Very interesting read, thanks. That forecast chart is hilarious. “We’ll sell $35 billion! Okay, maybe $30 billion. Or is it $25 billion? 15? 10? 5? Hey, doesn’t anyone want an Itanium?”
It’s surprising that Intel so grossly misjudged the capabilities the Itanium design. Poor performance and poor scaling appear to be the key reason why the design failed. What went wrong there?
Then we have bizarre statements like this:
“I think it’s doing very well,” said Lisa Graff, general manager of Intel’s high-end server group. She points to gains in Itanium’s scaled-back mission of replacing Power, UltraSparc and other reduced instruction set computing (RISC) chips and observes that half of the world’s 100 biggest companies use Itanium systems.
“I think Itanium is still the architecture for the next 20 years,” Graff said. “It’s the newest architecture that has come out. It has the headroom. I think the RISC architectures will run out of steam.”
Why would anyone buy another expensive Itanium server when true 64 bit systems based on x86/64 are available?
The Itanium processors still have the oodles-of-cache advantage, which probably helps in some fairly specific applications and is really their only saving grace at this point.
I don’t think it’s really a big deal that it failed to deliver, since VLIW architectures are still fairly underresearched and experimental technologies often hit dead ends, but it’s the sheer magnitude of the effort that went into this that’s interesting. You’d think they’d have spent a bit more time in the lab trying to demonstrate clear gains before rolling it out and just hoping it gets better in the future…
(I also had no idea that it was originally HP’s idea, not Intel’s.)
The Itanium processors still have the oodles-of-cache advantage, which probably helps in some fairly specific applications and is really their only saving grace at this point.[/quote]
And then the question is why buy an Itanium when you could buy a POWER 4 or 5? :lol: