1572172

Elmer - RE: "Intel SpecInt95 scores are actually SPECint_base95 scores and the highest reported score is 24.4, not 24.3 so using the K7 SPECint_base95 score of 25.1 you would get: 25.1/24.4 = 102.86% or a 2.86% improvement, not +5.76 as claimed." No matter how much you try to close the performance gap, the K7's L2 cache ran at 1/3 clock speed so the numbers will only improve with production half and eventually full speed cache.

To: Elmer who wrote (62335)	6/19/1999 12:04:00 AM
From: Cirruslvr	Respond to of 1572172

Elmer - RE: "Intel SpecInt95 scores are actually SPECint_base95 scores and the highest reported score is 24.4, not 24.3 so using the K7 SPECint_base95 score of 25.1 you would get: 25.1/24.4 = 102.86% or a 2.86% improvement, not +5.76 as claimed." I have another message for you about the same quoted thing. JC wanted me to pass this along to you - "I noticed Elmer comparing Intel's best specint with what I had reported as K7's specint baseline. Please let him be informed that the 25.1 was a typo, and the int baseline is actually the same as the int peak (as it is on x86 products, generally). The value is 25.7." So if you take that into consideration, you still get a slightly better than 5% advantage for the K7 when compared to the highest SpedInt95 score which you quoted. Add that to the K7's slower than expected L2 cache and you can expect greater performance for the final, half-speed L2 cache K7 versus the full speed Xeon at the same MHz.