• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

AMD says GPU physics is dead until DirectX 11

lol, says the company who is lagging behind on a physics graphics solution and a physics processing solution (multi-cores).
 
Lagging behind or not, they are a CPU/GPU maker nonetheless. You're a disgrace if you insult AMD while you're in five AMD clubs.
 
Sounds realistic from AMD, Well the only people I see hyping about physics are the idiots who bought a card living in denial saying how great it is while all they have is UT 03 and finally one other game that came out. Money well spent to watch a bunch of demos of balls bouncing and a water tower falling. I see AMD and Nvidia integrating it into their cards or motherboards before long. I mean really physics is pretty worthless right now. But AMD stating the obvious I guess offended this guy.
 
Yup, pretty much have to wait for dx11 for that. Dx10 is still like a dx9c with extra sugar and spices on top. Dx11 will be night and day most likely.
 
Gee did anyone read the article?I dunno but from what I read the main reason AMD is saying not now is because they were waiting for a common ground which was Havok, now they are waiting on Micro$oft to provide one with DX11.

This has little to do with development but everything to do with marketing, Intel bought it for parallel computing it might not even see the day of light as a physic middleware but as a multicore parallel tool for general purposes, which would pay off a lot more to Intel.

In this multicore race, the first to provide true and better multicore support will big $$$.

Game developers want something easy that can apply to any platforms, they are not going to use a propriatary software that will limit them in supporting one hardware.Game development doesn't limit itself to PC-ATi-NVIDIA, it goes way beyond to arcade and consoles.That's the main reason PhysX is so appealing because it support hardware and software, windows, linux and ps3-wii-xbox360.
 
And still, when people hear the word AGEIA, they cannot stop the insulting and the bashing.
Physx is the best physics engine available, because HAVOK is dead.
 
And still, when people hear the word AGEIA, they cannot stop the insulting and the bashing.
Physx is the best physics engine available, because HAVOK is dead.
This is just flamebait.

Havok is far from dead when so many great games use it across all platforms, Windows, PS3, and Xbox. From the venerable Source Engine and all of those titles, to recent titles like Halo3, Bioshock, and Hellgate.
 
Strange?
iNtel rapes HavokFX and kills it.
Put's PPU direct compitor GPU back for years till DX11 is out, around wenn iNTel will be avaible with its larrabee to counter GPGPU computing.

While ageia and AMD are bashed yet again. Why?

Havoks hands are tight by iNtel to only support CPU till iNtel have something more ready to be supported.

Why AMD say this is because most devs depend on middleware and Havok is the defacto standard middle ware to do Physics. So iNtel hit Physics acceleration by GPU hard.

So it's obvious a way to steer the market away from GPU for now.
Because CPU is threatened by GPGPU, see Folding@home and Avivio videoencoding
And gaming is a billion doller bussnes wich has influence on the hardware market.

I don't know if DX11 will be that spacial. And for this case of discusion not relevant It will possible just come with the first MS DirectX Physics component.

And then, will it be a low level standerised Physics API or equal to PhysX and Havok and newton as a highlevel Physics API/LIB?
Because DirecX SDK isn't a game-engine or middle ware. But more a hardware abstraction layer.
Like the older Immidiate mode vs retainmode from Direct3D
 
Strange?
iNtel rapes HavokFX and kills it.
Put's PPU direct compitor GPU back for years till DX11 is out, around wenn iNTel will be avaible with its larrabee to counter GPGPU computing.

While ageia and AMD are bashed yet again. Why?

Havoks hands are tight by iNtel to only support CPU till iNtel have something more ready to be supported.

Why AMD say this is because most devs depend on middleware and Havok is the defacto standard middle ware to do Physics. So iNtel hit Physics acceleration by GPU hard.

So it's obvious a way to steer the market away from GPU for now.
Because CPU is threatened by GPGPU, see Folding@home and Avivio videoencoding
And gaming is a billion doller bussnes wich has influence on the hardware market.

I don't know if DX11 will be that spacial. And for this case of discusion not relevant It will possible just come with the first MS DirectX Physics component.

And then, will it be a low level standerised Physics API or equal to PhysX and Havok and newton as a highlevel Physics API/LIB?
Because DirecX SDK isn't a game-engine or middle ware. But more a hardware abstraction layer.
Like the older Immidiate mode vs retainmode from Direct3D

so in summary... huh?
 
I guess this confirms that the physics API that mircosoft is suppose to be working on will be included in DX11. And I think a common physics by Microsoft is the only sensible way ahead. Hopefully developers will then start supporting in.

However, this is nothing new or surprising. I don't believe that things would have been different had Intel not bought Havok. Microsoft has long been known to be developing their own physics API, so regardless of whatever happens to Havok, developers will eventually choose Microsoft API over Havok's FX.

Havok FX never really got any game to support it anyway, so in a way, GPU physics can't dies for now because it never did took off in the first place.
 
It makes a big difference.
MS DX11 Physics is far future.
For the near future HavokFX makes GPGPU compete with iNTels multicore CPU.
iNTel don't have a aswer to this in the near future.
But then they will have.
So they bend the market for the near future to favor CPU for computational task.
Wenn they can't hold that they will have there own GpGPU or CPU hybrid ready.

But this is to ensure the CPU bussnes will by a big money cow for them. Maxemise profits. Keep those shareholders happy every year.
 
Obviously you're not getting it so just exit the thread quietly.

Doesn't look like you got his point or my point either way. What I'm saying is he was sort of mocking the idea that AMD is lagging behind and mentions VIA, which was not even part of the discussion but okay.

I guess this confirms that the physics API that mircosoft is suppose to be working on will be included in DX11. And I think a common physics by Microsoft is the only sensible way ahead. Hopefully developers will then start supporting in.

However, this is nothing new or surprising. I don't believe that things would have been different had Intel not bought Havok. Microsoft has long been known to be developing their own physics API, so regardless of whatever happens to Havok, developers will eventually choose Microsoft API over Havok's FX.

Havok FX never really got any game to support it anyway, so in a way, GPU physics can't dies for now because it never did took off in the first place.

True - however we've seen many demonstrations when Intel/ATI tried to show off 3 video card systems while the top card does the physics (I can't remember, it was being used by their Bad Axe 975X board) and it seemed to work pretty well. A X1600 would deliver plenty of physics processing power but that idea isn't recognized yet. Also DX10 just got introduced, so I'm not sure how long it'll take for us to enter the world of DX11. Isn't Havok owned by Intel?
 
Who cares if you can use a GPU for physics, it isn't the kind of physics that actually affects gameplay anyways. In order to get physics that does anything to gameplay, you'll need the ageia physx card or continue to do it on the CPU which is the slowest way.
 
Physx is a very tricky subject, as the cards are not "needed" to enjoy the game. My graphics card draws the information on the screen, my processor calculates everything, my ram stores some of the data, and my hard drive is the main source for all of it. My sound card processes the sound.

Physx cards are not "needed" to actually play any game on the market. They are just sort of cosmetic, but in a deeper gameplay level. I dont think external physx cards will ever become mainstream, or atleast not for a very long time.
 
Physx is a very tricky subject, as the cards are not "needed" to enjoy the game. My graphics card draws the information on the screen, my processor calculates everything, my ram stores some of the data, and my hard drive is the main source for all of it. My sound card processes the sound.

Physx cards are not "needed" to actually play any game on the market. They are just sort of cosmetic, but in a deeper gameplay level. I dont think external physx cards will ever become mainstream, or atleast not for a very long time.
Your very quote was used as an arguement against video cards 10 years ago..
 
There is no need for an API when there is no competition. With that in mind, I think having an API would help.
 
"10 years ago you had GLQuake"
GLquake came out years after the first 3Dcards. We already had up to if not over 3 years of 3Dgames before that.
 
And still, when people hear the word AGEIA, they cannot stop the insulting and the bashing.
Physx is the best physics engine available, because HAVOK is dead.



So if it is the "best" why is it hardly used? since it is the best and all....
 
"10 years ago you had GLQuake"
GLquake came out years after the first 3Dcards. We already had up to if not over 3 years of 3Dgames before that.

The Diamond Monster 3D was on shelves 1996 in August.

"Together with the Voodoo 1 3Dfx delivered also its own software interface to their graphics cards: GLIDE" http://www.x86-secret.com/articles/divers/v5-6000/v56kgb-2.htm

"On January 22, 1997, id Software released GLQuake" - http://en.wikipedia.org/wiki/Quake#GLQuake_and_WinQuake

The first 3d cards in general released in 1995 - http://en.wikipedia.org/wiki/Graphics_card

In less than 2 years after 3d cards were released, GLQuake was released. The game that 'really' pushed 3d to the homes of thousands.
 
The Diamond Monster 3D was on shelves 1996 in August.

"Together with the Voodoo 1 3Dfx delivered also its own software interface to their graphics cards: GLIDE" http://www.x86-secret.com/articles/divers/v5-6000/v56kgb-2.htm

"On January 22, 1997, id Software released GLQuake" - http://en.wikipedia.org/wiki/Quake#GLQuake_and_WinQuake

The first 3d cards in general released in 1995 - http://en.wikipedia.org/wiki/Graphics_card

In less than 2 years after 3d cards were released, GLQuake was released. The game that 'really' pushed 3d to the homes of thousands.

Game visuals and game physcis are 2 different sizes that cannot be directly compared.
 
IT can be compared, but the question SHOULD it be compared.

Visuals for the gaming world greatly affect the player's interaction.

Physics CAN greatly affect the player's interaction if the game takes advantage of it. Just as in quake. It didn't originally take advantage of 3d cards at first.
 
IT can be compared, but the question SHOULD it be compared.

Visuals for the gaming world greatly affect the player's interaction.

Physics CAN greatly affect the player's interaction if the game takes advantage of it. Just as in quake. It didn't originally take advantage of 3d cards at first.

You are still comparing apples and oranges...
 
Visuals often don't influence interaction with the player. But can do in some ways. It often influence player emersion more. That why we want it to. Put Realism on a higher level.
Like Bump mapping, FSAA, AF HQ textures. Nice visual features that don't interact but make it look better, more real.
How ever. HDR and Fog do interact in a way. It can conseal you, hide in shadow. Or with the sun in your back you can blind your enemy for your attack so you have more chance of succes. Motionblur Fog or HDR must be able to be set fixed at gameserver. It can influence gameplay.
High resolution influence Friend foo identification and object identification on long distances.
So visuals can have gameplay influence and or interaction.
altho its in a restricted indirect manner.

Physics can be effect Physics, wich falls also under visual eye candy. And have no interaction. On player but offers yet again a large boost of emersion.

Gameplay Physcics do interact ofcourse.

So you have interactive visuals that influence the player.
And non interactive visuals that don't influence the player in game.

But most visual feature are non interactive or minimal.
And most physics features can be of heavy gameplay kind and most of them also be used as a effect feature. The difference using the retrieved Physic computation result in the gameplay logic/mech. or not.
 
Yup, pretty much have to wait for dx11 for that. Dx10 is still like a dx9c with extra sugar and spices on top. Dx11 will be night and day most likely.

I'm pretty sure they were saying the same garbage about dx10. I dont see any legitimate reason for why one needs to wait for dx11 to do what can already be done on dx9c.
 
You are still comparing apples and oranges...

Not really...

While the two products may be as different as apples are to oranges, the discussion on their introduction to the market is an apples to apples comparison.

10 years ago everyone was peaches and creams with a 2d gaming world. Why on earth would I ever want to buy 200 dollar add-in card to get 3d?

10 years later...

My linear non-interactive world is treating me just fine. It doesnt hurt me and I dont hurt it, shit is straight okee dokee with me! Why on earth would I want to increase my interactivity and enjoyment by buying a 200 dollar add-in card to get physics acceleration?

The two situations are near identical from a marketing standpoint. You have a market thats been at a certain status quo for ages and everyone is overly content with the status quo. Then you have a company thats trying to convince the legions of llamas that the status quo can be raised up to unspeakable levels if you would just stop being a group of llamas for a second.
 
Not really...

While the two products may be as different as apples are to oranges, the discussion on their introduction to the market is an apples to apples comparison.

10 years ago everyone was peaches and creams with a 2d gaming world. Why on earth would I ever want to buy 200 dollar add-in card to get 3d?

10 years later...

My linear non-interactive world is treating me just fine. It doesnt hurt me and I dont hurt it, shit is straight okee dokee with me! Why on earth would I want to increase my interactivity and enjoyment by buying a 200 dollar add-in card to get physics acceleration?

The two situations are near identical from a marketing standpoint. You have a market thats been at a certain status quo for ages and everyone is overly content with the status quo. Then you have a company thats trying to convince the legions of llamas that the status quo can be raised up to unspeakable levels if you would just stop being a group of llamas for a second.

Visuals != gameplay physcis.
Neither in apperence or affect on the gameplay.
(This is the part that has importance)
Untill you understand this fact, further debate with you is pointless
 
If a second generation ppu was adopted and used by the mass market it couldnt revolutionize the very way we think of games..... its not that physx aren't the future, it's the fact we, the consumer and the extremely biased and picky hardware consumer, don't want it to be. And for $100, to me it's a steal
 
Something worth noting is that you can SEE graphics, you can HEAR sounds from an audio card, but you can't FEEL the physics added. It's only experienced virtually. Sure it's cool, but it doesn't interact with any of your senses directly like the other cards do.
 
Something worth noting is that you can SEE graphics, you can HEAR sounds from an audio card, but you can't FEEL the physics added. It's only experienced virtually. Sure it's cool, but it doesn't interact with any of your senses directly like the other cards do.

It impacts my brain, all that matters...
 
Back
Top