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.
Politics : Formerly About Advanced Micro Devices -- Ignore unavailable to you. Want to Upgrade?


To: Petz who wrote (26342)12/2/1997 1:20:00 PM
From: Time Traveler  Read Replies (3) | Respond to of 1572777
 
Petz,

"I'm now convinced the ONLY reason they can get that CPU to 300 MHz is the extremely high current they pass through the transistors."

You have convinced yourself with nonsense. The reason why the over-all DC current is higher is because of higher frequency of operation. A few months ago, we have been discussion power consumption, and where have you been? Does the equation (P is proportional to C * V^2 * f) ring any bell? Why do you and Ali insist on residing in your fantasy-like make-believe world?

"...L2 cache, which is another 1.4 amps"

How much current do the 512K L2 cache on a Socket-7 motherboard and the support chips draw? More than 1.4 amps!

"A P2-300 draws 66% more current than a K6-233, although its clock rate is only 29% higher."

K6/233 and its L2 cache together consume more power than a P-II/300. Remember K6/233 runs at higher Vcc than P-II. P oc C * V^2 * f.

No doubt is that Slot I has a superior power handling capability. It is also a multi-chip/device module which means Intel can add these filter capacitors in this thermal brick to allow the CPU running at an ever higher frequency. From a technical point of view, it is absolutely a superior solution, not to mention the MIB as well. Slot I is NOT a kludge.

John.



To: Petz who wrote (26342)12/2/1997 1:21:00 PM
From: Ali Chen  Respond to of 1572777
 
John, <reason they can get that CPU to 300 MHz is the extremely high current they pass through the transistors. A P2-300 .. draws 66% more current than a K6-233, although its clock rate is only 29% higher.>
Yes, your observation opens another dimension to this complex problem.
Good observation, I agree. Thanks.

Ali



To: Petz who wrote (26342)12/2/1997 1:35:00 PM
From: StockMan  Respond to of 1572777
 
P,
Re -- The fact is, any semi company can design an FET with any Idsat they like, if they can come up with a thermal scheme to keep the chip from self-destructing.

Stop polluting this thread with your nonsense posts. Take your amateruish BS elsewhere.

Stockman



To: Petz who wrote (26342)12/2/1997 2:26:00 PM
From: Kevin K. Spurway  Read Replies (2) | Respond to of 1572777
 
Re: "The fact is, any semi company can design an FET with any Idsat they like, if they can come up with a thermal scheme to keep the chip from self-destructing. Intel chose the quick and dirty approach of max-ing out Idsat at the expense of requiring the expensive SLOT 1 packaging. Then they called SLOT 1 a feature rather than a kluge, and the majority of techno-lemmings believed them."

If the argument that you and Ali are making is correct (that the K6 is effectively working harder/more effectively at a given clock speed than the chip inside a Pentium II cartridge) then it stands to reason (since PII remains faster at a given clock than K6) that the PII derives *all* of its performance advantage from the Slot 1 architecture (L2, dual independent bus, thermal performance, etc.). If this is in fact the case, then the cartridge is probably well worth the extra cost.

In any case, it seems that the cartridge makes adequate comparisons of process technology between the PII and K6 almost impossible. To be honest , I'm beginning to think that at a high level of process technology (which both AMD and Intel possess) the difference between a good process and a great process is not measured in clock speed, but in the ability to get good yields of high end (high clock) chips.

Things will get interesting when AMD rolls out K7 and both companies are using the thermal brick.

Kevin