Reply to comment
Not for the bulk consumer market
October 22, 2007 by xgeorgio, 2 years 2 weeks ago
Comment id: 25610
Reply
About us
Science Blog was started in August 2002. It lives, breathes and eats press releases from research organizations around the globe. Most of what you read here are press releases from the outfits named in the stories themselves. Got a news story you think belongs here? Let's talk. The other half of the equation is blog posts from readers like you. So if you have an interest in science, please register and join others like you in an ongoing, vibrant dialog about what makes the world tick. Meantime, please take a minute to read our Privacy Policy and Site Disclaimer.Recommended
Andrew Sullivan Dan Sullivan Doug Arellanes Emmanuelle Richard Faith Sullivan Fred Bortz Heather Havrilesky Henry Copeland Kate Sullivan Ken Layne lagreenbean.com LAist.com Maggie Sullivan Matt Welch NewYorkology Opinion L.A. Ron Kaye Slashdot Steve Elzer & Family TheNewsroom.org Thor Garcia Tony Pierce Wonkette OlderThanks CNET for naming Science Blog one of your Blog 100!
Advanced branch prediction requires many cycles and large in-core cache, both mean very expensive h/w compared to current CPU technologies. Furthermore, this can only be truly effective when automatic deep branch prediction is required, like when using very high-level programming languages (usually for AI).
The current trend in desktop h/w is quite the opposite, that is to embed as many parallel cores as possible inside the PC, including general-purpose programmable GPU (graphics card) h/w. This can ease the burden of compatibility of instruction sets from classic x86 while exploiting the current CPU technologies to the max via massive parallelism.
Also, it should be noted that most heavy-processing applications today, like climate simulations, weather prediction, molecular dynamics, pattern recognition, etc, are designed DSP-like, mainly focusing on simple math instructions that can be easily ported to parallel or vector machines. Hence, a new instruction set for graph-like branch prediction seems too specialized and cost-inefficient for now.