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 215.84+0.4%12:03 PM EST

 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext  
To: Petz who wrote (105775)11/14/2003 5:46:25 AM
From: Joe NYCRead Replies (1) of 275872
 
Petz,

On 64 bit extension:

In terms of SSE(x) to do 64 bit integer arithmetics, it is a little kludgy. SSE resides in FPU, and has no communication with normal registers. It communicates only with 64 or 128 bit MMX registers or memory. You can't do branching based on it, and in general the operations on MMX registers are limited. Generally, an architecture is considered say 32 or 64 bit based on the size of the normal integer register. So the SSE is a non-starter on many fronts.

So you really need to redo the ALU and switch to 64 bit integer registers - which may brake a lot of code, and without 64 bit addressing, it is still not a full 64 bit architecture, and in general, if Intel is going to bite the bullet and do an x86 extension, I don't see any reason to do an half assed job, that you are suggesting they will do.

The reason is that to this day, Intel is not sure about Itanium. It is plan A, and there is no plan B. A half assed 64 bit extended x86 is still no plan B. So if plan A fails, Intel doesn't have a plan B, and would still need to come up with a completely new plan B to be a longer term solution.

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