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 : Advanced Micro Devices - Moderated (AMD)
AMD 221.39-0.1%3:59 PM EST

 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext  
To: Ali Chen who wrote (64270)11/23/2001 9:49:58 PM
From: kapkan4uRead Replies (1) of 275872
 
<It says that 0.18um (100nm Leff) technology produces SRAM that operates at barely 1GHz. How come P4 can have "full speed" caches if it is technologically impossible to run 0.18 SRAM at 2GHz>

Yes, 1.06GHz without a guard band. This is why 2GHz is the hard limit on 180nm.

If you follow Intel Development Forum link at chip-architect.com and look at the P4 die, you will see a huge area of the chip, almost 50%, is very likely running at 1GHz on a 2GHz P4. The half clocked parts of the chip most likely include from left to right, top down on the die:

1. FSB logic.
2. 256KB Unified L2.
3. Front-end branch prediction.
4. Instruction decoder.
5. Trace Cache branch prediction
6. Trace Cache.

It is even possible that the Floating-Point and MMX/SSE/SS2 unit is half clocked. If the FP unit is half clocked, then it would bring the half clocked die size to around 70% of the chip.

I think that this is a bomb that is waiting to explode in Intel's face in both PR and leagal sence.

Can you imagine if Class Action lawyers get a hold of this?

Kap
Report TOU ViolationShare This Post
 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext