• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

Behold the power of SHADER MODEL 3.0!

Soldato
Joined
30 Jul 2005
Posts
19,897
Location
Midlands
http://www.firingsquad.com/news/newsarticle.asp?searchid=9435

SANTA CLARA, CA—MARCH 20, 2006—NVIDIA Corporation (Nasdaq: NVDA), the worldwide leader in programmable graphics processor technologies, and Havok, the game industry’s leading supplier of cross-platform middleware, will be demonstrating a physics effects solution that runs completely on a graphics processing unit (GPU) —an industry first—at this year’s Game Developer Conference (GDC) in San Jose, California (March 21st through 24th). The result of an ongoing engineering collaboration between Havok and NVIDIA, this new software product from Havok—called Havok FX™—enables the simulation of dramatically-detailed physical phenomena in PC games, when powered by GPUs such as NVIDIA GeForce®7 or 6 Series GPUs and further amplified with NVIDIA SLI multi-GPU technology. The Havok FX product is currently in early release to select developers and is expected to be available this summer.

Through Havok FX, GPUs can simulate the interactions of thousands of colliding rigid bodies, a fundamental technique of physics computation seen in today’s latest games. It’s now possible to compute the components of friction, collisions, mavity, mass, and velocity that form the basis of rigid body physics. Havok FX is designed for GPUs supporting Shader Model 3.0, including the NVIDIA GeForce 6 and 7 Series GPUs.

Utilizing Havok FX and NVIDIA graphics technology, game developers can now implement sophisticated physical phenomenon such as debris, smoke, and fluids that add immense detail and believability to game environments. Game designers can include advanced physics effects without burdening the CPU and slowing game-play, since the effects are simulated and rendered on the GPU.

Tyler Thompson, Technical Director of video game developer Flagship Studios states, “With Havok FX, we can explore new types of visual effects that add realism into Hellgate: London. Given the widespread installed base of GPUs and the incredible performance of the new NVIDIA GeForce 7900 GTX and GT GPUs, Havok FX was a natural choice.”

“We are very excited about the quality and speed we are seeing on the NVIDIA GPUs. We’ve believed for some time that GPU technology had the potential to simulate physical effects and our collaboration with NVIDIA has proven that,” says Jeff Yates, VP of Product Management at Havok. “The large installed base of Shader Model 3 class GPUs and momentum by NVIDIA in the market make Havok FX an attractive solution for game developers looking for hardware-accelerated physics.”

“Moving physics processing to the GPU is a natural progression enabled by the high programmability in today’s GPUs,” said David Kirk, chief scientist at NVIDIA. “By combining expertise with Havok, we have produced a fantastic solution for game developers that will lead to more compelling game-play and more realistic gaming experiences.”

The award-winning NVIDIA GeForce 7 and GeForce 6 graphics architectures deliver advanced technologies including full support for Microsoft® DirectX® 9.0 Shader Model 3.0, enabling advanced shading programs for more realistic effects. By using the advanced programmable resources available in current GPUs, developers can harness the massive parallel computation capability of NVIDIA GPUs. The performance available to developers is further enhanced through NVIDIA SLI technology. This revolutionary platform innovation allows users to intelligently scale graphics performance by combining multiple NVIDIA graphics solutions in a single system with an NVIDIA nForce® SLI MCP.




looks like a sm3.0 gfx card could end up rivalling the ageia physx processor.
the new havok FX engine makes good use of SM3.0

glad i bought a 6800 card now. :D
 
I doubt it, plus if you offload more and more onto the GPU, it will impact your frame rates. At 1920x1200+, some games already stress out 7900 SLI based systems, so for the power user a dedicated card will be the way forward.
 
i for one can;t see how this is going to work? how can shader model 3.0 impact physics? i thought all it does is speed up shading such as shadows etc?
 
I asked someone I know about the new PhysX processors, whether they'd be a success. He reckons they won't come to anything in the end, since GPUs have the capability to do all this stuff anyway.

I guess this is what he was talking about.

Hav
 
GPUs seem to be very good at raw calculations.... but I dont see a GPU being able to render a scene to the screen whilst also working out what should go in what direction etc.
No doubt it will help slightly... but nothing beats a solution that is designed for that very purpose. As a previous poster has pointed out -> i also struggle to see the link between phsyx and sm3.0.... wonder if it will be driver locked?
 
BubbySoup said:
I doubt it, plus if you offload more and more onto the GPU, it will impact your frame rates. At 1920x1200+, some games already stress out 7900 SLI based systems, so for the power user a dedicated card will be the way forward.


I agree this won't be a good solution. When an explosion takes place in a game both the graphics and physics caluclations will be fighting for GPU time.
 
Well, skimming through the [H] article on it, in SLI based systems, one card will be used as a GPU, the other mainly as a PPU. I just don't see how this is going to work much more than a gimmick. Effectively, you have paid £350+ for a PPU - granted you don't have to use it like that, but then why bother implementing this in the first place?

I for one would rather higher image quality and smoother frame rates than more rocks bouncing about. If I want my bouncy rocks, I'll go for a plug-in card.
 
Looking at it, it doesn't compare to a PhysX card. It just does Physics for graphic effects (so eye candy), the PhysX card does real physics for game effects (interaction between game objects, etc.). This is mentioned in an article on the inq.
 
This has been talked about for quite a while, and I find it very interesting - maybe this will bring dual core gpu's to the forefront? not in the case of one card for an SLI setup as is available now in several flavours, but more towards
one totally combined gpu / ppu - after all its been done in conjunction with nvidia themselves and ppu's (at first anyway) will be targeted at the high end and this audience as stated above wont want to lose framerates etc which has to have some kind of impact.

I just really hope that it doesnt become propreitary, and this only works on nv's cards - its a great avenue of opportunity but if its software they should develope it down strict guidelines that work across all sm3.0 cards or whatever (even though I can see this might cause nV to lose out)

to my mind an add-in card will always be beneficial all round rather than overloading an already stressed componant.
 
Seems like a good idea. For those with high end sli setups they are usually cpu limited.So taking the limit off the cpu and transfering the work to the gpu seems like a good idea as gpu's work better with vectors.(can do one thing in the time it takes a cpu do 32 things or something like that).

Im sure nvidia can work on it and work on load balancing,when a scene has high visual needs the two cards work at 100% on graphics but on a more vector intense area one card work at 100% and the other at 50% and the other 50% on physics.

We'll see how it goes anyway. :cool:

It wont work well with single cards cause you will be taking one gpu's power and wasting it,no doubt with a FPS hit aswell.
 
Ok, lets get things straight.

ALL physics calculations will *eventually* be done on the GPU.

At the moment a GPU based solution might not be the most efficient or powerful solution, but will eventually be the norm.

A PPU card will be the most powerful solution for the next couple of years.
 
MetalStorm said:
Ok, lets get things straight.

ALL physics calculations will *eventually* be done on the GPU.

At the moment a GPU based solution might not be the most efficient or powerful solution, but will eventually be the norm.

A PPU card will be the most powerful solution for the next couple of years.
Sorry, but that's rubbish. If PPU cards become "the most powerful solution for the next couple of years" then that is how things will stay. Why take a step backwards in terms of performance a few years down the line? Using either a GPU or another core from a dual/quad core CPU will never be as efficient as a piece of silicon 100% designed & dedicated to the task.

The closest thing I can see happening to your statement, IF they take off, is the inclusion of the PPU onto the graphics board, possibly even as a socket on the MB.
 
Ati were messing around with this months back as well. Multi function graphics\physics cards sooner or later/ =/
 
Eventually just like CPUs, GPUS will have dual cores, this is very early technology that is possible right now using SLI. They can get the technology working under SLI now, so once we do start seeing dual/quad GPUS, the technology will be there to use them.

As at the end of the day, GPU techology is already ahead of games technology, it could spare a few 1000 gpu cycles if you turned down the AA of AF, giving you better physics.

Maybe in the future, in game settings, you could turn AA down to x2 in favour of more particles ingame, or keep it at x4 etc and have less particles and effects. Essentially balancing yoru dual cpu/gpus systems for your own preferance.

And this isnt a Nvidia only thing, its just that NV have been giving Havok loads of dosh for research, so Havok will give them BIG HELLO in there press releases. Plus SLI has been available to Havok to play with longer than Crossfire has.
 
Back
Top Bottom