Advanced Micro Devices Inc (AMD)

[neye_eve]

not addressed to me, but I'll answer

What made you rout for AMD in 2002 when they were well behind Intel and losing droves of money?

don't forget - many of us didn't, and sat it out.

I think it's amazing, though woefully typical, to see the same people who did rout for AMD in 2002 based on empty promises, now grow skeptical that Intel can ever turn around in the same way.

Ok, you're using the word "ever". I'd say most people here think AMD will have an absolute x86-** advantage for a few months, but don't stretch it into ever. Again, I understand you addressed the post to fpg, so this may not be relevant to your post.

Though you have to admit that if Intel shrink Northwood and gave it the same micro-architectural enhancements that Prescott has, it would be 10-15% higher performance per clock and probably not be faced with the same power issues, allowing it to hit 4GHz even easier.

I feel like Elmer should be here, saying "woulda shoulda coulda", or something like that

I believe AMD will continue to innovate, but I feel that Intel has more low hanging fruit just by moving a Dothan derivative onto the desktop.

won't agree or disagree with that, as I don't have a crystal ball, but it makes sense to me.

As for IPF, I am still surprised that last year's chip continues to outperform AMD's yet to be released Opteron 250 on several benchmarks, still.

I surprised that a puny little Opteron continues to outperform the big bad 6MB cache 1.5Ghz I2 on the same software platform. what a disgrace.

Yet IPF is also limited by a slower front side bus and memory interface, and software development is still in relatively early stages.

True, and aHT isn't supposed to increase in speed (oh, wait, it is), never use higher clocked memory (oh, wait, it is), and no apps will ever benefit from being compiled to 64bit native for x86-64 (oh wait, they will). sounds pretty ambiguous to me, not a clear cut win for IPF.

Moreover, it's a second generation micro-architecture, and many of the current learnings will be applied towards the third.

True, and K8s are much older than the Itanium family of chips :p Understood that IA64 is younger than x86, but the incarnation fo A64s are younger than then Itanium family, and can also benefit from optimized coding, compilers, and chip design.

There is literally tons of headroom left, while x86 is nearing a peak.

I think you accidentally left out some words. The proper quote from, oh, 10 years ago or so is:

"There is literally tons of headroom left in RISC chips, while x86 is nearing a peak."

Yeah, I think that's about what we heard from Jobs and mot and ibm in roughly 1994, when the first PPCs were being put into macs. Sucks that x86 has been sucking wind all these years and not increasing processing power, while the RISC chips have been real barn burners.

Amazingly, Intel is still making out very well with the Prescott core in spite of all its deficiencies. Intel is strong company with many assets, many strong engineers, and very good marketing.

Leave out the (IMO obsequious) word "amazingly", and you have complete argument from me in this statement. Actually, the rest of your post, too, I agree with:

The next couple years are AMD's to enjoy if they can execute, but Intel is working towards next generation hardware and they have a goal to exceed. That's a much different position as opposed to being on top and underestimating the competition.

neye