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: andreas_wonisch who wrote (43453)	6/8/2001 3:52:41 PM
From: Gopher Broke	Read Replies (1) \| Respond to of 275872

Andreas, I have only a layman's knowledge of electronics but what you say does not make sense to me. If AMD have changed the design rules for Palomino then the optimal operational voltage may well change as well. My understanding is that voltage is chosen as a compromise between clock speed and processor lifetime. Therefore comparing Palo and Tbird voltages is illogical.

To: andreas_wonisch who wrote (43453)	6/8/2001 5:00:44 PM
From: ptanner	Read Replies (1) \| Respond to of 275872

Andreas, Re: "power consumption is a quadratic function of voltage" I think you mean P = k * C * V^2 k = constant for particular chip C = clock speed V = core voltage Or so I have "learned" from others on the thread. Regards, -PT

To: andreas_wonisch who wrote (43453)	6/8/2001 7:17:28 PM
From: milo_morai	Read Replies (1) \| Respond to of 275872

Interview on Athlon MP digitimes.com

To: andreas_wonisch who wrote (43453)	6/9/2001 3:48:07 AM
From: Petz	Respond to of 275872

andreas, here's my guess at Palomino's clock frequency scaling problems. AMD lowered the drive current on them to meet the low power requirement -- after all, heat was the main reason that Athlon could not go higher than 1.33 GHz. The voltage breakdown was well above 1.75v, but the old Athlon would burn up under a worst case environment above 1.8 volts. (Although the new stepping does go beyond 1.33, I have no idea why, and that's another story.) The may have also decreased the transistor size slightly to go along with the lower current. So it was supposed to scale to higher MHz, because power was thought to be the biggest problem holding back MHz and because the reduced feature size should have sped up the transistors. I suspect that process variability has led to some transistors channel length becoming a MHz limit on the processor. Has AMD actually figured out how to reduce the variability so Palomino can scale to higher MHz? I suspect that have, at least partially if the 1.4 GHz server chip makes an appearance shortly. Meanwhile, it's safe to increase Vcc because the 1.75v limit for Athlon was not based on breakdown, it was based on heat, and Palomino has met it's goals for less power admirably. Sander's has been stressing Palomino as a notebook chip for quite a while now and I think we didn't want to hear that message. Petz