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 (55003)	4/9/1999 5:37:00 PM
From: RDM	Read Replies (1) \| Respond to of 1582415

We estimate that the K7 will be, on average, 40% faster than the PIII in 3D games. The PIII needs 14 cycles for 14 multiplies and 12 cycles for 12 adds to calculate one vertex (one point), or 26 cycles. The K7 needs 14 cycles for 14 multiplies, and while these 14 cycles are calculated, the 12 adds are also calculated. For me and my uses, the best part was the possibility of 2X Pentium III (at same clock rate) for double precision: This means that the K7 has 4 times the peak bandwidth of a PIII in double precision FP calculations! No wonder some rumors say that the high end Winstone '99 runs twice as fast on a K7-600 than on a PIII-500. Keep in mind, we are talking about rumors, but the K7 will kick some serious butt in the Intel workstation market. aceshardware.com

To: Tenchusatsu who wrote (55003)	4/9/1999 6:29:00 PM
From: Scumbria	Read Replies (2) \| Respond to of 1582415

Ten, I don't know anything about AMD's plans other than what they have published about their CPU architectures. They are impressive and based on very solid principles, which is more than I can say about Merced. I see AMD's competition as coming from Willamette, which is Intel's sensible answer to performance. System performance will become a much higher priority. It is not hard for me to imagine MP migrating onto the desktop in a relatively short time frame (i.e. 3-5 years.) I don't see IA-64 going anywhere besides servers in the near future. The IA-64 architecture offers little or no benefit for desktop applications, though it is well suited for the predictable tasks which servers are devoted to. Scumbria

To: Tenchusatsu who wrote (55003)	4/9/1999 6:33:00 PM
From: Gopher Broke	Read Replies (3) \| Respond to of 1582415

Your question makes the assumption that the super-$1000 marketshare will continue to exist. The current pricing of "high-end" processors is ludicrous. When a competitor emerges that can compete in all processor performance categories (no I am not saying this is necessarily AMD, Mr Engel, so hold the sarcasm) and we see true price competition, rather than the rather skewed competition we are seeing at present, then it is inevitable that all models or processor will come down in price substantially. Who of us does not want to own a gigahertz processor? Call it a server system if you want, but I will fork out a few hundred bucks extra for one. So, what I anticipate is not that AMD moves out of supplying the sub-$1000 PC category, but that Intel is increasingly forced down into it. We (as consumers) will do well. I see it as inevitable that Intel will do badly, simply because they cannot hold the monopoly they have enjoyed. AMD will only survive if they pull AHEAD of Intel in the performance war. If they only achieve parity then they will end up the loser in the price wars.