• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

larrabee thoughts

snarfbot

Senior member
for what its worth more competition is a good thing, so regardless of how it performs it should benefit us all.

since its coming from a company like intel, i doubt they would let it die like the s3 chrome, and kyro chips and the others that came and went over the years.

now for a trip down memory lane. back when quake 1 came out, i had an amd 486dx4 100, it ran it adequately i suppose, but it was optimized for pentium processors.

im not sure, but im think i remember reading benchmarks that said a pentium pro at 100mhz could run quake at 512x384 at 60fps or something, i cant remember exactly but it was alot better than what i was getting at 320x240.

the chips in larrabee are going to be clocked at 1000mhz minimum, that would mean each core could run its own instance of quake 1 at pretty good framerates at high resolution.

this of course is stupid and pointless but that brings me to this.

http://strlen.com/gfxengine/fisheyequake/compare.html

perspective correct rendering could be implemented very easily on larabee, anyone who games on a big screen and tries to turn the fov up to match knows how distorted things get in the peripherals. something like the above could be done in software for pretty much any game, since its already translating dx and gl instructions to its own software renderer.
 
for me i'm looking at larrabee from a "coprocessor" perspective.
not sure if it'll support opencl or some other implementation, but regardless it should be a number crunching workhorse 🙂
 
Back
Top