Print 46 comment(s) - last by Icelight.. on Jun 17 at 10:25 AM

The pair will investigate the use of Radeon GPUs for physics processing

Physics processing has promised to bring more realism into the world of PC games since AGEIA first announced its PhysX card. To date, physics processing in most video games hasn’t become the big draw that many hoped it would.

AMD and Havok announced this week that they plan to optimize physics processing for AMD hardware. Havok already has a well defined user base with over 100 developers using its Havok Physics engine and it says that 300 leading game titles currently use its Havok engine.

AMD says that Havok’s physics engine will scale well across the entire AMD line. This includes its processors and its ATI Radeon video cards. AMD says that it and Havok will investigate the use of the Radeon GPU to manage aspects of in-game physics.

AMD isn’t alone in looking to the GPU to process physics. Rival NVIDIA purchased physics hardware and software maker AGEIA. NVIDIA announced that it was purchasing AGEIA in February 2008. The purchase virtually ensures that physics processing makes it to NVIDIA graphics cards in force.

Intel purchased Havok in September 2007 and at the time Intel said that Havok would become a key element in its visual computing and graphics efforts. Despite the proclamation at the time the purchase was announced by Intel, physics has not been seen as a key component in Intel’s product strategy yet.

Comments     Threshold

This article is over a month old, voting and posting comments is disabled

RE: It is very simple.
By AnnihilatorX on 6/14/2008 7:24:38 AM , Rating: 2
So you are saying Processors are not hardware?
Well in any case I think only Intel's Labarre and AMD's Fusion CPUs would be able to handle future physics calculations.

RE: It is very simple.
By decapitator666 on 6/15/2008 4:40:23 AM , Rating: 2

RE: It is very simple.
By Icelight on 6/17/2008 10:25:00 AM , Rating: 2
It's a parallel project currently in the works at Intel France.

RE: It is very simple.
By decapitator666 on 6/15/2008 4:47:18 AM , Rating: 2
Why? Graphics chips are becoming fully programmable to do streamed floating point calculations at breakneck speed.. Whereas CPUs floating point capability will go towards the GPU model.. They will meet in the middle with each having advantages..

Look at the CUDA programming language for the newer nvidia GPUs Its just a question of programming. The physics calculations are similar to graphics calculations so why would a graphics card not be able to do it...

RE: It is very simple.
By wavetrex on 6/15/2008 5:01:28 PM , Rating: 2
I was talking about AMD's GPU hardware, not their processors, but it seems that the "Labarre" dude is unable to read properly, beside writing invented names.

And why the frak was I rated down ?

I explained a very good reason why Intel wants this licensing to go on.

RE: It is very simple.
By tmouse on 6/16/2008 8:44:59 AM , Rating: 2
And why the frak was I rated down

My guess; someone did not like your 2 year prediction.

"Well, there may be a reason why they call them 'Mac' trucks! Windows machines will not be trucks." -- Microsoft CEO Steve Ballmer
Related Articles
Update: NVIDIA to Acquire AGEIA
February 4, 2008, 5:31 PM

Most Popular ArticlesAMD, Zen Processor might power the upcoming Apple MacBook Pro
September 30, 2016, 5:00 AM
Leaked – Samsung S8 is a Dream and a Dream 2
September 25, 2016, 8:00 AM
Are you ready for this ? HyperDrive Aircraft
September 24, 2016, 9:29 AM
Inspiron Laptops & 2-in-1 PCs
September 25, 2016, 9:00 AM
Apple’s Siri Speaker is a Game Changer
September 26, 2016, 5:00 AM

Copyright 2016 DailyTech LLC. - RSS Feed | Advertise | About Us | Ethics | FAQ | Terms, Conditions & Privacy Information | Kristopher Kubicki