Nvidia Responds to AMD's Claim of PhysX Failure
AMD accuses Nvidia of disabling multi-core CPU support in PhysX API -- Nvidia denies all such claims.
With PhysX being an Nvidia property, there are obvious reasons why AMD wouldn't be first in line to sing the praises of that specific proprietary physics technology.
Earlier this month, AMD worldwide developer relations manager Richard Huddy said in an interview with Bit-tech that Nvidia is squandering away CPU resources.
"The other thing is that all these CPU cores we have are underutilised and I'm going to take another pop at Nvidia here. When they bought Ageia, they had a fairly respectable multicore implementation of PhysX. If you look at it now it basically runs predominantly on one, or at most, two cores," said Huddy. "It's the same thing as Intel's old compiler tricks that it used to do; Nvidia simply takes out all the multicore optimisations in PhysX. In fact, if coded well, the CPU can tackle most of the physics situations presented to it."
We asked Nvidia for its response to the allegations made by AMD, and Nadeem Mohammad, PhysX director of product management, stepped up to the mic in hopes of setting the record straight:
I have been a member of the PhysX team, first with AEGIA, and then with Nvidia, and I can honestly say that since the merger with Nvidia there have been no changes to the SDK code which purposely reduces the software performance of PhysX or its use of CPU multi-cores.
Our PhysX SDK API is designed such that thread control is done explicitly by the application developer, not by the SDK functions themselves. One of the best examples is 3DMarkVantage which can use 12 threads while running in software-only PhysX. This can easily be tested by anyone with a multi-core CPU system and a PhysX-capable GeForce GPU. This level of multi-core support and programming methodology has not changed since day one. And to anticipate another ridiculous claim, it would be nonsense to say we “tuned” PhysX multi-core support for this case.
PhysX is a cross platform solution. Our SDKs and tools are available for the Wii, PS3, Xbox 360, the PC and even the iPhone through one of our partners. We continue to invest substantial resources into improving PhysX support on ALL platforms--not just for those supporting GPU acceleration.
As is par for the course, this is yet another completely unsubstantiated accusation made by an employee of one of our competitors. I am writing here to address it directly and call it for what it is, completely false. Nvidia PhysX fully supports multi-core CPUs and multithreaded applications, period. Our developer tools allow developers to design their use of PhysX in PC games to take full advantage of multi-core CPUs and to fully use the multithreaded capabilities.
- TSMC Says 40nm Issues are Resolved
- Windows 7 SP1 Screenshots Leaked
- xpPhone Will Cost $400-650, Run Windows 7 Too
- 17-year-old Windows Flaw Affects All Since NT
- Macs Finally Get Windows 7 Support in Boot Camp
- HP's "Wall of Touch" That You Don't Need to Touch
- Man Sells Steam Account for $1,000 on eBay
- Nvidia CEO: The PC Has Lost its Magic
- We've Updated Our Graphics Card Charts!
- Asus Launching Ion 2 Eee PC in April
- Build Your Own: Introducing Tom's Hardware's BestConfigs!
- Nvidia Driver Bug Stops Overclocking; Fix Soon
- System Specs Revealed for Aliens vs Predator
- Drool: Maingear's New OC'd Core i7 Gaming Rigs
- Clinton Wants Chinese Explanation of Hacking
- Firefox 3.6 is Officially Ready for Download
- Galaxy Launches 2-Year-old Graphics Card. Why?
- AMD Posts First Profit in Three Years
Who to believe?
I'm going with AMD on this one.
"This can easily be tested by anyone with a multi-core CPU system and a PhysX-capable GeForce GPU."
So to prove they are not messing with things we need a GeForce GPU first? Is he admitting guilt or just stupid?
"This can easily be tested by anyone with a multi-core CPU system and a PhysX-capable GeForce GPU."So to prove they are not messing with things we need a GeForce GPU first? Is he admitting guilt or just stupid?
If you don't have a GeForce GPU, what do you care anyway?? So, who is the stupid one here..........
If you don't have a GeForce GPU, what do you care anyway?? So, who is the stupid one here..........
Maybe you shouldn't comment if you actually don't understand the topic?
Summary of AMD's statement: PhysX was originally capable of running on anything useful and not just Nvidia GPU's. Now Nvidia has favored their own GPU's so much that it doesn't run efficiently on anything (CPU/GPU) else even though it could.
PhysX can run on Intel/AMD CPU's or ATI GPU's. Nvidia just doesn't want it to be good enough cause it would stop them from making you believe their cards are the only viable way. You obviously fell for Nvidia's brainwashing. Calling you stupid in return may be a bit too much, gullible though I'd dare to throw out.
Yeah, glad someone else commented on that blakar, I was thinking the same thing!
Thirded.
Yeah, glad someone else commented on that blakar, I was thinking the same thing!
What do you care meaning what do you care how well it runs on Nvidia compared to your ATI? Are you happy with the PhysX-performance on your ATI? I bet you are going to answer "NO" now "because you know that it could run faster because Nvidia is not supporting other GPUs in the same way". You could write to the CEO of AMD and ask to develop a similar system if you are all so sure that Nvidia is "cheating"
And for the rest, I'm sorry to disappoint you, but I have a Quadro in my Workstation.......
OK, somebody clarify something for me. The NVIDIA guy said:
"One of the best examples is 3DMarkVantage which can use 12 threads while running in software-only PhysX. This can easily be tested by anyone with a multi-core CPU system and a PhysX-capable GeForce GPU."
Does that mean you still have to have a GeForce GPU present in your system to utilise PhysX AT ALL, even if you programme it to run software-only on the CPU?
If that's the case then yes the AMD guy is correct - the programming model of PhysX restricts its operation UNLESS an NVIDIA GPU is present. But then PhysX IS NVIDIA tech so they are within their rights to do such a thing, however pathetic it may be.
But does poor PhysX underutilisation condone AMD blocking NVIDIA GPUs when a Radeon GPU is present?
I wasn't aware they were, more like the other way around.
I wasn't aware they were, more like the other way around.
I thought there was a hooha about ATI drivers blocking NVIDIA GPUs for PhysX if Radeons were the primary card? Or do I have that totally confused?
Been a long week - brain's turned to jam!
I wouldn't know about that one, but I did hear that nVidia were blocking ATi GPUs being present in the same system in some sort of fashion, presumably so they couldn't be made to work together with Lucid's Hydra solution. I'm not sure if it was ever proven, though (it's early morning and I can't be bothered to go search for it... lol).
And I know about the brain thing... can't wait for this week to end, to be honest!
I going for AMD on this one!
Purely because Dirt 2 runs s**t when physx is installed with my 5870!
Yes Silverblue thanks for reminding me. I got the culprit the wrong way round - NVIDIA drivers blocked ATI cards. The end result is the same though, which is what I was getting at: you can't run a GeForce card as a dedicated PhysX processor if you're using ATI GPU cards.
I did think for a minute I had it the wrong way round.
Apparently nVidia weren't blocking Hydra, but whether we believe that or not is entirely up to us. With nVidia's chipset business in tatters and their SLi licence fee being (supposedly) the highest (ATi < Lucid < nVidia) PLUS Hydra's ability to mix two very different cards in terms of architecture and capability, nVidia would probably feel justified in trying to stifle this threat, especially considering they're very much behind with GT100. They're facing cheaper competition from ATi, along with Hydra's ability to potentially stop you from needing to buy two cards for a dual GPU setup as you can use your existing card and thus deprive nVidia of even more money.
I'd have preferred if nVidia had licenced PhysX out as a tool to implement in games regardless of the card they were using instead of bolting it onto their GPUs. If CPUs are easily capable of running it (especially now tri- and quad-core offerings are widespread) then it'd be far more accessible, AND it possibly would've made nVidia far more money.