Does Physx really make a difference?

MelonSplitter

[H]ard|Gawd
Joined
Aug 6, 2006
Messages
1,142
I just played the Batman demo which looked pretty good on my 4870x2. When will ATI develop something like Physx.
 
Never. But DX11 makes provisions for physics compute on GPU's, so we should be seeing the taken up by developers in the next 1+ years.
 
well I have a physx card and I have yet to buy a game that uses it so I consider it worthless. and with cpu physics coming along nicely I don't see the need. but as said open cl and DX11 will both allow it.
 
well I have a physx card and I have yet to buy a game that uses it so I consider it worthless. and with cpu physics coming along nicely I don't see the need. but as said open cl and DX11 will both allow it.

Bet On Soldier ;)

as far as AMD with physics, they're just laying GPGPU framework for software to come along and make use of it, just like nVidia (while they own PhysX, nothing uses it, and its not an open standard, so I doubt it'll go anywhere), give 'em a few years and it'll be ridiculous

as far as having the 4870X2 itself doing this stuff, doubt we'll ever see that, sadly
 
Yes, in the games that support PhysX. Such as, Mirror's edge. Or the Physx specific UT3 stages. Otherwise no, you'll see no difference.
 
PhysX is implemented using CUDA by nVidia. ATi had the chance and opportunity to do the same using their Stream GPGPU solution, but refused and are still stuck with CPU-only Havok.

DX11 will include no physics API, nor will OpenCL. OpenCL and DX11 (OpenGL GLSL equivalent) however can be used to implement a physics API like PhysX.

I think that at this point we'll see PhysX begin making Havok irrelevant as hardware accelerated physics will be the next (only?) big new thing in games. We have pushed graphical realism as far as we can so far, it's time to make the physics side catch up, and the only way to do that right now is through PhysX.
 
at first I thought it was just another silly implementation of Physx but after going back through the Batman Arkham Asylum demo I am quite impressed. this video illustrates it better than words. http://www.youtube.com/watch?v=vINH6Z9kqgI

I would like to see a comparison of the of normal or high Physx settings in Batman Arkham Asylum though. just to be clear for those that havent played the demo there are 3 settings for Physx...off, normal, and high. normal recommends a gtx260 and high recommends the addition of a 9800gtx+ dedicated for Physx.
 
ATI will never use PhysX....They will only use it if its not from nVidia :p

anyway, PhysX doesn't bring any major differences, and it will lag your game in most part unless you have a dedicated GPU for it....
its really not worth to have a dedicated GPU for physX, seriously...it only bring a minor visual differences, like flying plastic that flow nicely.... :D

games are more toward to Havoc, lots of games we see these days that have physics engine in there are using Havoc...
 
After finishing the demo, I have yet to be really blown away by this PhysX thing.

Let me bring up the new Ghostbusters game as an example: This was some of the most fun i've had with game physics since HL2 came out 5 years ago. Nearly every object aside from the walls themselves could be destroyed or manipulated. The game also runs impressively well on my aging rig so my CPU can't really be TOO strained by the physics.

Also, the framerate loss i've experienced when using PhysX is not worth enabling the feature in my opinion. Sure you could add a dedicated card for it, but it's not cheap.

If you are a dev and have a potentially physics-heavy game, make a tri-core+ setting for the extra bit of flare. Something like: Low - Single core | Med - Dual core | High - More than 2 cores.
 
Except the most powerful 8 thread/4 physical core CPU processes on the order of ~85 GFLOPS. A mid-range GPU processes between 10 times that and 13. Do the math regarding CPU physics and you'll see that for large interactions, you will need a co-processor assist.
 
Last edited by a moderator:
After finishing the demo, I have yet to be really blown away by this PhysX thing.

Let me bring up the new Ghostbusters game as an example: This was some of the most fun i've had with game physics since HL2 came out 5 years ago. Nearly every object aside from the walls themselves could be destroyed or manipulated. The game also runs impressively well on my aging rig so my CPU can't really be TOO strained by the physics.

I think I read somewhere that the infernal engine can automatically scale the physics realism (or at the very least take full advantage of more cpu cores) depending on the number of cpu cores and their performance. When physx first came out it seemed like a really interesting technology. Honestly I thought if they priced those cards under 150 dollars and had a killer-app type deal launch with the cards, they'd make a killing and advance pc gaming physics by years. Unfortunately the cards were more expensive and the killer app just never came. CPU performance and physics has also come along nicely, so physx for physx's sake is just a waste of money at this point. Nice to have if you already own a couple nvidia cards though.
 
Except the most powerful 8 thread/4 physical core CPU processes on the order of ~85 GFLOPS. A mid-range GPU processes between 10 times that and 13. Do the math regarding CPU physics and you'll see that for large interactions, you will need a co-processor assist.

Indeed. If I recall correctly it was something around 5,000 physics objects/particles for a modern CPU and 500,000+ for a decent GPU in a test we ran with our in-house game engine (using PhysX) before we hit a wall. I completely disagree that PhysX slows down a game, even using just a CPU for PhysX is plenty fast. If the game is slow with PhysX-effects enabled, the implementation just sucks as we have encountered no such issues with our games.

To add to the statistics, current-gen GPUs are ~1 TF for high-end models (HD4870x2, GTX285), the upcoming cards (870, GT300) are expected to double that number. It'll be a long time off before CPUs will get anywhere near that number on integer or floating point arithmetic.
 
PhysX is implemented using CUDA by nVidia. ATi had the chance and opportunity to do the same using their Stream GPGPU solution, but refused and are still stuck with CPU-only Havok.

DX11 will include no physics API, nor will OpenCL. OpenCL and DX11 (OpenGL GLSL equivalent) however can be used to implement a physics API like PhysX.

I think that at this point we'll see PhysX begin making Havok irrelevant as hardware accelerated physics will be the next (only?) big new thing in games. We have pushed graphical realism as far as we can so far, it's time to make the physics side catch up, and the only way to do that right now is through PhysX.

Havok is not an ATI owned technology, its owned by Havok (the company) and Intel, and they were/are developing a GPGPU variant which will run nVidia or ATI GPUs (and ATI/Intel are working on an OpenCL/Stream SDK variant to utilize Havok SDK as well (I see no reason for them to shaft nVidia, unless nVidia themselves simply say no))

ATI is not "stuck" with anything, as a physics API is something a developer chooses to utilize, only a few games actually use PhysX, even less on GPU, and there are MANY more games with Havok support as opposed to PhysX (so I'm calling fud on your assertions), because any of these games will run fine with nVidia or ATI hardware, basically getting the same performance



Indeed. If I recall correctly it was something around 5,000 physics objects/particles for a modern CPU and 500,000+ for a decent GPU in a test we ran with our in-house game engine (using PhysX) before we hit a wall. I completely disagree that PhysX slows down a game, even using just a CPU for PhysX is plenty fast. If the game is slow with PhysX-effects enabled, the implementation just sucks as we have encountered no such issues with our games.

To add to the statistics, current-gen GPUs are ~1 TF for high-end models (HD4870x2, GTX285), the upcoming cards (870, GT300) are expected to double that number. It'll be a long time off before CPUs will get anywhere near that number on integer or floating point arithmetic.

HD4870X2 is advertised at 2TFlops, but this is unrealistic to compare to a CPU, as we're talking DSPs, its sort of like CBEA (CBEA is a better "middle ground" device though), stuff thats disgustingly parallel will scream along, otherwise, welcome to the land of pipeline stalls, NOPs, and generally sitting on your hands waiting for the DSP to do its thing, whereas modern CPUs can reach ~20 GFlops in "general" execution, and ~100+ with SSE (and ALU #'s are going to be a closer match to the 20 GFlop figure, not everything is FP math ;))

ultimately the real "jump" for everyone is the unreleased Intel/AMD CPUs with many (many many many) cores (and they aren't that far away, and should easily bring GPCPUs right up to the level of modern GPUs), which are closer to CBEA/T2 in design, and will more or less tear through anything (and nVidia may have a chip to add to this arena, or may not, not much is being said though), ultimately the separate API is a waste of resources and time, it incurs a considerable processing expense (so even if you aren't hitting performance, you still need a co-processor to handle it), doesn't add much to games (oh wow, an extra 10 particles, so amazing) over fully software solutions, and not many developers are even bothering with it (because not many users have support for it)

my thought on PhysX is that its the last in a line of retarded dot-com ideas that sadly made enough noise to get pushed into production, just like A3D, EAX, GLiDE, and so on, a highly specialized proprietary API will always run "better" on custom fitted hardware, and it looks great in reviews, but has little to no general appeal or application, before you argue with that, none of my examples are still considered "in development" or "active"
 
Last edited:
PhysX is implemented using CUDA by nVidia. ATi had the chance and opportunity to do the same using their Stream GPGPU solution, but refused and are still stuck with CPU-only Havok.

DX11 will include no physics API, nor will OpenCL. OpenCL and DX11 (OpenGL GLSL equivalent) however can be used to implement a physics API like PhysX.

I think that at this point we'll see PhysX begin making Havok irrelevant as hardware accelerated physics will be the next (only?) big new thing in games. We have pushed graphical realism as far as we can so far, it's time to make the physics side catch up, and the only way to do that right now is through PhysX.

LOL

OpenCL had adopted Havok's API (for GPU and CPU). It will load balance varying on the hardware detected using the balance which provides the best performance.

PhysX never really got a start and it seems it will die out in the next 1-3 years.
 
I would say not too much yet. However, after just watching the youtube comparison of PhysX on and off, the steam and fog would make the game much more immersive for me.
 
Hardware based physics acceleration can and will be a great thing for games. The problem now is that it's still new, and the market is too segregated. Most game companies aren't going to spend the time and money implementing a feature that only a fraction of customers will be able to use. Until we get more interoperability, things are going to stay the same. It'll be used as an optional extra, and not an integral part of the gameplay.
 
physx just slows down performance :( looks maybe nice with extra effects but not worth the silly fps drops.
 
I would say not too much yet. However, after just watching the youtube comparison of PhysX on and off, the steam and fog would make the game much more immersive for me.

Steam and fog have been in games for years. It's the fact that when you walk through it with Physx, it reacts. However, the dev for the batman game linked above decided to remove things completely if Physx is not enabled.

For that matter, it seems like physx is trying to be forced, by either using it, or not replacing it with a counterpart if not using physx.
 
Steam and fog have been in games for years. It's the fact that when you walk through it with Physx, it reacts. However, the dev for the batman game linked above decided to remove things completely if Physx is not enabled.

For that matter, it seems like physx is trying to be forced, by either using it, or not replacing it with a counterpart if not using physx.

well if you spent thousands to millions of dollars on licencing fees paid out to a company that spent millions of dollars to buy up the technology in a bid to own a larger stake of computer gaming, only to learn that it isn't really going to pay out, what do you think you'd do? be reasonable about it? or try to force it down everyone's throats with an iron fist?
 
Havok is not an ATI owned technology, its owned by Havok (the company) and Intel, and they were/are developing a GPGPU variant which will run nVidia or ATI GPUs (and ATI/Intel are working on an OpenCL/Stream SDK variant to utilize Havok SDK as well (I see no reason for them to shaft nVidia, unless nVidia themselves simply say no))
Havok is an AMD/ATi-licensed technology, owned by Intel. Havok, the company, was bought by Intel a while ago. Havok (the company) was developing Havok FX, but after they got bought out by Intel all work on this API stalled. A version of Havok (API) has been shown to work on GPU by AMD, but it was only a proof of concept, and there are AFAIK no plans to develop it commercially.

ATI is not "stuck" with anything, as a physics API is something a developer chooses to utilize, only a few games actually use PhysX, even less on GPU, and there are MANY more games with Havok support as opposed to PhysX (so I'm calling fud on your assertions), because any of these games will run fine with nVidia or ATI hardware, basically getting the same performance
Correct. Havok is the older physics API of the two, first becoming popular in 2006 with Half Life 2 and has been used in many games. PhysX didn't enjoy popularity until 2008 when the company which created PhysX got bought by nVidia. Since then there has been a huge uptake of PhysX, with it becoming available for SDKs on all current consoles and huge publishers/developers like EA licensing it as well. A major advantage PhysX has over Havok especially for smaller developers like my company is that using Havok requires a license, which isn't cheap. PhysX can be used commercially without any charge. This is why I think that PhysX will enjoy more growth than Havok the coming years.


my thought on PhysX is that its the last in a line of retarded dot-com ideas that sadly made enough noise to get pushed into production, just like A3D, EAX, GLiDE, and so on, a highly specialized proprietary API will always run "better" on custom fitted hardware, and it looks great in reviews, but has little to no general appeal or application, before you argue with that, none of my examples are still considered "in development" or "active"

I'm not so sure. There are piles of physics libraries, yet PhysX at this point is the only one which can be hardware-accelerated right now. Havok FX is dead and buried and Intel isn't going to let AMD resurrect it. Not until it can benefit from it with Larrabee or so.

My company uses PhysX in all of our games. We are planning games which utilize PhysX each to a different extent. We are even considering games which will require an nVidia GPU for HW PhysX since the market for it is reasonably large enough.

We didn't pick Havok because it's expensive, its API isn't as elegant as Havok's and it is limited to CPU-only. I'm fairly certain that these same points will make PhysX more attractive for other companies as well.

well if you spent thousands to millions of dollars on licencing fees paid out to a company that spent millions of dollars to buy up the technology in a bid to own a larger stake of computer gaming, only to learn that it isn't really going to pay out, what do you think you'd do? be reasonable about it? or try to force it down everyone's throats with an iron fist?
You don't seem to realize that it is incredibly hard to 'scale' the amount of physics in a game dynamically. You'd have to swap out physics-enabled objects for static ones or ones with a less fine mesh or so, or remove them all together. The latter option is far easier and thus is chosen because the former option would add months to the development time.
 
here is a comparison for Cryostasis http://www.youtube.com/watch?v=kufpPwho9Ec&NR=1


this game of course is very sluggish on highest settings with advanced Physx effects turned on.

The water droplet movements are fairly nice, but this has two shortcomings:

1. There's not enough water droplets to give those streams the right density. I think we need another jump in hardware capability before we can really model water well in real-time.

2. The water looks like crap because GPU shaders are not yet up to the task of rendering water droplets and streams in real-time.

Right now it's a gimmick because the hardware isn't there yet. But give it a few generations (and hopefully DirectX 11 / OpenCL), and I think we'll see good real-time water.
 
PhysX is implemented using CUDA by nVidia. ATi had the chance and opportunity to do the same using their Stream GPGPU solution, but refused and are still stuck with CPU-only Havok.

DX11 will include no physics API, nor will OpenCL. OpenCL and DX11 (OpenGL GLSL equivalent) however can be used to implement a physics API like PhysX.

I think that at this point we'll see PhysX begin making Havok irrelevant as hardware accelerated physics will be the next (only?) big new thing in games. We have pushed graphical realism as far as we can so far, it's time to make the physics side catch up, and the only way to do that right now is through PhysX.

I agree with all of this except the part about Havok becoming irrelevant. I think it will do just fine on the CPU accelerated side of things. For the more advanced physics that the video cards are doing with PhysX, I do think they are way far behind though.
 
The problem for Physx is that developers are not intergrating it enough. it is not being used to to the extent that it has to be used in the game. Developers are doing that with havoc.:eek:
 
I would say not too much yet. However, after just watching the youtube comparison of PhysX on and off, the steam and fog would make the game much more immersive for me.

well having played with it a little I can already tell you I am turning it off. not worth the performance hit. the game looks better then I thought it would though. I am going to have to play with the settings more.
 
The problem for Physx is that developers are not intergrating it enough. it is not being used to to the extent that it has to be used in the game. Developers are doing that with havoc.:eek:

It just seems that way because PhysX can be used both for CPU and GPU physics so when you turn off the fancy physics effects in say Mirror's Edge you're turning off the physics objects (models) which would have turned the CPU into a smouldering piece of silicon and switches back to maps which can be done on the CPU. Using PhysX.
 
The question isn't how many games use Havok vs PhysX. It's how many "new" games use Havok vs PhysX. As far as I know it's quite a tossup. For one, every single UE3 game uses PhysX by default.
 
That batman video was a jokem nothing there was anything other than gfx which you do not need a co-processor to acheive.

Cloth effects are nothign new, paper being disturbed is not anything much, still only gfx and the damage to the ground has been done for years unless you can pick up the debris and use it as a weapon.

Absolutely pathetic, when will we see a game that "needs" gpu accelerated physics. Most of that stuff like steam could be done quite easily I would imagine on a standard pc, it is nothing impressive.
 
Your imagination is wrong. Interactive steam would hit a CPU very hard. A lot of people make the mistake of equating "faked" effects with the stuff PhysX is doing. For example, while the floor tile destruction is pretty lame visually it's fully dynamic and you can interact with the broken pieces of tile (break them into smaller pieces). In the past if you did that you'd get a "broken tile" texture decal that was flat and static.
 
Your imagination is wrong. Interactive steam would hit a CPU very hard. A lot of people make the mistake of equating "faked" effects with the stuff PhysX is doing. For example, while the floor tile destruction is pretty lame visually it's fully dynamic and you can interact with the broken pieces of tile (break them into smaller pieces). In the past if you did that you'd get a "broken tile" texture decal that was flat and static.

There's a big difference between rigid bodies, soft bodies, fabric and particle simulation as well. CPU physics engines are pretty much limited to the first one, with some use of the second. Fabric and particle (steam, water) would be absolute murder on a CPU and the latter already strains GPUs (try a real-time fluid simulation for example).

Breaking up tiles would be rigid body simulation, and could be done on CPU only to some extent (until you define a too detailed mesh and have 10,000 tiles which you destroy at once). Fabric is also something which CPUs can't do realistically, like simulating for example a hallway with curtains along both sides which are being moved by airflow.

To get back to your example, what many game devs resort to is 'faking' things like fog using methods built into OpenGL and D3D. What this lacks is interaction. To have fully interactive fog or steam (swirling dynamically around objects and such) would require quite hefty calculations and would be difficult even with GPU acceleration (fluid dynamics!).
 
Yes, but it is not interactive, it does nothing but clear away unless i am mistaken. Why would you need to do anything more than detect when he enters an area where the steam is and do the same animation?

You would notice no difference.

I also cannot see the need for the floor tiles, what purpose do they serve? it is no different than shooting something and it generating dust or whatever, it does nothing.

Sorry but I am just not easily impressed by this sort of thing, a shiny new visual effect is one thing, i can see that, however, I do not want to have to waste power on useless visual effects which is all these are.
 
Indeed. If I recall correctly it was something around 5,000 physics objects/particles for a modern CPU and 500,000+ for a decent GPU in a test we ran with our in-house game engine (using PhysX) before we hit a wall. I completely disagree that PhysX slows down a game, even using just a CPU for PhysX is plenty fast. If the game is slow with PhysX-effects enabled, the implementation just sucks as we have encountered no such issues with our games.

To add to the statistics, current-gen GPUs are ~1 TF for high-end models (HD4870x2, GTX285), the upcoming cards (870, GT300) are expected to double that number. It'll be a long time off before CPUs will get anywhere near that number on integer or floating point arithmetic.

That is awesome. :cool:

Honesty, I have to say that PhysX is really cool, but it only adds about 1% of content to any game and isn't necessary to run on a GPU or PPU specifically. For the cost/performance ratio to specifically buy a GPU JUST for this isn't really required yet. A CPU can do just fine as long as enough cores are available.

If a spare GPU is available, I totally think it's worth it and if it REALLY makes a difference in a game.

I'm just saying, for the cost/performance ratio, it's not worth it.

But yeah, the CPU could never do what a GPU could do in terms of raw performance.
 
Last edited:
PhysX is not worth the trouble right now.What few games out there
thtat do support physx will make your fps lag.

Damn if you do and damn if you dont.
 
Honesty, I have to say that PhysX is really cool, but it only adds about 1% of content to any game and isn't necessary to run on a GPU or PPU specifically. For the cost/performance ratio to specifically buy a GPU JUST for this isn't really required yet. A CPU can do just fine as long as enough cores are available.

To get all technical and stuff, graphics is a subset of physics. Both use linear algebra to calculate vectors and matrices. In the Real World (tm) visualization ('seeing') forms only a minor fraction of the entirety of our physical experience of and interaction with our environment. The part of the EM spectrum we perceive ('light') ranges only across a few hundred nm. Physics stretches far beyond just that. In essence you could say that our eyes just allow us to see what the other 99.99% of physics affects and changes in our environment. In that respect it's somewhat crazy to consider that in computer games 99.99% is focused on visualization and the actual environmental interactions crammed in as an afterthought. Understandable considering the limitations of the medium (2D, only visual and auditory feedback, limited processing power), but still interesting.

The most realistic game will therefore be basically a big physics simulation :)
 
Steam and fog have been in games for years. It's the fact that when you walk through it with Physx, it reacts. However, the dev for the batman game linked above decided to remove things completely if Physx is not enabled.

For that matter, it seems like physx is trying to be forced, by either using it, or not replacing it with a counterpart if not using physx.

lol the way it's meant to be played. AMIRITE?
 
LOL

OpenCL had adopted Havok's API (for GPU and CPU). It will load balance varying on the hardware detected using the balance which provides the best performance.

PhysX never really got a start and it seems it will die out in the next 1-3 years.

"Open CL adopted Havok's API" ? :rolleyes:

You don't even seem to know what Open CL is. Open CL doesn't adopt anything. It's just a standard to use GPU resources directly. A physics API is third-party and can only be used on a GPU through OpenCL, if it's translated to either CUDA or Brook+ and this is done through drivers. NVIDIA has that option. AMD does not at this point.

As for PhysX not getting a start, again you don't even know what PhysX is. PhysX is a physics API. It's highly irrelevant if it's done on the CPU, GPU or PPU, but it is much more flexible than Havok since the user can choose where he/she wants to execute the physics calculations. And no! Despite your red fan dreams, what AMD showed with OpenCL and Havok running on GPU, that's simply not available for anyone.
 
That is awesome. :cool:

Honesty, I have to say that PhysX is really cool, but it only adds about 1% of content to any game and isn't necessary to run on a GPU or PPU specifically. For the cost/performance ratio to specifically buy a GPU JUST for this isn't really required yet. A CPU can do just fine as long as enough cores are available.

If a spare GPU is available, I totally think it's worth it and if it REALLY makes a difference in a game.

I'm just saying, for the cost/performance ratio, it's not worth it.

But yeah, the CPU could never do what a GPU could do in terms of raw performance.

You're saying that because you are only looking at what you've been given this far. We never had realistic physics effects before. The Ghostbusters game seems to be mentioned as a poster child, for the continued use of CPU for physics calculations, but the people that advocate it as an example, conveniently forget the main point. Ghostbusters physics effects ARE not realistic in any way. You can knock almost anything to the ground, but that's the same physics effects found in games like Doom 3 and HL2. You shoot at a table towel and it just changes color. It doesn't burn slowly with all the effects we see in RL. You also don't have realistic cloth or hair simulations, that add so much realism to a NPC in game. Water is another one of the very hard effects. We never had a game with realistic effects. Bioshock came close, but they are static effects for eye candy only. There's no variation. But they can be more realistic with proper physics calculations.

All these effects are huge problems in the making of 3D movies for example. Pixar (as an example) loses a great amount of time to achieve realistic cloth and hair simulations for their movies (I know that for The Incredibles, production was having some setbacks regarding Violet's hair). And we are talking about animation that takes roughly 20-30 hours to render each frame. What PhysX through the GPU offers, are those realistic physics simulations in real time. Obviously, you'll get a performance hit if you do those physics calculations, on a graphics card that's doing both graphics and physics at the same time, but even then, performance is already better than even on the fastest CPU doing those physics calculations by itself.
As hardware (in this case GPUs) get more powerful and more efficient for these types of calculations (NVIDIA is investing in that ever since G80, but especially GT200), physics calculations will be executed much more smoothly, even when the GPU is also doing graphics.
 
In the end, having physics calculated on the GPU means that graphics and physics are directly competing with one another... which is a messy situation. If PhysX ever becomes a "big thing" I fear we'll all need SLI setups to handle it without wrecking graphical performance. Which would be great news for NVIDIA, of course!

Aside from that, though, I don't like the situation we're in at the moment where NVIDIA cards do PhysX acceleration and ATI cards don't. There needs to be univeral support for physics acceleration. If there isn't universal support, I'm tempted to say that I'd rather not have physics acceleration at all (though I know that's a very argumentative thing to say!). Basically I don't think having proprietary physics acceleration is a much better situation than having no physics acceleration. With things the way they are at the moment, something has to give way... and I don't think ATI getting muscled out of the market would be a better outcome than PhysX dying out.
 
In the end, having physics calculated on the GPU means that graphics and physics are directly competing with one another... which is a messy situation. If PhysX ever becomes a "big thing" I fear we'll all need SLI setups to handle it without wrecking graphical performance. Which would be great news for NVIDIA, of course!

Aside from that, though, I don't like the situation we're in at the moment where NVIDIA cards do PhysX acceleration and ATI cards don't. There needs to be univeral support for physics acceleration. If there isn't universal support, I'm tempted to say that I'd rather not have physics acceleration at all (though I know that's a very argumentative thing to say!). Basically I don't think having proprietary physics acceleration is a much better situation than having no physics acceleration. With things the way they are at the moment, something has to give way... and I don't think ATI getting muscled out of the market would be a better outcome than PhysX dying out.

The funny thing about your argument, is that you don't even mention Havok in there and it's as proprietary as PhysX is...So I'm sure you refused to play all the games that supported Havok until now, because they were using proprietary tech...:rolleyes:

Also, PhysX is a physics API (how many times does this have to be said ????). You don't need to have a NVIDIA graphics card to run it, much less SLI, but if you do, you have the felixibility to choose where you want the physics calculations to be done. If you have graphics power to spare, do it on the GPU. If not, use the CPU or even a PPU.

"ATI cards don't" because AMD refused to use PhysX, when they were offered the chance to do it. They preferred to choose Havok instead, which is owned by Intel and, once again, is as proprietary as PhysX is.

You're never going to get a "free" Physics API. Havok isn't one and neither is PhysX, but you already have a standard to "play" with GPU resources directly. And it's called OpenCL, which both NVIDIA and AMD support. PhysX won't die out and neither will Havok (because they are physics APIs!!!!), but the flexibility of one of the over, will give it momentum and PhysX surely has that over Havok right now (deals signed with EA and 2K games for example). But the games that use it, don't come out over night. They take time to develop.
 
Back
Top