wbmw,
the first from a so-called employee that happened to get quoted from the Mercury Times (and subsequently reproduced at all the major news publications), and the second from an off-hand comment by an HP exec, that just happened to get exclusively captured by our favorite Internet Media Tabloid (oops, I mean second favorite - sorry, Mike).
Why are you trying to belittle this? The first story is just how most of the stories break, the second story gives remarkable corroboration by somehow who is in the know, in a direct, quoted, attributed interview. (I bet the guy got into a lot of trouble).
So now that the cat is out of the bag, Joe, what does Intel gain by hiding it?
The usual, "It's unconfirmed report, Intel does not comment on unannounced products, Itanium is Intel's 64bit strategy, so go ahead and by it. You can't go wrong by buying Itanium. It will not be canceled. Really. We don't want you to wait and see what develops. Nothing is going to develop. We want you to commit to Itanium and buy it now. Resistance is futile. Itanium is it. Damn it. Itanium is the only thing on our roadmap, just like Rambus was (oops didn't mean to remind you of that), just like there was no point in going to PC-133, because it was unreliable technology, and it will never make it to Intel line up to supplement / replace Rambus, Intel will never release 815 chipset supporting PC-133, so buy Rambus! Damn it! Oops, didn't mean buy Rambus NOW. Buy Rambus when we tell you to buy Rambus, discard it when we are ready to discard it. Now we are telling you to buy Itanium, and we are not even going to mention anything else that could distract you from the highest priority of today, which is to buy Itanium. I can promise you you will not see me here 2 years from now telling you to buy something else. 2 years from now, I will either be telling you to buy Itanium, or I will be at a different position, wearing dual hats of Chief Tchnology Architect at large - Emeritus, and Intel campus dog catcher"
Actually, it doesn't take a genius to realize that something doesn't add up. There is more here than meets the eye, yet you are willing to jump to the obvious conclusion.
When something else develops, new information is released, making some other conclusion obvious, I will probably jump to that. What's wrong with that? This is not something like a philosophy of life that you refine, build upon. The question is much simpler. Will Intel expand their 32 bit instruction set to 64 bit? Yes or No. The information that is out in the open cat point to one or the other. Right now it points to "Yes".
Can you explain why you are so hesitant to admit that the facts are leaning to "Yes" answer, when it is the obvious conclusion as of today, and lean to "No" answer, for which the only evidence is "No Comment"?
That's because you just can't get over how great it is that Intel could have no faith in their own ISA, and the very notion that a Plan B exists confirms all of your preconceived beliefs. Therefore, you are now convinced that IA-64 will fail, and all the architectures that Intel managed to tumble in the process - Alpha, PA-RISC, maybe others - will leave plenty of room for AMD to corner the market.
It does not take a genius to realize that Itanium has been a complete and utter failure up to this point. After years of development and 100s of millions if not billions spent, years of delays, there is nothing to show for it.
Some re-evaluation is in order. Few people argue with success, but a failure of such a huge proportion necessarily has a lot of second guessing. Having second thought is logical thing. I doubt that Intel realized day before yesterday that Itanium is in trouble, so Plan B must have started some time ago.
As far as implications for AMD, you may have noticed that I have been very skeptical about pouring resources to server market at expense of weaker mainstream to performance desktop, beyond dual configuration, since these need to be tackled before higher end segment will even consider AMD. That's why I am stressing over lack of dual channel DDR in Clawhammer, since this is AMD's bread and butter, while Sledgehammer with Dual channel DDR, high end server market is a pie in the sky.
Sound too good to be true, Joe? Maybe that's because it is.
Well, you will see me asking questions about this, since I want more information, not less, if I am to invest in AMD in the future, and buy computers in the future.
Joe |
