PHYSX NEWS PHYSX SDK
PROJECTS TABLE
GPU PHYSX
GAMES INFO
PHYSX
ARTICLES
PHYSX WIKI FORUM
РУССКИЙ ENGLISH


:: Back to news index ::

AMD and PhysX: History of the Problem

with 19 comments

Recent provoking claims by AMD regarding GPU PhysX and all hype around Open Physics Initiative are causing me some Deja Vu feeling.

Why ? Because, actually, hardware accelerated PhysX was criticized by AMD/ATI throughout it’s lifespan (more or less). Let’s take a retrospective look at escalation of the conflict:

In June 2006, just after Ageia PPU launch, ATI unveils their GPU physics processing conception called “Boundless Gaming” (also known as Asymmetric Physics Processing) based on Havok FX engine (technology was supported by Nvidia as well).

X1900 XT GPU was promised to deliver over 9x performance of a PhysX PPU card and games augmented with Havok FX physics were supposed to come out in 2007.

Result ? Havok FX died with Intel aquisition of Havok company, no games were released, 16 titles with Ageia PPU support from hardware PhysX side.

To be honest, AMD was threating PhysX enough politely those times, even considering to buy whole Ageia company:

Richard Huddy saying, “we’ve had that discussion, yes. It’s a discussion that goes round every three months – someone turns to me and says “why don’t we buy Ageia?”

Things started to get worse when Nvidia bought Ageia with all patents, personell, software and hardware developments. GPU PhysX was buried alive by AMD beforehand

“There is no plan for closed and proprietary standards like PhysX,” said [Godfrey] Cheng “As we have emphasized with our support for OpenCL and DX11, closed and proprietary standards will die.”

To support its position, AMD comes to GDC 2009 with GPU accelerated Havok demonstrations, this time with promises to port it to OpenCL platform.

Result ? 7 complitely new titles with GPU PhysX support in 2009, no games with OpenCL Havok, but “PhysX will become irrelevant” claims again and hollow “Accelerated physics processing” label to HD4xxx GPU series feature set.

Meanwhile, whole year passed since than, and now new Open Physics strategy is on the GDC 2010 scene, accompanied by more intensive GPU PhysX abuse

And where is GPU Havok now ?! Following Havok FX in it’s way to oblivion, I guess.

So, why do you think, readers – will this hardware accelerated Bullet & DMM duet took up this time, or it’ll end like other AMD’s attempts to settle on GPU physics field for past 4 years ? Are they really planning to bring peace and unity to developers absolutely disinterestedly, while Nvidia is using GPU PhysX to make money ?

Written by Zogrim

March 11th, 2010 at 8:59 pm

Posted in Articles, Reviews

Tagged with , , ,

19 Responses to 'AMD and PhysX: History of the Problem'

Subscribe to comments with RSS

  1. im really angry …i cant speak english very well :D

      

    MicrO

    11 Mar 10 at 10:33 pm

  2. Zogrim, thanks for that physx/physics resume and i´m really curious about free DMM2. Lookin forward to see it in many future games (or some similar technology).

      

    Vojtech

    11 Mar 10 at 11:26 pm

  3. Why are AMD doing this ‘disinterestedly’? Havok may not be publicly being talked about, but that might not mean nothing is being done about it.

      

    caveman-jim

    12 Mar 10 at 12:27 am

  4. @caveman-jim

    there is no reason for one to assume “something” is being done about Havok, so its only logical to assume to opposite.

      

    applejack

    12 Mar 10 at 1:46 am

  5. Vojtech

    12 Mar 10 at 2:27 am

  6. Hey Caveman- you really get around ol buddy! Is there a site you’re NOT on?!?

    In any case, ATi has been flip flopping on physics for years.

    I think when they were having those “Should we buy Ageia” talks someone in accounting piped up and said “With what?!”.

      

    Rollo

    12 Mar 10 at 3:35 am

  7. applejack
    My thoughts exactly)

    Vojtech
    Here, first demo video of DMM2 and Open Physics from GDC 2010
    (in the middle of the page)

      

    Zogrim

    12 Mar 10 at 6:55 am

  8. When i think about it, i really can’t say that ati/amd is to be blamed for absence of game titles with their physics technology. You see, the work of GPU vendor is to invent and deliver such technologies. Promoting is somewhere beyond this line for me.

    In any case, why would GPU vendor even want to help game developers with anything? Regular Q/A may be settled with a couple of emails. Developing the game is the work of game designers, and nobody sould bias them into using any proprietary technology.

    Now imagine, you are developer who is creating some sort of AA/AAA-game title. Some GPU vendor is contacting you and says that if you choose PhysX, they’ll help you with resources and integration. What would you do? And were you even planning to integrate any physics mechanics into your game before that?

    Yes, perhaps some PhysX implementations feel great in some titles. But did anyone ever try to understand why would a normal unbiased game dev even want to integrate a GPU-based physics processing of tree leaves into their RPG project? And cloth effects into parkour-on-the-roofs action? (come on, breaking simple glass shouldn’t really halve your fps)
    And maybe someone thinks that spreading 30k particles from killed aliens is really a good idea?
    I may be wrong, but i think that not all these ideas came to developers from their own mind. And sometimes it feels more like another tech-demo instead of a game.

    I think, if PhysX@GPU somehow would be *freely* available for both vendors by 2008 already, we would have more games utilizing it by now. And they wouldn’t even need to “help” developers with the implementation.
    But i’m sure, even years after first dx11 GPUs they will try to push PhysX hard again, limiting the competitition opportunity. At this rate this will only slow down the game industry imo.

    Good thing is promoting open-standart-based engines, yes. But where is nvidia now? They only say “Yeah, we support that DirectCompute and OpenCL. But guys, just look at our PhysX, it’s awesome!”
    Can we say that not-promoting open-standart-based things is a bad thing? No, of course, “we can’t really blame nvidia for that”… i see this phrase too often lately.

      

    GenL

    12 Mar 10 at 8:30 pm

  9. GenL
    When i think about it, i really can’t say that ati/amd is to be blamed for absence of game titles with their physics technology
    But they can be blamed for constant griefing of opponents technology. Especially when attack mostly have black-PR nature.

    I may be wrong, but i think that not all these ideas came to developers from their own mind. And sometimes it feels more like another tech-demo instead of a game.
    It’s obvious that mostly effects are implemented with help of Nvidia content-devs, but people prefer to see negative sides from this, and how about positive:
    1) GPU PhysX titles are really coming out (not on slides or paper), and such effects are really echancing immersion.
    2) GPU PhysX content supply free CPU/console PhysX SDK and following tools. Isn’t that good for developers ?
    Compared to other free physics engines, PhysX SDK (with all its Ageia/Nvidia proprietary stuff) is much much more widely adopted.
    Even semi-free PC Havok initiative (started in May 2008 by Intel), supposed to kill or at least counter free PhysX SDK, haven’t much effect.

    I think, if PhysX@GPU somehow would be *freely* available for both vendors by 2008 already, we would have more games utilizing it by now.
    I don’t think so. History is showing that mostly devs aren’t interested in hardware accelerated physics of any kind. Havok is a good indicator:
    Devs require better authoring tool ? They created Havok Cloth and Destruction
    Devs require better integration of physics with other game engine systems ? They created Havok Animation, AI and Behaviour
    Devs require hardware physics ? Seems no, and all such Havok projects failed (btw, forced by GPU manufacturers all the time).

    From my POV, it’s better to have at least something, than nothing :)

      

    Zogrim

    13 Mar 10 at 2:28 pm

  10. Black-PR? Maybe. But i personally prefer to support people who says “proprietary standarts should and will die” because i agree, while others say “PhysX is the only right way for physics, and we’ll do anything we can to provide support for devs with it”.

    PhysX titles? Yes, they are coming out, but that is only causing further confusion for PC gamers. You know – PC, not %gpu_vendor_here% PC.

    Free SDK tools? I think they always should be free, that’s why everyone should support open standarts. I know, for example, DX itself is not really *open*, but at least it is supported by all vendors from the beginning.

    Devs are not interested? Maybe true, but i’m sure there will be at least more progress for fully-destructable-environment projects, because hardware acceleration would minimize optimization needs for the engine. And these projects could be used by many games already, developers won’t hesistate to use solutions which are not vendor-locked.

    For now i just think we all should hope that many good projects with OpenCL/DirectCompute will emerge, and that no one will try to prevent this.
    When you must buy a new hardware to upgrade your PC, it’s fine. But when for specific features of the gameplay experience you have to stick with some particular vendor, something is wrong. And DX11 is not a case.

      

    GenL

    13 Mar 10 at 8:24 pm

  11. GenL, the only thing “wrong” here is ATI f*cking their customers by refusing to support PhysX.

    if Nvidia refused to support DX11, would you then think DX11 should die ?

      

    Gibor

    13 Mar 10 at 9:51 pm

  12. But i personally prefer to support people who says “proprietary standarts should and will die” because i agree, while others say “PhysX is the only right way for physics, and we’ll do anything we can to provide support for devs with it”.
    I actually prefer people who not just say, but do ;)

    As for rest, well, lets just see how it will turn out)

      

    Zogrim

    13 Mar 10 at 9:54 pm

  13. Gibor, they already said that nvidia lied about that proposition. Source: http://www.bit-tech.net/bits/interviews/2010/01/06/interview-amd-on-game-development-and-dx11/1

    As for DX11, it’s just silly to assume they would refuse it. They refused DX10.1, and look – there are almost no games using it right now. We could play Assassin’s Creed I with higher fps, but they won’t allow us. A key word here is “limiting”, not a “progress”.

      

    GenL

    14 Mar 10 at 5:24 pm

  14. GenL, you didn’t expect ATI to say “we refused the proposition then, now its too late”, right ?
    so they said “nvidia lied” and you eat that without looking back. typical for blind ATI fans.
    if the proposition was a lie, you would know that back then. long time has past, not only they (probably) refused the proposition, but also virtually killed PhysX and spawn false accusation.
    sure, they can now “go whistle”, as they DO NOT DESERVE another PhysX proposition from NVIDIA.

    regarding DX10.1, if you accuse NVIDIA for not supporting it back then, you should also accuse ATI for not supporting PhysX back then.
    only difference is NVIDIA made sure GPU PhysX is integrated in games, while ATI haven’t done sh*t with DX10.1, as they always talk more than do.

      

    Gibor

    15 Mar 10 at 7:51 pm

  15. Gibor, it’s too easy to say that amd lied too. But then nvidia would have already responded to that publically. When they say they had conversations about technology, it’s not something real people would lie about.

    Comparing DX10.1 with PhysX is useless. First one is a generic gfx standart of Misrosoft OS which is not a property of any GPU vendor.
    It’s actually quite obvious that nvidia would never allow any other vendor to support PhysX@GPU, no matter what. And you’ll also see that it won’t be ported to anything but CUDA, no matter what. Because that’s what nvidia about.

      

    GenL

    15 Mar 10 at 8:07 pm

  16. Gibor, give me a PhysX driver from my 9800GTX that works along side a 5890 on Win 7. Otherwise, I support the “nvidia lied” quote!

      

    Emm

    23 Mar 10 at 4:00 pm

  17. idk why I wrote 90… it’s a 5870, so? an answer?

      

    Emm

    25 Mar 10 at 3:49 am

  18. GenL

    [Black-PR? Maybe. But i personally prefer to support people who says “proprietary standarts should and will die” because i agree, while others say “PhysX is the only right way for physics, and we’ll do anything we can to provide support for devs with it”.]

    Yes, but if you’re saying that “proprietary standards should and will die” and you are using Havok for your “physics on GPU” implementation, something is wrong with what you’re saying.

    And…. helping Bullet dev is not an AMD prerogative, in fact the developer is using NVIDIA hardware and their code sample.

    http://www.fudzilla.com/content/view/15642/1/

    http://tech.icrontic.com/news/nvidias-take-on-amds-open-source-bullet-physics/

      

    Stefem

    30 Mar 10 at 7:01 pm

  19. Nvidia bought out Ageia PhysX so it would be propritery so that AMD/ATi Cant use thos technologies, i know because i bought an ageia PhysX card, just a bit too late, i bought one few months after nvidia boughtthem out, my ageia PhysX card WOULD NOT WORK WITH MY ATI/AMD Video Card cause ageia didnt support thr driver anymore i had to download the driver from nvidias web site and it wouldnt even detect the Ageia PhysX card with My ATI GPU installed, i replaced my ATI card with an nvidia GPu and OH SHIT, the AGeia card was suddenly detected…

    thier was/ probably still is and ATI PHYSX LOCK in the Nvidia Driver software????

    HERE IS THE LINK STATING THAT NVIDIA CUT PHYSX FROM AMD SYSTEMS???
    CLICK
    http://www.techspot.com/news/36436-nvidia-cuts-physx-if-amd-card-is-present.html

      

    krazispeed

    11 Aug 12 at 6:56 am


Leave a Reply

*
Copyright © 2009-2014. PhysXInfo.com | About PhysXInfo.com project | Privacy Policy
PhysX is trademark of NVIDIA Corporation