Bookmark and Share

Tags

32nm 40nm 45nm AMD Apple ASUS ATI ATIC Atom Business Cypress E-Book Evergreen Fermi Flash Geforce Globalfoundries GT300 Intel Microsoft Mubadala Nokia Nvidia Radeon Semiconductor Sony SSD TSMC USB Windows

News

NVIDIA Corp.’s chief architect David Kirk said in an interview that the company would make a graphics processor with unified pixel and vertex processing engines – an approach backed by Microsoft Corp. and ATI Technologies, but denounced by NVIDIA earlier – in future. The executive said that from manufacturing standpoint such a chip would be easier to make, but harder to design.

“We will do a unified architecture in hardware when it makes sense. When it’s possible to make the hardware work faster unified, then of course we will. It will be easier to build in the future, but for the meantime, there's plenty of mileage left in this architecture,” David Kirk said in an interview with Bit-tech.net web-site.

Late last year ATI and NVIDIA revealed their standpoints on the so-called unified shader processing engines within a graphics chip. ATI said that such an approach was very efficient and would allow to dynamically allocate resources of a chip. The company has developed a graphics processor for Microsoft’s Xbox 360 console, which actually has unified shader processors. NVIDIA late last year said that unified shader processor was not really efficient for both vertex and pixel processing compared to specifically optimized pipes.

Microsoft pushes unified shader language for pixel and vertex shaders in its next-generation Xbox 360 game console ad well as graphics API of Windows Longhorn – Windows Graphics Foundation 2.0. As a result of that graphics hardware designers should deliver their chips with unified shader engines at some point in future in order to more efficiently support the new API. But at this point NVIDIA thinks the approach is not something required tremendously in short-term future, which may be a feasible, as Longhorn is not expected to be launched earlier than in mid-2006.

“It’s far harder to design a unified processor - it has to do, by design, twice as much. Another word for ‘unified’ is ‘shared’, and another word for ‘shared’ is ‘competing’. It’s a challenge to create a chip that does load balancing and performance prediction. It’s extremely important, especially in a console architecture, for the performance to be predicable. With all that balancing, it’s difficult to make the performance predictable,” Mr. Kirk said.

Discussion

Comments currently: 0

You must log in to add comments.

Forgot password? Registration

remember me



Related news

Latest News

Sunday, November 22, 2009

11:34 am | Voice Communications Set to Decline in Two Years – Analysts. Analysts Predict Dramatic Increase of VoIP Popularity

Saturday, November 21, 2009

11:44 pm | Barnes and Noble: Nook E-Book Readers Are Sold Out. E-Book Reader from B&N Sold Out Before Release

7:53 am | Cell Network Operators Set to Become Largest Mobile Internet Devices Sellers – Analysts. Mobile Network Operators to Gain Strength in Devices

Friday, November 20, 2009

10:11 pm | ATI Seeks Its Best to Ensure More Radeon HD 5-Series Supplies – Company. Additional Number of DirectX 11 Graphics Boards is Incoming

11:56 am | Fusion-io’s SSD Setup Reaches 1TB/s Aggregate Bandwidth. Fusion-io Gets Contracts from Government, Creates World’s Fastest SSD Setup

10:06 am | Notebook – the Most Desired Christmas Gift, Says CEA. Notebooks, Players and HDTVs Top Christmas Presents Wish List

9:11 am | Ebay Completes Skype Sell Off. Skype No Longer Belongs to Ebay