X
Innovation

Tuesday - Blue Screen Of Death

Tuesday 15/04/2003We've got one of them new-fangled Intel 3GHz Pentium 4s with the 800MHz front side bus, nicely fitted inside a big fast box from one of Intel's friends. Slap on the benchmarks, press the big red button, hang onto hats, socks and other easily detached items of peripheral clothing, and...
Written by Rupert Goodwins, Contributor
Tuesday 15/04/2003
We've got one of them new-fangled Intel 3GHz Pentium 4s with the 800MHz front side bus, nicely fitted inside a big fast box from one of Intel's friends. Slap on the benchmarks, press the big red button, hang onto hats, socks and other easily detached items of peripheral clothing, and... bang! Blue Screen Of Death. It's not a normal BSOD. No reams of hex dumps, register contents and stack vomit, just a terse message and a couple of strings of zeros. Hm. Try again. And again, the latest and greatest ends up in a virtual heap on the floor, more Norman Wisdom than Paula Radcliffe. We're digesting this when the news comes in that Intel, bless, has withdrawn the chip due to a "small anomaly observed in validation testing on a small number of chips" We have no trouble believing this, but despite much needling we're unable to find out what the anomaly is, nor why it would only appear on a few chips -- and if so, why they couldn't just junk those. Nobody gets 100 percent yield, specially not on a new and particularly fast design. So it must be something more than that -- but at least Intel hasn't got as far as an FDIV recall. We don't know that our ill computer is really demonstrating whatever it is that Intel is trying to hide. But it remains a possibility. If only I had a scope to hang off the bus...
Editorial standards