Jump to content

PhysX Shenanigans


sdy284

Recommended Posts

Sorry, I worded this wrong.

 

Yes, the software is free, but unless you're Microsoft there isn't much money to be made on software alone anyway. 24/7, on site, technical support has a price though and it's not cheap!

 

Anyway, open source graphics libraries on the PC haven't enjoyed much success so I don't see why physics would be an exception, but physics API's are going anywhere at all right now so who knows...

It's true, I've seen some linux distros make a ton of money just from selling books for their distro.

Share this post


Link to post
Share on other sites

  • Replies 47
  • Created
  • Last Reply

Top Posters In This Topic

Top Posters In This Topic

Yes, the software is free, but unless you're Microsoft there isn't much money to be made on software alone anyway. 24/7, on site, technical support has a price though and it's not cheap!

this sounds a lot more like the model to which i am accustomed, but i can't see how it could be applied effectively in this context.

 

physics API's are going anywhere at all right now so who knows...

LOL! next to nobody creates their own physics engines from scratch these days!

 

hardware accelerated physics might not be huge right now, but they absolutely will be at some point, so getting a foot in the door at this stage is incredibly important. with the GPGPU power fermi is likely to be touting this may well be the generation we see some really stunning implementations hitting the market, and with the next generation of consoles not too far away, it may be the feature that puts an nvidia GPU into the XBOX720 and PS4.

Share this post


Link to post
Share on other sites

However you look at it I think we can all agree that an open standard would benefit everyone and hurt no one.

 

The problem with open source is that snyone can drop the ball and quit on the project halfway through it and have development fizzle to a halt.

 

 

:blink: Open source / open standard benefits everyone and breeds competition FROM everyone...

 

Not really see above!

 

Exactly. I know I wouldn't accept a deal where the standard is owned by my main (and sole) competitor. NVIDIA could optimize all they want, while ATI would get the binaries as is and have to hope it works with their software and hardware. In the end, it would still work like crap on ATI...

 

One more thing, we keep hearing PhysX this, PhysX that and how much potential it has... it's been out for how many years now and how many titles actually implement it so actually enhances the game? Yeah... It's obviously not getting anywhere

 

 

If you give it away for free then why can ATI not do the work to make it work for them and then you have something both sides can use and benefit from. problem is that ATI/AMD cant or wont have the resources so they want it open source so they dont foot the development budget. Foolish to put something that important out of your house and into someone else hands but if you dont have the money..........

 

Tried the effects in Dark Void?

Share this post


Link to post
Share on other sites

I don't blame NVIDIA. I mean, it puts them in a position to rake in the money at the expense of the customer (who is forced to cough up money for an NVIDIA card).

 

They're a corporation. I don't really think they care about the little guy.

Share this post


Link to post
Share on other sites

If you give it away for free then why can ATI not do the work to make it work for them and then you have something both sides can use and benefit from. problem is that ATI/AMD cant or wont have the resources so they want it open source so they dont foot the development budget. Foolish to put something that important out of your house and into someone else hands but if you dont have the money..........

 

Tried the effects in Dark Void?

 

If I understand it from the ATi/AMD point of view it is to say that they were committed to havok from the get go even before NVIDIA had CUDA or PhysX. Mainly I see it as them holding onto and doing what they said they were going to do. Say ATi did pick it up. Not only would that be beneficial to NVIDIAs IP but it would probably destroy any idea of ever implementing havok physics on their GPUS.

 

Bottom line why should ATi drop the standard they intended to support and start backing their competitors and yes by picking up physx they would be essentially dropping havok. I have no problem with ATi/AMD's decision here its nice to see a company commit its self.

 

I think the reason why they went open source to begin with isn't necessarily to drop costs and to not hire software developers. Probably the best reason why its opensource is it allows game developers to get at code rather than binaries.

Edited by Compxpert

Share this post


Link to post
Share on other sites

LOL! next to nobody creates their own physics engines from scratch these days!

I meant none of the existing physics API's really stand out in their adoption rate. (PhysX != physics)

 

Yeah, there are a few good PhysX titles and it's pretty cool what it allows, but I don't see PhysX really taking off with only one vendor backing it.

 

Of course I'd like a vendor agnostic physics API to gain some real traction and I don't care which, just come up with something ATI + NVIDIA! Until then... we'll continue to see only a few titles here and there with good effects.

Share this post


Link to post
Share on other sites

If I understand it from the ATi/AMD point of view it is to say that they were committed to havok from the get go even before NVIDIA had CUDA or PhysX. Mainly I see it as them holding onto and doing what they said they were going to do. Say ATi did pick it up. Not only would that be beneficial to NVIDIAs IP but it would probably destroy any idea of ever implementing havok physics on their GPUS.

 

Bottom line why should ATi drop the standard they intended to support and start backing their competitors and yes by picking up physx they would be essentially dropping havok. I have no problem with ATi/AMD's decision here its nice to see a company commit its self.

 

I think the reason why they went open source to begin with isn't necessarily to drop costs and to not hire software developers. Probably the best reason why its opensource is it allows game developers to get at code rather than binaries.

 

The funny thing is ATI talked about Havok long before PhysX and they said they were going to change gaming. Well they sat around and did nothing and Nvidia came out of nowhere with PhysX and stole the show, now everyone is bitching and moaning about it. Suck it up; ATI dropped the ball they had their chance and they blew it. Oh well to bad, time to move on simple as that.

 

From a personal stand point it would be nice to see it work for both companies but I live in reality and I just don't see that happening anytime soon :( unless the whole idea of PhysX and Havok is changed.

Share this post


Link to post
Share on other sites

I meant none of the existing physics API's really stand out in their adoption rate. (PhysX != physics)

 

Yeah, there are a few good PhysX titles and it's pretty cool what it allows, but I don't see PhysX really taking off with only one vendor backing it.

 

Of course I'd like a vendor agnostic physics API to gain some real traction and I don't care which, just come up with something ATI + NVIDIA! Until then... we'll continue to see only a few titles here and there with good effects.

 

Vendor agnostic would be great but i dont see an open source solution.

Share this post


Link to post
Share on other sites

unless the whole idea of PhysX and Havok is changed.
Vendor agnostic would be great but i dont see an open source solution.

Maybe if a big player *coughmicrosoftcough* would step in...

Share this post


Link to post
Share on other sites

Maybe if a big player *coughmicrosoftcough* would step in...

 

 

There you go stirring the pot even thicker.... Everyone would have some kind of problem with that ,much the same way they have a problem with everything else that MS does.

Share this post


Link to post
Share on other sites

There you go stirring the pot even thicker.... Everyone would have some kind of problem with that ,much the same way they have a problem with everything else that MS does.

Yes and no...it entirely depends on how they handle it. If they add it to DX12, for example, I don't think anyone would have a problem with that.

Share this post


Link to post
Share on other sites

The problem with open source is that snyone can drop the ball and quit on the project halfway through it and have development fizzle to a halt.

Uhh...no. That's what happens with proprietary stuff. With open source anyone can contribute and any one person leaving isn't going to screw the project over. The only time anything really gets abandoned is when nobody wants to work on it anymore - this generally coincides with nobody wanting to use the software/project. If someone wants to use it; someone will work on it.

 

 

There you go stirring the pot even thicker.... Everyone would have some kind of problem with that ,much the same way they have a problem with everything else that MS does.

Adding support for GPU accelerated physics is exactly what they've done with DirectCompute in DX11.

Edited by Waco

Share this post


Link to post
Share on other sites

Please sign in to comment

You will be able to leave a comment after signing in



Sign In Now

×
×
  • Create New...