If Itanium fails and Hammer succeeds, Intel is in a world of hurt.
Well, when you put it that way, of course, but what are the odds of that happening? We can put it in binary:
Itanium Hammer
0 0
0 1
1 0
1 1
1 = success. 0 = failure.
I think some form of the bottom case will happen. The worst case you cite, or both failing are pretty remote. Itanium a success and Hammer a failure would take an AMD of K5 or K6 days to pull off. But Intel is still Intel, and they can make things happen through their OEM partnerships that AMD doesn't have a chance for. Of course, McKinley and supporting infrastructure, and Madison and other subsequent chips have to be production, performance and RAS worthy on some kind of schedule from now on.
1. Dell's dropping of the IA-64 workstation product, apparently due to stillborn sales. This by itself is not conclusive, as Dell is a "high volume" company, not a big iron company, and the decision to drop a particular box is not terribly significant. For example, they may be regrouping with a McKinley version. Or they may wait for volumes in general to increase.
Better be the latter selections. Who else does Dell turn to for bigger than PC or up to 8-way 32 bit servers. I think they'll be there with McKinley boxes. Can't imagine they'll pass on 64 bits.
2. IBM continues to push their Power4 architecture for high-end use in their own systems. Why is the IA-64 not gaining in such uses?
IBM has some pretty good McKinley based products in the works, like the Summit chipset based one, which they advertise as having mainframe-like RAS features, scalable to 16 way, maybe more. If they gave up on their Power4 architecture, everyone would be astounded.
3. According to someone whose expertise I value, who works on the IA-64, "it's just a dog." (I can't say much more for fear of making his identity apparent to some Intel employees who may be reading this.)
That's certainly disturbing to hear. Like the other guys said, is it a current (McKinley) opinion. It's also quite different from what Compaq said last summer when they announced they were throwing everything 64 bit to Itanium. They were satisfied with McKinley's performance, but even more by Intel's roadmap for performance improvement for IA64. They said they saw an approximate doubling every year, for a while. Has any of this changed, I wonder?
I'm not trying to downplay the importance of Itanium. Heck, I'm a server guy, big servers, and I've been excited about its prospects ever since I heard of Merced. Itanium failure or severe future slippage would be severe to an Intel stockholder, but I don't think quite as severe as Hammer failure, or future severe slippage.
Tony |
