Slashdot Mirror
NVidia Cripples PhysX "Open" API
An anonymous reader writes "In a foot-meet-bullet type move, NVidia is going to disable PhysX engine if you are using a display adapter other than one that came from their company. This despite the fact that you may have an NVidia card on your system specifically to do this type of processing. 'For a variety of reasons some development expense, some quality assurance and some business reasons Nvidia will not support GPU accelerated PhysX with Nvidia GPUs while GPU rendering is happening on non-Nvidia GPUs.' Time to say hello to Microsoft dx physics or Intel's Havok engine."
Worse than that even, this is using your strength in one industry segment (physics acceleration) to support sales of an arguably different segment (graphics acceleration).
Which is nasty and unethical to be sure, but it's not illegal unless it can be legally shown that Nvidia is a monopoly. It's amazing to me how many slashbots don't understand this distinction.
I'm pissed at ATI for dropping binary support for FGLRX for Linux kernels later than 2.6.29, and was considering getting an Nvidia GPU in my next laptop, but now it looks an awful lot like Intel is getting my $50....
I have no problem with your religion until you decide it's reason to deprive others of the truth.
It is no standard. PhysX was an API made by a company (Ageia) who wanted to cell physics acceleration cards. Their cards never sold well, but the free beer software libraries were used by a number of people (the libraries supported CPU execution as well). Then NVIDIA bought them and ported the thing to run on their GPUs. So I see this ending up like the 3Dfx Glide API for 3D graphics - some historic games used it, such as Mechwarrior, but no one uses it anymore.
physx seemed nice until they tried to close source it. Does Nvidia have anything left this round? Bad Yields, physx being stupid and abusive when disabled (it only uses 1 cpu core when on AMD for example instead of even all threads). Not to mention their crippling of batman as well.
So what's left for Nvidia? I don't see a whole lot.
No. The framework would only run on their GPUs. However, you could have one of their cards in the system to do purely physics calculations, and then use a competitor's card to do the actual display and 3d rendering. They've now disabled this, so if your monitors are connected to, say, and ATI card, you can no longer use the Nvidia card in your system for physics processing.
Before you discount this as an unlikely scenario, consider motherboards with onboard NVidia chipsets. These are usually underpowered for full time duty, but are perfectly suited to being used for physics calculations while a more powerful ATI card in the PCI-E slot does the graphics rendering. This is actually a fairly likely setup these days, and NVidia has just said they're going to block it.
Personally, I agree with others who have pointed out this must be an anti-trust issue. Intel and Microsoft have both been fined heavily recently for doing exactly this kind of anti-competitive behaviour.
-- sudo.ca
OpenGL3 is the first time that companies are breaking away from windows.
It seems like OpenAL was the first. Creative have been visibly pushing it now that Vista's forced-software-only sound API has made their sound cards pointless.
http://www.bulletphysics.com/
I don't have any affiliation with the project other than I've used it in my homegrown game engine that has never left my hard drive. It is however rather easy to use. When I was looking for a physics engine, Bullet turned out to be the best license, code base, and documentation set out there for no cost.
Persistent Volume manager for Kubernetes - https://github.com/dwimsey/openshift-pvmanager
From the NVIDIA PhysX FAQ:
Can I use an NVIDIA GPU as a PhysX processor and a non-NVIDIA GPU for regular display graphics?
No. There are multiple technical connections between PhysX processing and graphics that require tight collaboration between the two technologies. To deliver a good experience for users, NVIDIA PhysX technology has been fully verified and enabled using only NVIDIA GPUs for graphics.
Read the f... Phoronix article ;-)
Yes, it seems the binary is really crappy in this case and the OSS driver at least passable. Although I'm not familiar with those benchmarks and how they measure up against similar software on Windows.
3D still seems to lag behind, otherwise we could officially forget the Catalyst driver and use the OSS driver exclusively for Linux. But I think we'll get there.
C - the footgun of programming languages
An OpenCL implementation of Bullet physics is coming. It's Open Source and is already being used in commercial games -- once it gets GPU acceleration there will probably be little demand for PhysX.
They didn't drop support for "CRT's". They decided that their stereographics driver would only work in the following configurations:
- anaglyph glasses with a "whatever" monitor (horrible color distortion and headache-inducing ghosting ensues).
- *THEIR* shutter glasses, with *THEIR* overpriced "partnered" LCD monitors.
Now, what is the difference (tech-wise) between their shutter glasses and mine? Only the fact that theirs send a specific "yes I'm nvidia" signal back to the card. What is the capability difference between their overpriced "partnered" LCD monitors and my 120Hz-capable CRT? Two things: Jack and Shit.
This is not about "dropping support for outdated technology." Prior to what they pulled, I could plug in an industry-standard shutter glasses set made by any of a number of manufacturers, combine them with any monitor capable of 120-Hz refresh (whether CRT, LCD, certain televisions, or even a few projector models), and enjoy stereoscopic gaming. After their "update" to the drivers and subsequent "update" to the stereoscopic drivers, the Nvidia cards would only recognize *THEIR* proprietary glasses (which again, hardware-wise are no different than the old type save for sending a "hi I'm from nvidia" signal to the card) and would only interoperate with a precious few "specially chosen" 120Hz LCD's.
This had nothing to do with "dropping support" for "obsolete equipment" (which wasn't in any way, shape, or form) and everything to do with trying to milk people for $500+ on a new rig by forcibly crippling industry-standard hardware.
Its anti-consumer, but that doesn't trigger an anti-trust charge, they don't have a monopoly.
The perpetrator doesn't have to have a monopoly for tying to be illegal - in U.S., for example, you only need "sufficient market power" to affect "not insubstantial amount of interstate commerce in the tied product market". I dare say that NVidia has pretty damn substantial market power in GPU niche, and it is quite likely to affect sales of all other GPUs in a significant way. In the end, of course, it's up to the courts to decide, if it comes to that, but the allegation is not without merit.