Just pray that plan A works for Intel
Intel has mooted a 'Plan B' in case its Itanium processor doesn't live up to hype, David Ludlow investigates
As all secret agents, master villains and politicians know, you have to have a plan B, just in case it all goes horribly wrong.
Intel seems to understand this and has set to work on its own plan B. It covers the company's back in the eventuality that its Itanium processor doesn't turn out to be the success the company told everyone it would be.
The story was broken by the San Jose Mercury News, which discovered that Intel was starting to integrate 64-bit functions into the next version of its Pentium family, Prescott.
Of course, Intel refused to comment on the story. Yet it did not deny that its actions were taken in the light of the poor performance of the Itanium so far. But, if you read the signs, there has been a wavering in confidence at Intel about Itanium for a while.
Itanium was hyped as the next big thing a long time before it was ready. But come the launch date, where was it? Intel decided a stealth launch would be better, and played down interest, preferring to call the first incarnation of the chip a prototype.
But if all the rumours are true and Intel is working on something else, I think we should be worried. Itanium, whether it's commercially successful or not, has already claimed the lives of its big competitors.
Hewlett-Packard was the first to jump on the bandwagon, helping Intel develop the chip using its own PA-Risc architecture and a considerable slice of investment. Now, all HP's PA-Risc servers are Itanium-ready, pushing customers towards an Intel future.
Next came Compaq, which sold its Alpha product line to Intel. This forces future developments of Tru64 Unix to run on an Itanium platform.
In the meantime, it's not at all clear where we stand. Itanium is poised to take over the enterprise world, but it has to wait. It has barely learnt to crawl so it will be some time before it emerges wearing suits and impressing people at cocktail parties.
If the decision is made to drop the line in favour of plan B, a similar amount of time will presumably elapse before the new processors are mature enough. Until then there would be a very uncomfortable gap.
Of course, Sun and IBM would do very well out of it, but they don't have the economies of scale in chip production that Intel enjoys. Itanium is supposed to give us high-powered performance at mass-production costs, while widening software choice.
If the processor was to fail, then the potential headaches, not to mention lack of choice in OS for the user, doesn't make this situation very funny. I, for one, hope that rumours of Itanium's death have been greatly exaggerated.