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.
Technology Stocks : Intel Corporation (INTC) -- Ignore unavailable to you. Want to Upgrade?


To: Win Smith who wrote (148875)11/19/2001 8:58:52 PM
From: muzosi  Respond to of 186894
 
go read the latest jsscc & isscc proceedings about the p4 design and decide for yourself if intel is not doing a full-custom design. all indications are that for a very large portion of the design intel is pushing each transistor polygon by hand (except memory of course a small portion of which is designed by hand and then replicated).



To: Win Smith who wrote (148875)11/19/2001 9:11:47 PM
From: Tenchusatsu  Read Replies (3) | Respond to of 186894
 
Win, <FUD or complete cluelessness? I doubt if anybody has done a "layout the chip by hand" job on a major microprocessor since the original MC68000.>

Before you continue on with your juvenile retorts, you should know that the K7 design guys did indeed layout the chip by hand. This was revealed during the Dirk Meyer dinner presentation around May or June of 1999, as well as a Fred Weber presentation at MIT made months later. In the face of the trend toward further automation, the methodology was considered to be very unconventional. But obviously it succeeded.

It is unknown whether AMD is continuing to use hand-layout or not, especially with Hammer. I wouldn't doubt it, considering how the Hammer design borrows heavily from the original K7.

Tenchusatsu



To: Win Smith who wrote (148875)11/20/2001 6:37:19 AM
From: Dave  Read Replies (1) | Respond to of 186894
 
Win,

Pardon my ignorance, but...
Intel does full custom design, and any change like that is a major hassle. One of the big advantages that AMD has is they use a cell based datapath methodology, which allows for much faster design turnaround times.

If the "cell based" methodology is superior to Intel's custom design, why doesn't Intel implement this means?