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: Elmer who wrote (135512)5/19/2001 12:37:18 PM
From: muzosi  Read Replies (1) | Respond to of 186894
 
you should really take off your "intel can do no wrong" blinders from time to time. your interpretation makes sense only if intel gave you a clear pmax and had a way to turn off thermal monitor. the reality is in a cooling system designed for 64 w (ptyp for 1.75v 1.7ghz p4), thermal monitor is required and it will turn on when you run some applications. from another document developer.intel.com we infer that the pmax is over 85 w for the same processor. how much do you think it is going to cost to design a cooling system for that and why isn't intel giving the oems an option to design for it and turn off thermal monitoring ?



To: Elmer who wrote (135512)5/19/2001 12:38:22 PM
From: Dan3  Read Replies (2) | Respond to of 186894
 
Re: Let me walk you through it slowly and then maybe it will make sense.

Hmmmm.

Consider this link:
cluboc.net

Factory spec 1.0 to 1.33GHZ Athlons run at 1.4GHZ to 1.6GHZ with a good fan/heatsink combo.

Meanwhile we're learning that factory spec 1.7GHZ P4s sometimes run at 1.2GHZ under load.

If AMD specified heatsink/cooling systems as large as the ones Intel specifies for P4, and then added throttling to reduce performance under load, thunderbird would bin out at about the same speeds as P4.

Palamino would bin out 20% higher than that.

IPC for thunderbird is higher than P4.

IPC for Palamino is higher than thunderbird (and much higher than P4).

Thunderbird/Palamino is about 2/3s the size of P4 and consequently less than half as expensive to produce (yields go down as size goes up). Usually this isn't a big deal for parts as profitable as CPUs, but as the price war extends to mobile and SHV servers, the fact that AMD' revenue needs are about 1/10 those of Intel's could become significant. That 1/10 revenue need is supporting a marketshare of 1/5 to 1/4.

The architecture for the Athlon core design and the processes for its manufacture were developed while AMD was nearly bankrupt and resource starved.

Now AMD has paid off its debt (except for a small subsidized Dresden loan) and has earned substantial new resources to accelerate design, process, and manufacturing development.

Despite the intensifying price war with Intel AMD still looks to be in excellent shape.

OTOH, we are in the early part of a period in which Intel is attempting to reverse AMD's gains by exhausting what started out 6 months ago as a $13 Billion war chest it accumulated over 20 years. (Current Assets - Current Liabilities) At present rates, we may be as little as 12 months away from AMD's financial position being superior to Intel's, but it's going to be a brutal 12 months, and Intel will likely regain some marketshare as it exhausts its $13 Billion.

Earnings from AMD this year may well disappoint, while the losses from Intel may be downright horrible.

Dan

PS - If the IBM/AMD .13 / copper / SOI process were to pan out as hoped, AMD could dominate the mobile and SHV server markets 18 months from now.

In which case, Intel and VIA can battle it out for the 20% of revenues available to the lower 75% of the market.

e.g. We Will Bury You!

:-)