The Inquirer-Home

Intel claims it invented 64-32 technology

We love the sound of breaking, [Eva] Glass
Mon Dec 06 2004, 11:59
WE REPORTED last Friday on a conversation we had with Ron Curry, here.

Ron Curry was formerly on the Intel Itanium team and said that he'd talked to software ISVs about the Itanium shortly after the firm announced its plans for the chip.

He said a far more interesting thing too. According to Curry, Intel's "competitor" - that's AMD of course, took advantage of the work it had done. He said that the competitor announced its move into the 64-32 arena largely based on the work Intel had done.

We've been scratching our head about this one all weekend. On the 25th of August 2003, the Reuter news wire reported CEO Craig Barrett as saying that it had no plans to add 64-bit extensions to its existing technology.

As early readers of the INQ will know, we'd been hearing rumours about Intel's Yamhill project way back even when we were at the "Other Plaice".

At the February Intel Developer Forum, Dr Barrett confirmed that Intel was indeed adding 64-bit address extensions for the Xeon and the Prescott.

At a Q&A held at the same time, Dr Barrett said that 64-32 architecture was compatible with AMD's architecture, or rather broadly compatible. AMD's comment then, as we reported here, described Intel's adoption of 64-32 tech as the biggest about face ever.

We contacted AMD for comment on Intel's claims. Richard Baker, marketing manager for AMD Northern Europe, said: "That's interesting. It's only the last beta build [of Windows XP Pro 64] that has been certified for Intel's Nocona". Previous versions of the preview OS were certified only for AMD microprocessors, he said.

So frankly we're a little puzzled as to how AMD could possibly have stolen Intel's thunder on this work. Intel's stance on the 64-32 technology its competitor was undertaking was, for several years, that this wasn't the right approach to microprocessor architecture.

In a sense, Intel is still maintaining that stance, as you can see from the slide we published in our Friday story. But be that as it may, the fact of life is that by the end of 2005 Intel's desktop and server processors will have this technology inside. A little like the existing internal designs of the most current Intel microprocessors are already dabbling with dual core tech. µ


Share this:

blog comments powered by Disqus
Subscribe to INQ newsletters

Sign up for INQbot – a weekly roundup of the best from the INQ

INQ Poll

Heartbleed bug discovered in OpenSSL

Have you reacted to Heartbleed?