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: Tenchusatsu who wrote (54976)	4/9/1999 2:46:00 PM
From: Scumbria	Read Replies (2) \| Respond to of 1583294

Rookie, I personally predict that the K7 could initially achieve similar integer performance as a K6-III clock-for-clock, and perhaps a 20% increase in floating-point over a Pentium III. That's my WAG, but it seems like a reasonable hype-to-truth ratio. However, I don't believe the K7 will really debut above an initial clock speed of 550 MHz. As much as the K7 tries to imitate the Alpha, it's never going to be a high MHz Alpha. In the absence of any information other than a bogus Winstone 43 from JC and another bogus story from Sharky, how can you make such claims? Scumbria

To: Tenchusatsu who wrote (54976)	4/9/1999 2:59:00 PM
From: Kevin K. Spurway	Read Replies (1) \| Respond to of 1583294

Re: "I personally predict that the K7 could initially achieve similar integer performance as a K6-III clock-for-clock, and perhaps a 20% increase in floating-point over a Pentium III. That's my WAG, but it seems like a reasonable hype-to-truth ratio. However, I don't believe the K7 will really debut above an initial clock speed of 550 MHz." I think I agree! But the integer performance will be that of a K6-III with 2 MB L2 (i.e. rather faster than any PIII out there). I also think that the K7 architecture will scale MUCH better than the PIII architecture as memory technology improves between July 1999 and Q3-Q4 2000 (when we'll see Willamette). So, oveer time, we'll see an effect similar to what has happened with the K6 vs. the Pieceofsh*t-II/III given the K6's socket 7/100 MHz FSB limitations, but in the other direction (i.e. to AMD's advantage). Kevin

To: Tenchusatsu who wrote (54976)	4/9/1999 3:10:00 PM
From: Scumbria	Read Replies (1) \| Respond to of 1583294

Ten, The 21264 actually got to around 24 SPECint The Compaq Digital Alpha Server 8400 with one 21264 CPU scores SPECint95 30.3 spec.org The main reason for the bloated estimates of 21264 performance were due to expectations of higher clock speeds. Scumbria