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.

Technology Stocks : Advanced Micro Devices - Moderated (AMD) -- Ignore unavailable to you. Want to Upgrade?

To: Petz who wrote (20202)	11/22/2000 12:17:32 PM
From: fyodor_	Read Replies (2) \| Respond to of 275872

<Petz: The P4 1.5 GHz on 0.18u has extremely little headroom, even for a 1.6 GHz part, unless Intel implements Herculean cooling technology.> Didn't Intel formulate some new requirements, where the heatsink had to be coupled to the case or something? -fyo

To: Petz who wrote (20202)	11/22/2000 12:18:20 PM
From: andreas_wonisch	Read Replies (1) \| Respond to of 275872

Petz, Re: The 70 degC and 72 degC case temperature limits aren't quite as bad as the P3 in the sense that they are case temperature limits instead of junction temperature limits. The Tj can probably be 5 degrees higher. T_j of 1.4 and 1.5 GHz P4 is at 95°C as listed in the techdocs. This is the same temperature AMD specifies, so I think with the current stepping both CPUs are roughly equally scaleable in a relative sense since they consume the same power (this will change, of course, with Palomino). Edit: Actually it's 85°C (see ftp://download.intel.com/design/pentium4/datashts/24919801.pdf maximum ratings). Andreas

To: Petz who wrote (20202)	11/22/2000 12:43:30 PM
From: Dan3	Read Replies (1) \| Respond to of 275872

Re: The P4 1.5 GHz on 0.18u has extremely little headroom, even for a 1.6 GHz part Wow, Scumbria was very right and Paul DeMone was very wrong. Scumbria called both the IPC problem and the scaling limits. I don't think the die size issue has been fully appreciated, either. If their process is working every bit as consistently on those 3GHZ gates as it was on the 1GHZ P3 gates, they're still going from about 200 chips/wafer to about 70 chips/wafer. 2 new FABs and a moderation in the rate at which demand is increasing will help them stay in the market share fight, but it's still going to be a tough couple of quarters for Intel. Everybody starts fresh on .13 in 2H 2000, so they'll have another chance to catch up then. It should be easy for Intel to move from AL to CU and .18 to .13 concurrently. Meanwhile, AMD faces the challenge of a straight shrink. :-) Dan