Nvidia Discusses DirectX in Windows 7
PrintAccording to Nvidia, Windows 7, which recently reached RTM, will be the catalyst that propels the concept of GPGPU computing into the mainstream.
"Previously, GPUs were almost exclusively limited to rendering and accelerating graphics and video," Chris Daniel, product manager for software at Nvidia, wrote in a Microsoft Partner blog. "With the introduction of Windows 7, the GPU and CPU will exist in a co-processing environment where each can handle the computing task they are best suited for. The CPU is exceptionally good at performing sequential calculations, I/O, and program flow, whereas the GPU is perfectly suited for performing massive parallel calculations."
Nvidia went on to say that by introducing the DirectX Compute in Windows 7, Microsoft is providing a huge shot in the arm for developers to make better use of the GPU for more than just graphics acceleration. Such tasks include high-quality video playback, high performance transcoding, enabling new media scenarios, and offering extended control over media libraries.
"As an example of the real world benefits of DX Compute, you will be able to use the massive parallel capabilities of the GPU to significantly reduce the time it takes to manager your media files compared with just using the CPU alone," Nvidia added.
Image Credit: Nvidia
More like this
lien_meat
July 28, 2009 at 6:13pm
@Gailim: So by "new universal standard" I'm guessing you mean just on windows, and just by cards that support something like cuda or stream? Pretty sure DX compute is gonna have a fair amount of competition with openCL being supported by all the big players already for a while now, on EVERY platform. (meaning, every OS, and every device including cell phones and desktops alike...). Yeah, openCL doesn't get any press, but be sure, there are plenty things in the works for it(OSX snow leopard to name one).
### I'm an idiot, and I approve this message ###
SpazzAttack
July 28, 2009 at 12:51pm
Can the GPU processing be used for real-time applications? Applications like real-time machine vision and image processing will always be bound to the CPU if the memory in the video card is still a lot slower to read from than it is to write to. Anybody actually know what the story is with the memory?
Gailim
July 28, 2009 at 12:07pm
they have spent years promoting CUDA, and this new protocol is essentially the nail in the coffin for it. on the one hand it's great because GPGPU is going to have a chance at the mainstream, but on the other hand DX compute works equally well with ATI cards, whereas CUDA was nvidia's baby. I think on the whole it will help more than hurt them, (video cards will be viewed as more essential in the future) but it must hurt a little to see this new universal standard supplant thier internal one
Deanjo
July 28, 2009 at 1:15pm
Not quite, CUDA = Compute Unified Device Architecture. It doesn't matter if your using DX Compute or openCL on a nvidia card you will still be using CUDA. Your confusing 'C for CUDA' with CUDA. Niether openCL nor DX Compute spell the end to CUDA as it is required to use either of them on a nvidia card.
Connect with MaximumPC
Featured Content
We introduce Intel's latest class of tablet killers and review 4 of the first Ultra...
Where have all the memorable videogame enemies gone?
This month's issue
Feature
11 Hardware Hacks
Feature
Overclock Your CPU
How-To
Supercharge Your Smartphone
Build It
A Powerful $830 Rig
Most Commented Articles
Latest Max PC Tweets
- maximumpc: Analysts: 9-Inch Kindle Fire This Summer http://t.co/QDYApwCZ2 hours 6 min ago
- maximumpc: Micron: DRAM Prices Likely as Low as They're Going to Get http://t.co/fFUWuFOm5 hours 3 min ago
- maximumpc: Yar, Mateys! All Of Pirate Bay Made Available As A Single 90MB Zip File http://t.co/j5LHlXEZ6 hours 17 min ago
