 | This articleneeds additional citations forverification. Please helpimprove this article byadding citations to reliable sources. Unsourced material may be challenged and removed. Find sources: "Physics processing unit" – news ·newspapers ·books ·scholar ·JSTOR(March 2015) (Learn how and when to remove this message) |
Aphysics processing unit (PPU) is a dedicatedmicroprocessor designed to handle the calculations ofphysics, especially in thephysics engine ofvideo games. It is an example ofhardware acceleration.
Examples of calculations involving a PPU might includerigid body dynamics,soft body dynamics,collision detection,fluid dynamics, hair andclothing simulation,finite element analysis, and fracturing of objects.
The idea is having specialized processors offload time-consuming tasks from a computer's CPU, much like how aGPU performs graphics operations in the main CPU's place. The term was coined byAgeia to describe itsPhysX chip. Several other technologies in the CPU-GPU spectrum have some features in common with it, although Ageia's product was the only complete one designed, marketed, supported, and placed within a system exclusively being a PPU.
An early academic PPU research project[1][2] named SPARTA (Simulation of Physics on A Real-Time Architecture) was carried out at Penn State[3] and University of Georgia. This was a simpleFPGA based PPU that was limited to two dimensions. This project was extended into a considerably more advancedASIC-based system named HELLAS.
February 2006 saw the release of the first dedicated PPUPhysX fromAgeia (later merged intoNvidia). The unit is most effective in acceleratingparticle systems, with only a small performance improvement measured for rigid body physics.[4] The Ageia PPU is documented in depth in their US patent application #20050075849.[5] Nvidia/Ageia no longer produces PPUs and hardware acceleration for physics processing, although it is now supported through some of their graphics processing units.
The first processor to be advertised being a PPU was named thePhysX chip, introduced by afabless semiconductor company calledAGEIA. Games wishing to take advantage of the PhysX PPU must use AGEIA'sPhysXSDK, (formerly known as the NovodeX SDK).
It consists of a general purpose RISC core controlling an array of customSIMD floating pointVLIW processors working in local banked memories, with a switch-fabric to manage transfers between them. There is nocache-hierarchy like in a CPU or GPU.
The PhysX was available from three companies akin to the wayvideo cards are manufactured.ASUS,BFG Technologies,[6] andELSA Technologies were the primary manufacturers. PCs with the cards already installed were available from system builders such asAlienware,Dell, andFalcon Northwest.[7]
In February 2008, afterNvidia bought Ageia Technologies and eventually cut off the ability to process PhysX on the AGEIA PPU and NVIDIA GPUs in systems with active ATi/AMD GPUs, it seemed that PhysX went 100% to Nvidia. But in March 2008, Nvidia announced that it will make PhysX an open standard for everyone,[8] so the main graphic-processor manufacturers will have PhysX support in the next generation graphics cards. Nvidia announced that PhysX will also be available for some of their released graphics cards just by downloading some new drivers.
Seephysics engine for a discussion of academic research PPU projects.
ASUS andBFG Technologies bought licenses to manufacture alternate versions of AGEIA's PPU, the PhysX P1 with 128 MB GDDR3:
TheHavok SDK is a major competitor to the PhysX SDK, used in more than 150 games, including major titles likeHalf-Life 2,Halo 3 andDead Rising.[12]
To compete with the PhysX PPU, an edition known asHavok FX was to take advantage of multi-GPU technology fromATI (AMD CrossFire) andNVIDIA (SLI) using existing cards to accelerate certain physics calculations.[13]
Havok divides the physics simulation intoeffect andgameplay physics, with effect physics being offloaded (if possible) to the GPU asShader Model 3.0 instructions and gameplay physics being processed on the CPU as normal. The important distinction between the two is thateffect physics do not affect gameplay (dust or small debris from an explosion, for example); the vast majority of physics operations are still performed in software. This approach differs significantly from the PhysX SDK, which moves all calculations to the PhysX card if it is present.
Since Havok's acquisition byIntel, Havok FX appears to have been shelved or cancelled.[14]
The drive towardGPGPU has made GPUs more suitable for the job of a PPU; DX10 added integer data types, unified shader architecture, and a geometry shader stage which allows a broader range of algorithms to be implemented; Modern GPUs supportcompute shaders, which run across an indexed space and don't require any graphical resources, just general purpose data buffers. NVidiaCUDA provides a little more in the way of inter-thread communication andscratchpad-style workspace associated with the threads.
Nonetheless GPUs are built around a larger number of longer latency, slower threads, and designed around texture and framebuffer data paths, and poor branching performance; this distinguishes them from PPUs andCell as being less well optimized for taking over game world simulation tasks.
TheCodeplay Sieve compiler supports the PPU, indicating that the Ageia physX chip would be suitable for GPGPU type tasks. However Ageia seem unlikely to pursue this market.
Although very different from the PhysX, one could argue thePlayStation 2'sVU0 is an early, limited implementation of a PPU. Conversely, one could describe a PPU to a PS2 programmer as an evolved replacement for VU0. Its feature-set and placement within the system is geared toward accelerating game update tasks including physics and AI; it can offload such calculations working off its own instruction stream whilst the CPU is operating on something else. Being a DSP however, it is much more dependent on the CPU to do useful work in a game engine, and would not be capable of implementing a full physics API, so it cannot be classed as a PPU. Also VU0 is capable of providing additional vertex processing power, though this is more a property of the pathways in the system rather than the unit itself.
This usage is similar to Havok FX or GPU physics in that an auxiliary unit's general purpose floating point power is used to complement the CPU in either graphics or physics roles.