Originally posted by: SunnyD
Wow, you really just don't get it do you.
LOL. I actually refuse to believe you're this ignorant, but you're convincing me pretty quickly.
#1 - It's a tick in the PhysX control panel. You know, the thing that gets installed with the stand-alone PhysX package? Yeah - it's a restriction imposed by Nvidia, if you'd bother to get your head out of Nvidia's ass and actually TEST the damn thing instead of spewing the marketing propaganda you're encouraged to spew.
And what do you think that PhysX standalone package is? Perhaps a hmmm --->
CUDA DRIVER <---- that Nvidia provided for their hardware to run on their proprietary GPGPU API when there were no viable alternatives? What hardware would PhysX run on if there was no Nvidia part present? Oh right, the CPU...in software. All the necessary runtimes are included in those games, no "stand-alone PhysX package" needed, just as there is none needed for the myriad other platforms like all 3 major consoles that the PhysX middleware supports without any driver from Nvidia.....
There is no testing needed, I'm well aware of what the ---->
CUDA PHYSX DRIVER <---- does and why its necessary for PhysX to run on Nvidia hardware, thanks. And none of that is marketing propaganda, the only encouragement was a little bit of research in order to not sound like a complete ignoramus on the topic.
#2 - As I said, the PhysX middleware is currently tied to Nvidia hardware, plain and simple. Stop trying to twist all this around to an ATI deficiency, it's not. If Nvidia were to port the middleware (read: the ACTUAL PHYSX SOFTWARE) to OpenCL, Nvidia wouldn't care what other vendors do as long as the vendors adhere to the OpenCL standards. If ATI were to mess it up, which arguably they have a lot invested in getting OpenCL right being a primary member of the working group, then Nvidia could laugh in ATI's face.
No, the PhysX middleware is not tied to Nvidia hardware, or you wouldn't be able to run any
of these games without an Nvidia DX10 or better GPU. No one is trying to twist this into an ATI deficiency, your claim that Nvidia is somehow blocking PhysX support with a driver lock is blatently false. Nvidia has stated numerous times they would be willing to work with ATI to support PhysX on their parts but were rebuffed each time.
ATI's previous statements indicated they did not want to support proprietary standards (read CUDA, the proprietary NV API currently used for PhysX). Nvidia's statement that they plan to port PhysX to OpenCL removes that last layer of FUD that shelters ATI and its apologists, at which point ATI and its fans like you can't claim lack of support due to "proprietary standards" or whatever other BS, as PhysX will be running on the open standard, OpenCL.
At that point, ATI needs to write an OpenCL driver for their hardware and support PhysX, or they need to explain to their customers why
they do not want to support PhysX. Balls in their court with no more excuses, simple as that.
As for messing up and laughing etc. blah blah, you're clearly wrong there as well. Both Dave Baumann on these very forums and Nadeem Mohammed
at Bit-Tech expressed their concerns about implementation, brand image, and performance with regard to PhysX and hardware support.
Please, stop trying to pretend you know anymore than your marketing agenda purports to know and just shut up.
Again, I'm absolutely shocked you would comment so ignorantly on a topic you clearly know nothing about. The only agenda here is to combat misinformation. Once Nvidia ports to OpenCL we'll have to revisit this thread and see just who should've shut up, and when.
Facebook
Twitter