Where's The Physics: The State of Hardware Accelerated Physics
by Ryan Smith on July 25, 2007 4:00 PM EST- Posted in
- GPUs
More than each other however, there's one other thing that threatens the camps offering hardware physics acceleration: the CPU. Recent years have seen CPUs going multi-core, first with two cores and this week has seen the introduction of the (practically) cheap four core Q6600 from Intel. Being embarrassingly parallel in nature, physics simulations aren't just a good match for GPUs/PPUs with their sub-processors, but a logical fit for multi-core CPUs.
While both AMD and Intel have stated that they intend to avoid getting in to a core war as a replacement for the MHZ race, all signs point to a core war taking place for the foreseeable future, with Intel going so far as to experiment on 80-core designs. With the monolithic nature of games these cores will all be put to work in one way or another, and what better way than physics simulations which can be split nicely among cores? While not the floating point power houses that dedicated processors are, with multiple cores CPUs can realistically keep the gap closed well enough to prevent dedicated processors from being viable for consumers. In some ways Havok is already betting on this with their software physics middleware already designed to scale well with additional CPU cores.
Furthermore the CPU manufactures (Intel in particular) have a hefty lead in bringing manufacturing processes to market and can exploit this to further keep the gap closed versus GPUs(80nm at the high end) and the PhysX PPU(130nm). All of this leads to multi-core CPUs being an effective and low-risk way of going about physics instead of a riskier dedicated physics processor. For flagship titles developers may go the extra mile on physics, on most other titles we wouldn't expect such an effort.
So what does all this mean for hardware physics acceleration overall? In spite of the original battle being between the PPU and the GPU, we're wondering just how much longer Ageia's PhysX software/hardware package can hold out before losing the war of attrition, at the risk of becoming marginalized before any decent software library even comes out. Barring a near-miracle, we're ready to write off the PPU as a piece of impressive hardware that provided a technological solution to a problem few people ended up concerned about.
The battle that's shaping up looks to be between the GPU and the CPU, with both sides having the pockets and the manufacturing technology to play for keeps. The CPU is the safe bet for a developer, so it's largely up to NVIDIA to push the GPU as a viable physics solution (AMD has so far not taken a proactive approach with GPU physics outside of Havok FX). We know that the GPU can be a viable solution for second-order physics, but what we're really interested in is first-order physics. So far this remains unproven as far as gaming is concerned, as current GPGPU projects working with physics are all doing so as high performance computing applications that don't use simultaneous graphics rendering.
Without an idea of how well a GPU will perform with simultaneous tasks, it's too early to call to call the victor. At the very least, developers won't wait forever and the GPU camp will need to prove that their respective GPGPU interfaces can provide enough processing power to justify the cost of developing separate physics systems for each GPU line. However given the trend to move things back on to the CPU through projects such as AMD's forthcoming Fusion technology, there's an awful lot in favor of status quo.
Facebook
Twitter
RSS
32 Comments
View All Comments
Bladen - Thursday, July 26, 2007 - link
When I say first order physics, I mean the most obvious type: fully destructible environments.In UT3, you could have a fully destructible environment as an on/off option without making the game unbalanced in single player. The game is mindless killing, who care is you blow a hole through a wall to kill your enemy?
I guess you could have fully destructible environments processed via hardware and software, but I'd assume that the software performance hit would be huge, maybe only playable on a quad core.
Bladen - Thursday, July 26, 2007 - link
Whether or not the game has fully destructible environments, I don't know.Verdant - Thursday, July 26, 2007 - link
dedicated hardware for this is pointless, with GPU speeds and numbers of cores on a die increasing on CPUs, I see no point in focusing on another pipe.Plus the article has a ton of typographical errors :(.
bigpow - Wednesday, July 25, 2007 - link
Maybe it's just me, but I hate multiple-page reviewsShark Tek - Wednesday, July 25, 2007 - link
Just click on the "Print this article" link and you will have the whole article in one page.Visual - Thursday, July 26, 2007 - link
indeed thats what i do. almost.i hate that printarticle shows up in a popup, can't open it in a tab easily with a middle-click too... same as the comments page btw. really hate it.
so i manually change the url - category/showdoc -> printarticle, it all stays in the same tab and is great. i'm planning on writing a ".user.js" (for opera/greasemonkey/trixie) for fixing the links some time
Egglick - Wednesday, July 25, 2007 - link
Other than what was mentioned in the article, I think another big problem is that the PCI bus doesn't have enough bandwidth (bi-directional or otherwise) for a card doing heavy real-time processing. For whatever reason, manufacturers still seem apprehensive about using PCIe x1, so it will be rough for standalone cards to perform at any decent level.I've always felt the best application for physics processors would be to piggyback them on high-end videocards with lots of ram. Not only would this solve the PCI bandwidth problem, but the physics processor would be able to share the GPU's fast memory, which is probably what constitutes the majority of the cost for standalone physics cards.
This setup would benefit both NVidia/ATI and Ageia. On one hand, Ageia gets massive market penetration by their chips being sold with the latest videocards, while NVidia/ATI get to tout having a huge new feature. They could also use their heavy influence to get game developers to start using the Ageia chip.
cfineman - Wednesday, July 25, 2007 - link
I thought one of the advantages of DX10 was that it would allow one to partition off some of the GPU subprocessors for physics work.I was *very* surprised that the author implied that the GPUs were not well suited to embarrassingly parallel applications.... um.... what's more embarrassingly parallel than rendering?
Ryan Smith - Wednesday, July 25, 2007 - link
I think you're misinterpreting what I'm saying. GPUs are well suited to embarrassingly parallel applications, however with the core-war now you can put these tasks on a CPU which while not as fast at FP as a GPU/PPU, is quickly catching up thanks to having multiple CPU cores and how easy it is to put embarrassingly parallel tasks on such a CPU. GPUs are still better suited, but CPUs are becoming well enough suited that the GPU advantage is being chipped away.As for DX10, there's nothing specifically in it for physics. Using SM4.0/geometry shaders you can do some second-order work, but first-order work looks like it will need to be done with CUDA/CTM which isn't a part of DX10. You may also be thinking of the long-rumored DirectPhysics API, which is just that: a rumor.
yyrkoon - Thursday, July 26, 2007 - link
Actualy, because of the limited bandwidth capabilities of any GPU interface, the CPU is far better suited. Sure a 16x PCIe interface is limited to a huge 40Gbit/s bandwidth (asyncronous), and as I said, this may *seem* huge, but I personally know many game devers who have maxed this limit easily when experimenting with game technologies. When, and if the PCIe bus expands to 32x, and *if* graphics OEMs / motherboard OEMs implement it, then we'll see something that resembles the CPU-> memory of current bandwidth capabilities(10GB/s). By then however, who is to say how much the CPU -> memory bandwidth will be capable of. Granted, having said all that, this is why *we* load compressed textures into video memory, and do the math on the GPU . . .Anyhow, the whole time reading this article, I could not help but think that with current CPUs being at 4 cores, and Andahls law, that the two *other* cores could be used for this purpose, and it makes total sense. I think it would behoove Aegia, and Havok both to forget about Physics hardware, and start working on a liscencable software solution.