superkdogg
Gawd
- Joined
- Oct 14, 2004
- Messages
- 743
Here's the end-game of this situation that I'm surprised nobody is talking about yet (I'm an experienced user, with good everyday tech knowledge, but I'm admittedly out of my element on theory so maybe this is completely junk and I don't know it-if so tell me why) :
I think that the result of the PhysX movement is going to be Havoc and Nv/ATI working on hardware-acceleration of physics on GPU's (or their sexay-named new GPU-like cards). With the programmable nature of today's GPU's and the power available in said GPU's, I'm thinking that the current HavokFX and whatever ATI proposes are simply a stepping stone to full-on hardware acceleration on GPU's. There's also the alternative of ATI and NV offerring their own physics cards that use a new Havok API that allows for a different version of physics (one that does not exist yet) but still includes hardware acceleration. In the process, one or both of the big teams buys intellectual property rights from Ag and effectively ends the company's hardware efforts, but continues development of the technology and integrates it into future implementations. It's win-win-win for everybody as ATI and Nv get new market, Ag's venture capital pays off in the sale, consumers get hardware accelerated physics, dev's get another toy to play with and it's uniform across the board rather than two different standards, and everybody lives happily ever after.
I think that this is a logical outcome, but again, I'm no guru-what do you all think?
I think that the result of the PhysX movement is going to be Havoc and Nv/ATI working on hardware-acceleration of physics on GPU's (or their sexay-named new GPU-like cards). With the programmable nature of today's GPU's and the power available in said GPU's, I'm thinking that the current HavokFX and whatever ATI proposes are simply a stepping stone to full-on hardware acceleration on GPU's. There's also the alternative of ATI and NV offerring their own physics cards that use a new Havok API that allows for a different version of physics (one that does not exist yet) but still includes hardware acceleration. In the process, one or both of the big teams buys intellectual property rights from Ag and effectively ends the company's hardware efforts, but continues development of the technology and integrates it into future implementations. It's win-win-win for everybody as ATI and Nv get new market, Ag's venture capital pays off in the sale, consumers get hardware accelerated physics, dev's get another toy to play with and it's uniform across the board rather than two different standards, and everybody lives happily ever after.
I think that this is a logical outcome, but again, I'm no guru-what do you all think?