fluffy grue voted 1 on this story.
Yay! Ars Technica! I keep meaning to get in a habit of reading it to replace the tech news I'm missing from Slashdot. Most of Slashdot's tech "articles" were just links there anyway.
A bit more on-topic: it's nice to finally see what's probably a fair comparison of the different SIMD implementations. However, due to my crappy connection I've only managed to read the first page so far. I like their subtle reference early on to the Connections Machine, though it'd have been nice if they were a bit more explicit (and if they'd pointed out that in its peak, it was used more as a prop on the set of Jurassic Park than an actual computer). Also, they kind of make it sound that SIMD is some innovation new to microprocessors, but with the CM, the individual 1-bit ALUs (all 65536 of them) were separate chips, and that Crays have had SIMD capabilities since the Cray-1. I guess that's mostly trivia history, though, with little bearing on modern implementations (Cray's vector processing was more like explicit pipelining than parallel execution, for example).
"Is not a quine" is not a quine.
I have a master's degree in science!
[ Hug Your Trikuare ]