• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

Nvidia implementation of CUDA almost complete

I don't like it, I don't think I have any GPU cycles on even my 8800GTS as it is. Need everything I can for pixels and graphics...

Now if I could throw a $26 8400GS card in my system and use it as a PPU instead of a second GPU... and get huge physics performance gains... then I'd be all for it.
 
I don't like it, I don't think I have any GPU cycles on even my 8800GTS as it is. Need everything I can for pixels and graphics...

Now if I could throw a $26 8400GS card in my system and use it as a PPU instead of a second GPU... and get huge physics performance gains... then I'd be all for it.

Agreed.
 
Reading Theo on THG is just as bad as it at was at theinq. :barf:

bit-tech had coverage here: http://www.bit-tech.net/hardware/2008/04/14/nvidia_analyst_day_-_biting_back_at_intel/5
Hegde said that PhysX was the most pervasive and advanced physics API on the planet, claiming that it's featured in at least 140 titles across all platforms. He also revealed that the PhysX SDK had been completely ported over to CUDA, therefore enabling GPU accelerated physics, in just one month and he claimed that there has been an exponential increase in developer adoption because of >50m installed base of CUDA-ready GPUs.
Hopefully CUDA PhysX gets rounds of performance improvements through updates, instead of being used to push higher end hardware, and has flexibility in choosing from secondary cards installed.
 
Think of it like this. We all use anti aliasing. But there was a time when it took a huge performance hit. But today we can do AA with little performance hit or so small it dont matter. Graphics cards are so fast we just take AA for granted.

This is how I see PhysX. It'll take a performance hit but as cards get faster the performance hit wont matter anyway cause your games will be running at more than playable frame rates anyway. So in the beggining people will complain of the performance hit till Nvidia cards get so fast that the peformance hit is so small we'll take it for granted like AA.
 
I don't like it, I don't think I have any GPU cycles on even my 8800GTS as it is. Need everything I can for pixels and graphics...

Now if I could throw a $26 8400GS card in my system and use it as a PPU instead of a second GPU... and get huge physics performance gains... then I'd be all for it.

agreed.

an 8600GT would be fantastic in a Sugo03 case + Gigabyte G33 board as a CUDA Physx card.
 
This made me giggle:

While Intel's Nehalem demo had 50,000-60,000 particles and ran at 15-20 fps (without a GPU), the particle demo on a GeForce 9800 card resulted in 300 fps. If the very likely event that Nvidia's next-gen parts (G100: GT100/200) will double their shader units, this number could top 600 fps, meaning that Nehalem at 2.53 GHz is lagging 20-40x behind 2006/2007/2008 high-end GPU hardware. However, you can't ignore the fact that Nehalem in fact can run physics.

Did the 9800 run without a CPU?
Or the PR cover that part up?
*smiles smurky*
 
Did the 9800 run without a CPU?
Or the PR cover that part up?
*smiles smurky*

Yeah you're right he should have said 300fps minus the 20fps provided by the cpu.
280fps is so much closer to 20fps than 300fps, damn biased journalists!
:D
 
Yeah you're right he should have said 300fps minus the 20fps provided by the cpu.
280fps is so much closer to 20fps than 300fps, damn biased journalists!
:D

Not my point, try running without the CPU.

0 FPS < 20 FPS...
 
Not my point, try running without the CPU.

0 FPS < 20 FPS...

Try running w/o the monitor, try w/o RAM or try w/o a computer at all, what's the point, GPU is not competing for CPU, he's just pointing the fact CPU suxor at parallel processing in the current state.
 
Try running w/o the monitor, try w/o RAM or try w/o a computer at all, what's the point, GPU is not competing for CPU, he's just pointing the fact CPU suxor at parallel processing in the current state.

No need to tell me that, since I own a PPU, I know how much the current CPU's sucks at Graphics/Physics.

But based on what I have picked up so far NVIDIA is in a thight spot.
AMD has a platform (CPU, Chipset, GPU)
Intel has has a platform (CPU, Chipset, GPU(and Larrabee comming))
NVIDIA has 2/3 platform (Chipset/GPU)

Unless NVIDIA get a DECENT CPU, they will be banned to a sideline with Intel as the big boy and AMD trying to push NVIDA out.

Looking at the big picture, Intel is on a mission...AMD kicked them in the balls, but now they are fully awake and as I can see it they are the only ones with enough money, fabs, R&D ect to do the "full monthy" in a timely manner.

When ever I hear NVIDIA bark, i sense fear in their barking...
 
Methink Intel wants more chipset coverage, that is where nvidia is working very hard, the nForce series have been very popular so far.

As for onboard gpu...There is always the Hybrid thingy under Vista which nvidia is pushing, I don't see Intel competiting for gaming market, that segment is shrinking anyway in favor of console these days.Not gonna start an argument which one is better, but fact remains Intel wants chipset and onboard supremacy, they're not aiming at high end 3D onboard.

Look at it this way, how the hell can you redo the whole computer architecture without breaking the OS and API, just take a look at how hard it was for a small company to even "add" a new API to the mix like Ageia did with the PPU, and thats not remotely close to rebuilding a computer architecture.

Computer architecture branching is something we saw in the 80's because there was no unified OS, yeah i hate to say it, but Winblows kinda ruined it, M$ owns the platform market therefore it indirectly dictate how it's gonna happen.

Let's say tomorrow morning a company comes into play with this awesome computer without any legacy background, built ground up with new technologies, with theorical breakthrough performance, but the damn thing doesn't run on windows and the unit price is 3 times the price of your average PC because of steep developement cost although the performance is twice your current high end system...

It just gonna fail or find a niche market... SGI, SUN, Amiga... you get the picture.
 
Methink Intel wants more chipset coverage, that is where nvidia is working very hard, the nForce series have been very popular so far.

Nehalem does make the chipset less important, and besides if yiou want the best, most table chipset for your intel CPU, you go with an chipset made by Intel..ask around...hell ask Dan?

As for onboard gpu...There is always the Hybrid thingy under Vista which nvidia is pushing, I don't see Intel competiting for gaming market, that segment is shrinking anyway in favor of console these days.Not gonna start an argument which one is better, but fact remains Intel wants chipset and onboard supremacy, they're not aiming at high end 3D onboard.

Dumping the GPU, and going multi core will make PC gaming more widespread.
Even Tim Sweeny see the GPU as a temporay solution to something tha is really a CPU problem.

Look at it this way, how the hell can you redo the whole computer architecture without breaking the OS and API, just take a look at how hard it was for a small company to even "add" a new API to the mix like Ageia did with the PPU, and thats not remotely close to rebuilding a computer architecture.

It's runnig today...slow, but still running...no need to paint a false picture..all we need now is soemthing like the Teracale CPU..multiple cores galore...

Computer architecture branching is something we saw in the 80's because there was no unified OS, yeah i hate to say it, but Winblows kinda ruined it, M$ owns the platform market therefore it indirectly dictate how it's gonna happen.

Red Herring...you can run DX on the CPU today...no need for a change..why do you think NVIDIA is whining?
(Hint: they don't have a CPU)

Let's say tomorrow morning a company comes into play with this awesome computer without any legacy background, built ground up with new technologies, with theorical breakthrough performance, but the damn thing doesn't run on windows and the unit price is 3 times the price of your average PC because of steep developement cost although the performance is twice your current high end system...

It just gonna fail or find a niche market... SGI, SUN, Amiga... you get the picture.[/quote]
 
Ok... so next year 18 April 2009 you post back here with a computer that has nothing but a CPU, a motherboard w/o chipset-souncard-gpu.

As for the rest of your reply, please take no offense but I have no clue what you're trying to say.So I don't see this discussion going any further in a constructive way.Ciao.
 
Ok... so next year 18 April 2009 you post back here with a computer that has nothing but a CPU, a motherboard w/o chipset-souncard-gpu.

As for the rest of your reply, please take no offense but I have no clue what you're trying to say.So I don't see this discussion going any further in a constructive way.Ciao.

Subscribed so I can see this in a year.
 
Ok... so next year 18 April 2009 you post back here with a computer that has nothing but a CPU, a motherboard w/o chipset-souncard-gpu.

As for the rest of your reply, please take no offense but I have no clue what you're trying to say.So I don't see this discussion going any further in a constructive way.Ciao.

have fun upgrading that combo..
 
Ok... so next year 18 April 2009 you post back here with a computer that has nothing but a CPU, a motherboard w/o chipset-souncard-gpu.

Didn't think I needed to adress this fallcy, but it seems i have to.
Inventing goals, shifting the goal post and setting a deadliine of the the thin blue air that has nothing to do with what I said is a fallacy of magnetiudes, please refrain from doing that or assuming that I will humor your fallcy.
 
Back
Top