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: A. A. LaFountain III who wrote (52630)	3/15/1999 10:21:00 AM
From: Elmer	Respond to of 1578303

Re: "1) SSM doesn't compete against the Xeon. In fact, it could eventually support it. It does compete with Intel's shared-bus architectures (450NX chip set for 4-way and Intel's Corollary unit's Profusion product for the 8-way market). With a bus-oriented approach, there would be a need for arbitration and the greater the load, the greater the arbitration. If I understand it correctly, this leads to a scalability question, which Profusion will address by use of a second bus." All P6 generation processors have a private bus between the processors for arbitration etc. It doesn't require any FSB cycles. There is FSB arbitration with the bus controller when a snoop takes place but not between the processors. Re: "2) The network approach (SSM) obviates the need for arbitration due to the fundamental difference in architecture. Scalability is constant." There would still be a need for arbitration when a snoop is required, just like the P6 bus. EP

To: A. A. LaFountain III who wrote (52630)	3/15/1999 11:10:00 AM
From: DRBES	Respond to of 1578303

re: "Pretty interesting, don't you think? - Tad LaFountain" Yes, Tad yes. Regards, DARBES

To: A. A. LaFountain III who wrote (52630)	3/15/1999 2:19:00 PM
From: Petz	Read Replies (1) \| Respond to of 1578303

Tad, re:SSM, I actually have a patent on an 8x8 crossbar switch for use in multiprocessor systems. patents.ibm.com I notice that it is currently referenced by 34 other patents, including 1 from Intel ("US5604878: Method and apparatus for avoiding writeback conflicts between execution units sharing a common writeback path", granted 2/18/97", patents.ibm.com Unfortunately, Intel doesn't have to pay any royalties to me or anyone else; Oryx went belly up and no one acquired the patent. Petz

To: A. A. LaFountain III who wrote (52630)	3/15/1999 2:26:00 PM
From: gbh	Read Replies (1) \| Respond to of 1578303

so they sell for $500 each, while cost per part averages $200; Tad, where do you get the $200/part cost? This seems very higher to me. Microprocessor Report, the best source I know for this type of info, shows a K6-3 manufacturing cost of $45. The K7 will be roughly the same die size, therefore roughly the same cost. Of course the Slot card that the K7 lives on will add cost, but not $150. Gary

To: A. A. LaFountain III who wrote (52630)	3/16/1999 2:21:00 PM
From: DRBES	Respond to of 1578303

I would like to refer you to my post # 52682 : Message 8346000 I do not claim this is a likelihood but it drops out of the numbers that I am able to generate. Remember, iNTEL is encumbered by a strong case of NIH and evolutionary legacy technology; while AMD has started out with a clean sheet of paper and some non-ignorable encumbrances as well. I am looking at a, less than likely, best possible case scenario. ANY THOUGHTS? (that you can share here) Regards, DARBES