Bookmark and Share

Tags

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

News

Graphics chip designer NVIDIA is describing opportunities for general purpose computations on graphics hardware, something which has a prerogative of enthusiasts for years.

One of the courses NVIDIA hosted at Siggraph conference presented a detailed introduction to general-purpose computation on graphics hardware (GPGPU), with emphasis on core computational building blocks ranging from linear algebra to database queries. The course was held by NVIDIA’s engineers, which may indicate that computations beyond computer graphics on graphics chips are explored at NVIDIA.

Modern graphics processing units (GPU) are far more complex than contemporary central processing units (CPUs) and even though they work on much lower frequencies, the number of operations they can perform at once may exceed than of the CPUs. Unfortunately, in order to make GPU calculate something other than graphics, a completely new programming language, or a special translating software, should be written.

Several groups of enthusiasts have explored possibilities to use graphics processing units for general purpose computing, however, no project has actually become a success.

Discussion

Comments currently: 2
Discussion started: 08/11/05 06:34:07 PM
Latest comment: 08/12/05 03:20:13 AM

[1-2]

1. 
would be very cool if some big software development house such as Adobe would figure out a way to do this.. maybe not all operations but just offloading the heavier operations from the cpu if possible, would be good to know that the most expensive part of the computer would be used for more than gaming =)
[Posted by: Silver  | Date: 08/11/05 06:34:07 PM]

2. 
"Modern graphics processing units (GPU) are far more complex than contemporary central processing units (CPUs)"

Not even close... just because they have more transistors? I don’t think designing a pipeline and vertex processor, and then REPLICATE it several times is very complex....
I will not even talk about the AA "engine" that have been the same since Geforce4 < COPY PASTE

"and even though they work on much lower frequencies, the number of operations they can perform at once may exceed than of the CPUs."

Not even close, CPU can process anything throne at it very fast, look at the software that exists in the all world.
GPU is more like an DSP, do that and only that fast enough.
[Posted by: kaz  | Date: 08/12/05 03:20:13 AM]

[1-2]

You must log in to add comments.

Forgot password? Registration

remember me



Related news

Latest News

Wednesday, November 25, 2009

5:50 pm | U.S. Patent Office Again Rejects Rambus’ Claims Against Nvidia. Nvidia Wins Another Round in Patent Dispute with Rambus

2:36 pm | EA Montreal to Concentrate on High-Def Games, Lower Focus on Wii. Large Video Game Developer to Re-Focus on HD Blockbuster Titles

11:58 am | AMD to Describe 32nm x86-64 Processor at Chip Conference [UPDATED]. AMD to Reveal Power Trimming Technologies of Next-Generation Mobile Chip

Tuesday, November 24, 2009

11:50 pm | Nvidia to Start Shipping Next-Generation Tegra to Developers “Soon”. Nvidia Readies Second-Generation Tegra SoC for Handhelds

10:37 pm | Despite Netbook Popularity, Consumers Still Want Notebooks – IDC. Even in Asia, Consumers Still Prefer Notebooks over Netbooks

4:04 pm | Imagination Intros Processors for “Internet Everywhere” Consumer Electronics. Imagination Presents Connected Processors for CE Devices

3:33 pm | Sub-$99 Blu-Ray Players Black Friday Deals Available, But Not a Lot. Walmart to Sell BD Players for $78 on Black Friday

12:27 pm | Microsoft Sued for Banning Third-Party Xbox Memory Cards. Memory Cards Supplier Sues Microsoft

11:55 am | OCZ to Release External USB 3.0 Solid-State Drive. OCZ USB 3.0 SSD Incoming for Consumer Electronics Show

7:52 am | Nvidia’s CEO Expects Underpowered Mobile Devices to Gain Popularity. PC of the Future – Web-Based Device with 4G Connectivity, Says Chief Exec of Nvidia