Graphics processing unit 1

Graphics processing unit

GeForce 6600GT (NV43) GPU

A graphics processing unit (GPU), also occasionally called visualprocessing unit (VPU), is a specialized electronic circuit designed torapidly manipulate and alter memory to accelerate the creation ofimages in a frame buffer intended for output to a display. GPUs areused in embedded systems, mobile phones, personal computers,workstations, and game consoles. Modern GPUs are very efficient atmanipulating computer graphics, and their highly parallel structuremakes them more effective than general-purpose CPUs for algorithmswhere processing of large blocks of data is done in parallel. In apersonal computer, a GPU can be present on a video card, or it can beon the motherboard or—in certain CPUs—on the CPU die.

The term GPU was popularized by Nvidia in 1999, who marketed theGeForce 256 as "the world's first 'GPU', or Graphics Processing Unit, asingle-chip processor with integrated transform, lighting, triangle setup/clipping, and rendering engines that arecapable of processing a minimum of 10 million polygons per second". Rival ATI Technologies coined the termvisual processing unit or VPU with the release of the Radeon 9700 in 2002.

History

1980sIn 1983, Intel made the iSBX 275 Video Graphics Controller Multimodule Board, for industrial systems based on theMultibus standard.[1] The card was based on the 82720 Graphics Display Controller, and accelerated the drawing oflines, arcs, rectangles, and character bitmaps. The framebuffer was also accelerated through loading via DMA. Theboard was intended for use with Intel's line of Multibus industrial single-board computer plugin cards.Released in 1985, the Commodore Amiga was one of the first personal computers to come standard with a GPU. TheGPU supported line draw, area fill, and included a type of stream processor called a blitter which accelerated themovement, manipulation, and combination of multiple arbitrary bitmaps. Also included was a coprocessor with itsown (primitive) instruction set capable of directly invoking a sequence of graphics operations without CPUintervention. Prior to this and for quite some time after, many other personal computer systems instead used theirmain, general-purpose CPU to handle almost every aspect of drawing the display, short of generating the final videosignal.In 1986, Texas Instruments released the TMS34010, the first microprocessor with on-chip graphics capabilities. Itcould run general-purpose code, but it had a very graphics-oriented instruction set. In 1990-1991, this chip wouldbecome the basis of the Texas Instruments Graphics Architecture ("TIGA") Windows accelerator cards.In 1987, the IBM 8514 graphics system was released as one of the first video cards for IBM PC compatibles toimplement fixed-function 2D primitives in electronic hardware.

Graphics processing unit 2

1990s

Tseng Labs ET4000/W32p

S3 Graphics ViRGE

Voodoo3 2000 AGP card

In 1991, S3 Graphics introduced the S3 86C911, which its designersnamed after the Porsche 911 as an implication of the performanceincrease it promised. The 86C911 spawned a host of imitators: by1995, all major PC graphics chip makers had added 2D accelerationsupport to their chips. By this time, fixed-function Windowsaccelerators had surpassed expensive general-purpose graphicscoprocessors in Windows performance, and these coprocessors fadedaway from the PC market.

Throughout the 1990s, 2D GUI acceleration continued to evolve. Asmanufacturing capabilities improved, so did the level of integration ofgraphics chips. Additional application programming interfaces (APIs)arrived for a variety of tasks, such as Microsoft's WinG graphicslibrary for Windows 3.x, and their later DirectDraw interface forhardware acceleration of 2D games within Windows 95 and later.

In the early- and mid-1990s, CPU-assisted real-time 3D graphics werebecoming increasingly common in computer and console games, whichled to an increasing public demand for hardware-accelerated 3Dgraphics. Early examples of mass-marketed 3D graphics hardware canbe found in fifth generation video game consoles such as PlayStationand Nintendo 64. In the PC world, notable failed first tries for low-cost3D graphics chips were the S3 ViRGE, ATI Rage, and MatroxMystique. These chips were essentially previous-generation 2Daccelerators with 3D features bolted on. Many were evenpin-compatible with the earlier-generation chips for ease ofimplementation and minimal cost. Initially, performance 3D graphicswere possible only with discrete boards dedicated to accelerating 3Dfunctions (and lacking 2D GUI acceleration entirely) such as the 3dfxVoodoo. However, as manufacturing technology continued to progress,video, 2D GUI acceleration, and 3D functionality were all integrated into one chip. Rendition's Verite chipsets werethe first to do this well enough to be worthy of note.

OpenGL appeared in the early '90s as a professional graphics API, but originally suffered from performance issueswhich allowed the Glide API to step in and become a dominant force on the PC in the late '90s.[2] However, theseissues were quickly overcome and the Glide API fell by the wayside. Software implementations of OpenGL werecommon during this time, although the influence of OpenGL eventually led to widespread hardware support. Overtime, a parity emerged between features offered in hardware and those offered in OpenGL. DirectX became popularamong Windows game developers during the late 90s. Unlike OpenGL, Microsoft insisted on providing strictone-to-one support of hardware. The approach made DirectX less popular as a standalone graphics API initially,since many GPUs provided their own specific features, which existing OpenGL applications were already able tobenefit from, leaving DirectX often one generation behind. (See: Comparison of OpenGL and Direct3D.)

Over time, Microsoft began to work more closely with hardware developers, and started to target the releases of DirectX to coincide with those of the supporting graphics hardware. Direct3D 5.0 was the first version of the burgeoning API to gain widespread adoption in the gaming market, and it competed directly with many more-hardware-specific, often proprietary graphics libraries, while OpenGL maintained a strong following. Direct3D 7.0 introduced support for hardware-accelerated transform and lighting (T&L) for Direct3D, while OpenGL had this

Graphics processing unit 3

capability already exposed from its inception. 3D accelerator cards moved beyond being just simple rasterizers toadd another significant hardware stage to the 3D rendering pipeline. The Nvidia GeForce 256 (also known as NV10)was the first consumer-level card on the market with hardware-accelerated T&L, while professional 3D cards alreadyhad this capability. Hardware transform and lighting, both already existing features of OpenGL, came toconsumer-level hardware in the '90s and set the precedent for later pixel shader and vertex shader units which werefar more flexible and programmable.

2000 to 2005With the advent of the OpenGL API and similar functionality in DirectX, GPUs added programmable shading totheir capabilities. Each pixel could now be processed by a short program that could include additional image texturesas inputs, and each geometric vertex could likewise be processed by a short program before it was projected onto thescreen. Nvidia was first to produce a chip capable of programmable shading, the GeForce 3 (code named NV20). ByOctober 2002, with the introduction of the ATI Radeon 9700 (also known as R300), the world's first Direct3D 9.0accelerator, pixel and vertex shaders could implement looping and lengthy floating point math, and in general werequickly becoming as flexible as CPUs, and orders of magnitude faster for image-array operations. Pixel shading isoften used for things like bump mapping, which adds texture, to make an object look shiny, dull, rough, or evenround or extruded.

2006 to presentWith the introduction of the GeForce 8 series, which was produced by Nvidia, and then new generic streamprocessing unit GPUs became a more generalized computing device. Today, parallel GPUs have begun makingcomputational inroads against the CPU, and a subfield of research, dubbed GPU Computing or GPGPU for GeneralPurpose Computing on GPU, has found its way into fields as diverse as machine learning, oil exploration, scientificimage processing, linear algebra,[3] statistics,[4] 3D reconstruction and even stock options pricing determination.Nvidia's CUDA platform was the earliest widely adopted programming model for GPU computing. More recentlyOpenCL has become broadly supported. OpenCL is an open standard defined by the Khronos Group which allowsfor the development of code for both GPUs and CPUs with an emphasis on portability.[5] OpenCL solutions aresupported by Intel, AMD, Nvidia, and ARM, and according to a recent report by Evan's data OpenCL is the GPGPUdevelopment platform most widely used by developers in both the US and Asia Pacific.

GPU companies

GPU manufacturers market share

Many companies have produced GPUs under a number of brandnames. In 2009, Intel, Nvidia and AMD/ATI were the market shareleaders, with 49.4%, 27.8% and 20.6% market share respectively.However, those numbers include Intel's integrated graphics solutions asGPUs. Not counting those numbers, Nvidia and ATI control nearly100% of the market as of 2008.[6] In addition, S3 Graphics (owned byVIA Technologies) and Matrox produce GPUs.

Graphics processing unit 4

Computational functionsModern GPUs use most of their transistors to do calculations related to 3D computer graphics. They were initiallyused to accelerate the memory-intensive work of texture mapping and rendering polygons, later adding units toaccelerate geometric calculations such as the rotation and translation of vertices into different coordinate systems.Recent developments in GPUs include support for programmable shaders which can manipulate vertices and textureswith many of the same operations supported by CPUs, oversampling and interpolation techniques to reduce aliasing,and very high-precision color spaces. Because most of these computations involve matrix and vector operations,engineers and scientists have increasingly studied the use of GPUs for non-graphical calculations. An example ofGPUs being used non-graphically is the generation of Bitcoins, where the graphical processing unit is used to solvehash functions.In addition to the 3D hardware, today's GPUs include basic 2D acceleration and framebuffer capabilities (usuallywith a VGA compatibility mode). Newer cards like AMD/ATI HD5000-HD7000 even lack 2D acceleration; it has tobe emulated by 3D hardware.

GPU accelerated video decoding

The ATI HD5470 GPU (above) features UVD 2.1which enables it to decode AVC and VC-1 video

formats

Most GPUs made since 1995 support the YUV color space andhardware overlays, important for digital video playback, and manyGPUs made since 2000 also support MPEG primitives such as motioncompensation and iDCT. This process of hardware accelerated videodecoding, where portions of the video decoding process and videopost-processing are offloaded to the GPU hardware, is commonlyreferred to as "GPU accelerated video decoding", "GPU assisted videodecoding", "GPU hardware accelerated video decoding" or "GPUhardware assisted video decoding".

More recent graphics cards even decode high-definition video on thecard, offloading the central processing unit. The most common APIsfor GPU accelerated video decoding are DxVA for Microsoft Windowsoperating system and VDPAU, VAAPI, XvMC, and XvBA for Linux and UNIX-based operating systems. Allexcept XvMC are capable of decoding videos encoded with MPEG-1, MPEG-2, MPEG-4 ASP (MPEG-4 Part 2),MPEG-4 AVC (H.264 / DivX 6), VC-1, WMV3/WMV9, Xvid / OpenDivX (DivX 4), and DivX 5 codecs, whileXvMC is only capable of decoding MPEG-1 and MPEG-2.

Video decoding processes that can be accelerated

The video decoding processes that can be accelerated by today's modern GPU hardware are:•• Motion compensation (mocomp)•• Inverse discrete cosine transform (iDCT)

• Inverse telecine 3:2 and 2:2 pull-down correction• Inverse modified discrete cosine transform (iMDCT)• In-loop deblocking filter•• Intra-frame prediction• Inverse quantization (IQ)• Variable-length decoding (VLD), more commonly known as slice-level acceleration• Spatial-temporal deinterlacing and automatic interlace/progressive source detection• Bitstream processing (Context-adaptive variable-length coding/Context-adaptive binary arithmetic coding) and

perfect pixel positioning.

Graphics processing unit 5

GPU forms

Dedicated graphics cardsThe GPUs of the most powerful class typically interface with the motherboard by means of an expansion slot such asPCI Express (PCIe) or Accelerated Graphics Port (AGP) and can usually be replaced or upgraded with relative ease,assuming the motherboard is capable of supporting the upgrade. A few graphics cards still use Peripheral ComponentInterconnect (PCI) slots, but their bandwidth is so limited that they are generally used only when a PCIe or AGP slotis not available.A dedicated GPU is not necessarily removable, nor does it necessarily interface with the motherboard in a standardfashion. The term "dedicated" refers to the fact that dedicated graphics cards have RAM that is dedicated to thecard's use, not to the fact that most dedicated GPUs are removable. Dedicated GPUs for portable computers are mostcommonly interfaced through a non-standard and often proprietary slot due to size and weight constraints. Such portsmay still be considered PCIe or AGP in terms of their logical host interface, even if they are not physicallyinterchangeable with their counterparts.Technologies such as SLI by Nvidia and CrossFire by ATI allow multiple GPUs to be used to draw a single image,increasing the processing power available for graphics.

Integrated graphics solutions

Layout

A motherboard withintegrated graphics, whichhas HDMI, VGA and DVI

outs.

Integrated graphics solutions, shared graphics solutions, or Integratedgraphics processors (IGP) utilize a portion of a computer's system RAMrather than dedicated graphics memory. Most are integrated into themotherboard, though exceptions include AMD's IGPs that use dedicatedsideport memory on certain motherboards, and APUs, where they areintegrated with the CPU die. Computers with integrated graphics account for90% of all PCshipments.[7]Wikipedia:Manual_of_Style_(dates_and_numbers)#Chronological_itemsThese solutions are less costly to implement than dedicated graphicssolutions, but tend to be less capable. Historically, integrated solutions wereoften considered unfit to play 3D games or run graphically intensiveprograms but could run less intensive programs such as Adobe Flash.Examples of such IGPs would be offerings from SiS and VIA circa 2004.However, modern integrated graphics processors such as AMD AcceleratedProcessing Unit and Intel HD Graphics are more than capable of handling 2Dgraphics or low stress 3D graphics, having improved performance that theycan match or exceed that of older dedicated graphic cards, although they arestill far less capable than current generation dedicated GPUs.[citation needed]

As a GPU is extremely memory intensive, an integrated solution may finditself competing for the already relatively slow system RAM with the CPU, asit has minimal or no dedicated video memory. IGPs can have up to 29.856GB/s of memory bandwidth from system RAM, however graphics cards canenjoy up to 264 GB/s of bandwidth over its memory-bus. Wikipedia:Pleaseclarify Older integrated graphics chipsets lacked hardware transform andlighting, but newer ones include it. Wikipedia:Please clarify

Graphics processing unit 6

Hybrid solutionsThis newer class of GPUs competes with integrated graphics in the low-end desktop and notebook markets. Themost common implementations of this are ATI's HyperMemory and Nvidia's TurboCache. Hybrid graphics cards aresomewhat more expensive than integrated graphics, but much less expensive than dedicated graphics cards. Theseshare memory with the system and have a small dedicated memory cache, to make up for the high latency of thesystem RAM. Technologies within PCI Express can make this possible. While these solutions are sometimesadvertised as having as much as 768MB of RAM, this refers to how much can be shared with the system memory.

Stream Processing and General Purpose GPUs (GPGPU)It is becoming increasingly common to use a general purpose graphics processing unit as a modified form of streamprocessor. This concept turns the massive computational power of a modern graphics accelerator's shader pipelineinto general-purpose computing power, as opposed to being hard wired solely to do graphical operations. In certainapplications requiring massive vector operations, this can yield several orders of magnitude higher performance thana conventional CPU. The two largest discrete (see "Dedicated graphics cards" above) GPU designers, ATI andNvidia, are beginning to pursue this approach with an array of applications. Both Nvidia and ATI have teamed withStanford University to create a GPU-based client for the Folding@home distributed computing project, for proteinfolding calculations. In certain circumstances the GPU calculates forty times faster than the conventional CPUstraditionally used by such applications.GPGPU can be used for many types of embarrassingly parallel tasks including ray tracing. They are generally suitedto high-throughput type computations that exhibit data-parallelism to exploit the wide vector width SIMDarchitecture of the GPU.Furthermore, GPU-based high performance computers are starting to play a significant role in large-scale modelling.Three of the 10 most powerful supercomputers in the world take advantage of GPU acceleration.NVIDIA cards support API extensions to the C programming language such as CUDA ("Compute Unified DeviceArchitecture") and OpenCL. CUDA is specifically for NVIDIA GPUs whilst OpenCL is designed to work across amultitude of architectures including GPU, CPU and DSP (using vendor specific SDKs). These technologies allowspecified functions (kernels) from a normal C program to run on the GPU's stream processors. This makes Cprograms capable of taking advantage of a GPU's ability to operate on large matrices in parallel, while still makinguse of the CPU when appropriate. CUDA is also the first API to allow CPU-based applications to directly access theresources of a GPU for more general purpose computing without the limitations of using a graphics API.Since 2005 there has been interest in using the performance offered by GPUs for evolutionary computation ingeneral, and for accelerating the fitness evaluation in genetic programming in particular. Most approaches compilelinear or tree programs on the host PC and transfer the executable to the GPU to be run. Typically the performanceadvantage is only obtained by running the single active program simultaneously on many example problems inparallel, using the GPU's SIMD architecture. However, substantial acceleration can also be obtained by notcompiling the programs, and instead transferring them to the GPU, to be interpreted there.[8] Acceleration can thenbe obtained by either interpreting multiple programs simultaneously, simultaneously running multiple exampleproblems, or combinations of both. A modern GPU (e.g. 8800 GTX or later) can readily simultaneously interprethundreds of thousands of very small programs.

Graphics processing unit 7

References[1] Michael Swaine, "New Chip from Intel Gives High-Quality Displays", March 14, 1983, p.16[2][2] 3dfx Glide API[3][3] "Linear algebra operators for GPU implementation of numerical algorithms", Kruger and Westermann, International Conf. on Computer

Graphics and Interactive Techniques, 2005[4] "ABC-SysBio—approximate Bayesian computation in Python with GPU support", Liepe et al., Bioinformatics, (2010), 26:1797-1799 (http:/ /

bioinformatics. oxfordjournals. org/ content/ 26/ 14/ 1797. full)[5] Khronos Group (http:/ / www. khronos. org/ opencl/ )[6] Q3 Sales Report from Jon Peddie Research via TechReport.com (http:/ / techreport. com/ discussions. x/ 15778)[7] AnandTech: µATX Part 2: Intel G33 Performance Review (http:/ / www. anandtech. com/ mb/ showdoc. aspx?i=3111& p=23)[8][8] V. Garcia and E. Debreuve and M. Barlaud. Fast k nearest neighbor search using GPU. In Proceedings of the CVPR Workshop on Computer

Vision on GPU, Anchorage, Alaska, USA, June 2008.

External links• NVIDIA - What is GPU computing? (http:/ / www. nvidia. com/ object/ what-is-gpu-computing. html)• The GPU Gems book series (https:/ / developer. nvidia. com/ content/ gpu-gems)• - a Graphics Hardware History (http:/ / titancity. com/ articles/ gfxcards. html)• General-Purpose Computation Using Graphics Hardware (http:/ / www. gpgpu. org/ )• How GPUs work (http:/ / www. cs. virginia. edu/ ~gfx/ papers/ paper. php?paper_id=59)• GPU Caps Viewer - Video card information utility (http:/ / www. ozone3d. net/ gpu_caps_viewer/ )• OpenGPU-GPU Architecture(In Chinese) (http:/ / www. opengpu. org/ bbs/ index. php)• ARM Mali GPUs Overview (http:/ / malideveloper. arm. com/ learn-about-mali/ about-mali/ arm-mali-gpus/ )

Article Sources and Contributors 8

Article Sources and ContributorsGraphics processing unit  Source: http://en.wikipedia.org/w/index.php?oldid=599612122  Contributors: 4th-otaku, 4wajzkd02, 5900FX, =Josh.Harris, AFBorchert, ALoopingIcon, Aayush.nitb,Academic Challenger, Accessory, Ae-a, Aeons, Agelu, Akkida, Alan Liefting, Alanau8605, Ale And Quail, Aleenf1, AlexKepler, Alf Boggis, Alison, Alistair1978, AlistairMcMillan, Anabus,Andrei.gheorghe, Andres, AndriusG, Andy16666, Angelic Wraith, Arch dude, Arnero, Arthur Rubin, Aughtandzero, Aunva6, Beland, Ben Ben, Berkut, Bevo74, Bigbluefish, Bitsmart, Bjorke,Braddodson, Braxtonw1, Calinou1, Can't sleep, clown will eat me, Canadianbob, CanisRufus, Chealer, ChrisfromHouston, Clark89, Codename Lisa, Codepro, Cody-7, Cogiati, Colinstu,Comp.arch, CoolingGibbon, Crazyideas21, CrystalCanine, Cxk271, Cybercobra, D'Agosta, DARTH SIDIOUS 2, DJSupreme23, Danrok, David Biddulph, David Eppstein, Dcirovic, Delirium ofdisorder, Des3dhj, Dholcombe, Diculous, Disavian, Djayjp, DopefishJustin, Dpmuk, Drhex, Dsimic, Dymatic, E Wing, Earthere, Easwarno1, Eberhart, EconomistBR, Edward, EdwardJK, Egil,Egomaniac, El Krem, Eleven even, Erud, Evan-Amos, Everyking, Evice, Eyesnore, FAMAS, FearTec, Fiftyquid, Flex Flint, Frap, Frood, Furrykef, Gadfium, Galoubet, Garde, GateKeeper,Gbrose85, GeoffMacartney, Gilliam, Gillwill, Gioto, GlasGhost, Glitchrf, Gogo Dodo, GoldDragon, Gracefool, Gronky, Gzkn, Hallowin, Handheldpenguin, HarisM, Harumphy, Harvester,Haseo9999, HeroTsai, Heron, Hexmaster, Hitachi-Train, Holden15, Hu12, Imroy, Isenherz, Isnow, IvarTJ, J.delanoy, J04n, JDP90, JJC1138, Jafet, Jagged 85, JanEnEm, Jannex, Jappalang,Jarble, Jdevesa, Jdog147123, JeGX, Jerome Charles Potts, Jesant13, Jesse V., Jesse Viviano, Jmankovecky, Jo7hs2, Joffeloff, Johan.Seland, John Buchan, Jpvinall, Jrockley, Jsmaye, Jtbandes,Kaf, Kangaroopower, Karada, Kbdank71, Kelly Martin, Khalid hassani, Kjkolb, Kremerica, Largecrashman, Lazulilasher, Lee Cremeans, Lightmouse, Lights, LilHelpa, Locke411,Lockeownzj00, Lurker, M412k, MER-C, MFNickster, Macronyx, Mahbubur-r-aaman, Mahjongg, Marasmusine, MarcusBritish, Markpapadakis, Marmoulak, Martial75, Masterofwiki666,Math1337, Matt Britt, Mattdj, Matthiaspaul, MaxDZ8, Mc6809e, Melter, Mentifisto, Michaelothomas, Mike92591, Mikkow, Mirror Vax, Mr. XYZ, MrOllie, MrRK, Murphykieran, Nakon,Nczempin, Nekura, Niceguyedc, Nigma2k, Nikevich, Nitro912gr, Nixdorf, Nnemo, Noctibus, NonNobisSolum, Nonnompow, Nopetro, OSXiOSMacFan, Octahedron80, Optim, Osarius,Oxymoron83, P99am, Panscient, Papagao, Paul Pogonyshev, Pavel Vozenilek, PavelSolin, Pgan002, Pharaoh of the Wizards, Phatom87, Phranq, Pi Guy 31415, Pinkadelica, Pixel8, Placi1982,Playwrite, Polluks, Publicly Visible, QTCaptain, Quadell, Qviri, R'n'B, RBBrittain, Ravenperch, Rchandra, Reatlas, Renku, RicReis, Richi, Rilak, Romdanen, Ronhjones, RoyBoy, Rreagan007,RubyQ, Ruw1090, Rzęsor, SCEhardt, SEG88, Sahrin, Salam32, Salamurai, Sango123, Scissorhands1203, Sciurinæ, ScotXW, Scott Paeth, ScottHolden, Sdornan, Serketan, Serpent's Choice,Sg227, ShadowHntr, Shandris, Sibekoe, Simoneau, Sjf, Skarkkai, SkyWalker, Socks 01, Socrates2008, Soumya92, Spiesr, Squalk25, Stargaming, Stoakron97, Stocbuster, Stormie, StuartBrady,Sugarbat, Suruena, Swaaye, TOR, Tagremover, Tandral, Taw, Tbird20d, Technobadger, Tempshill, The Thing That Should Not Be, Theonlyedge, Thue, Thumperward, Tigeron, Tijok,TimBentley, Tobias Bergemann, Tomy9510, Topbanana, Topeil, Toussaint, Tr-the-maniac, Trieste, Trilobite, Trusader, Twsl, Umawera, UnfriendlyFire, Unyoyega, Utcursch, Vapourmile,Varuna, Veikk0.ma, Veinor, Veritysense, Vespristiano, Victorbabkov, Vidsi, Vincentfpgarcia, Virek, Vitkovskiy Roman, Viznut, Vuurmeester, Wayne Hardman, Wbm1058, Webclient101,Wernher, Wibbble, Widr, WikipedianMarlith, Willking1979, Winstonliang, Wknight94, Woohookitty, Xbspiro, Xcentaur, Xowets, Yogi m, Yulu, Yyy, Zarex, Zephalis, Zodon, 605 anonymousedits

Image Sources, Licenses and ContributorsImage:6600GT GPU.jpg  Source: http://en.wikipedia.org/w/index.php?title=File:6600GT_GPU.jpg  License: GNU Free Documentation License  Contributors: User:BerkutFile:Dstealth32.jpg  Source: http://en.wikipedia.org/w/index.php?title=File:Dstealth32.jpg  License: Public domain  Contributors: Magog the Ogre, باسم

File:DIAMONDSTEALTH3D2000-top.JPG  Source: http://en.wikipedia.org/w/index.php?title=File:DIAMONDSTEALTH3D2000-top.JPG  License: Public Domain  Contributors: SwaayeFile:Voodoo3-2000AGP.jpg  Source: http://en.wikipedia.org/w/index.php?title=File:Voodoo3-2000AGP.jpg  License: GNU Free Documentation License  Contributors: Original uploader wasSwaaye at en.wikipediaFile:Marktanteil GPU-Hersteller.png  Source: http://en.wikipedia.org/w/index.php?title=File:Marktanteil_GPU-Hersteller.png  License: Creative Commons Attribution-Sharealike 3.0 Contributors: User:IsenherzImage:AMD HD5470 GPU.JPG  Source: http://en.wikipedia.org/w/index.php?title=File:AMD_HD5470_GPU.JPG  License: Creative Commons Zero  Contributors: highwycombe (talk)File:Motherboard diagram.svg  Source: http://en.wikipedia.org/w/index.php?title=File:Motherboard_diagram.svg  License: GNU Free Documentation License  Contributors: user:Moxfyre.Original uploader was Moxfyre at en.wikipediaFile:A790GXH-128M-Motherboard.jpg  Source: http://en.wikipedia.org/w/index.php?title=File:A790GXH-128M-Motherboard.jpg  License: Creative Commons Attribution-Sharealike 3.0 Contributors: Evan-Amos

LicenseCreative Commons Attribution-Share Alike 3.0//creativecommons.org/licenses/by-sa/3.0/