Intel Says Nuts To Reports Of 45-Nm Penryn DelaysIntel Says Nuts To Reports Of 45-Nm Penryn Delays

Unsubstantiated reports out of Taiwan claim that AMD's recent quad-core bug troubles have caused to Intel push back the launch of the first 45-nm versions of its desktop Core 2 Quad processors. Only problem is, Intel says it ain't so.The Digitimes story that's attracting attention today speculates that Intel sees no imminent competitive threat from AMD. The story draw from this the conclusion that Intel doesn't want to cannibalize its current top-of-the-line 65nm parts, and so it's delaying the launch of three desktop 45-nm Penryns from January until February or March.

This all sounds plausible enough so that it prompted me to go straight to Intel to get a yea or a nay out of them. Prefacing his comments with the caveat that Intel normally doesn't comment on such speculative stories, a spokesman at the chip giant e-mailed me back: "We've been public since November that we plan to introduce a host of new mobile and desktop processors based on our 45 nm and reinvented transistors in Q1, and there are no changes to that plan."

Another source close to the company told me that it's laughable to think that Intel would delay any shipments based on rumored delays of an AMD product that's not even widely available yet.

OK, but to make matters more confusing, there's another potential angle to the story. According to The Tech Report, which gets a shout-out for breaking the story about AMD's erratum 298 (aka the Barcelona bug), Intel's 45-nm Penryns are being pushed back due to a little erratum of their own, which requires attention.

Hey, I don't make the news, I only report it, so I'm sorry to have to be in a situation where I'm essentially writing about dueling speculative reports. But if you wanna know what I really think, it's this:

All processors have errata (bugs). With processors getting ever more complex, and with Intel and AMD under ever-increasing pressure to play "can you top this?" (not only external pressure, but internal, too), there will continue to be bugs. Most of them will be given workarounds; the more serious ones will fixed via mask revisions. Bugs are called bugs for a reason; they're mostly minor nits. Showstoppers are almost always buzzed out before fabrication.

The upshot is that no bug is ever again likely to cause the kind of backlash we saw from the 1994 Pentium FDIV glitch. (My 15-ms of fame, here.) That said, AMD has nevertheless taken a hit because its Barcelona/Phenom bug came amid the high-profile launch of the company's first quad-core parts. However, I continue to believe AMD will recover from what is more of a public-relations hit than a black eye to its technology.