The evolution of the fashionable graphics processor started with the introduction of the first 3D add-in playing cards in 1995. This growth coincided with the widespread adoption of 32-bit working methods and the affordability of private computer systems. The graphics business that existed earlier than the PC was largely dominated by extra prosaic 2D structure. Graphics boards have been recognized for his or her chip’s alphanumeric naming conventions and large worth tags. 3D gaming and PC virtualization graphics emerged from a spread of sources, together with arcade and console gaming, army purposes, robotics, house simulators, and medical imaging.
The early days of 3D shopper graphics have been marked by a “Wild West” of competing concepts, starting from {hardware} implementation to numerous rendering methods, software and knowledge interfaces, and chronic naming hyperbole. These early graphics methods featured a set perform pipeline (FFP) and an structure that adhered to a really inflexible processing path. There have been nearly as many graphics APIs as there have been 3D chip makers.
3D graphics reworked the considerably uninteresting PC business right into a spectacle of gentle and magic after generations of progressive endeavor. This article is the first installment of a TechSpot Special collection that extensively explores the historical past of the GPU. We revisit the early days of 3D shopper graphics, highlighting the game-changing affect of 3Dfx, and the business’s consolidation at the flip of the century. The collection concludes (for now) with a take a look at in the present day’s fashionable general-purpose GPUs, which we’ve got come to like and use for gaming. These GPUs are actually reworking the whole business.
TechSpot’s History of the GPU
The fashionable graphics processor has change into one of the largest, most complicated, and costliest elements present in nearly any computing system. From the early VGA days to the fashionable GPU. The historical past and evolution of the chip that got here to dominate gaming, and later AI and compute.
1976 – 1995
1995 – 1999
2000 – 2006
2006 – 2013
2013 – 2020
1976 – 1995:
The Early Days of 3D Consumer Graphics
The first true 3D graphics originated with early show controllers, often called video shifters and video handle mills. These gadgets acted as a pass-through between the important processor and the show. They transformed the incoming knowledge stream right into a serial bitmapped video output, which included luminance, coloration, and vertical and horizontal composite sync. This synchronization was essential for sustaining the alignment of pixels in a show technology, guaranteeing the orderly development of every successive line, and managing the blanking interval (the time between ending one scan line and beginning the subsequent).
A flurry of designs emerged in the latter half of the Seventies, laying the basis for 3D graphics as we all know them. One notable instance was RCA’s “Pixie” video chip (CDP1861), launched in 1976. It was succesful of outputting an NTSC-compatible video sign at a decision of 62×128, or 64×32 for the short-lived RCA Studio II console.
The video chip was quickly succeeded by the Television Interface Adapter (TIA) 1A in 1977. This chip, integral to the Atari 2600, was liable for producing the display show, creating sound results, and studying enter from controllers. Jay Miner, who later spearheaded the design of the customized chips for the Commodore Amiga pc, led the growth of the TIA.
In 1978, Motorola unveiled the MC6845 video handle generator. This system laid the basis for the Monochrome and Color Display Adapter (MDA/CDA) playing cards utilized in the IBM PC in 1981 and in addition supplied related performance for the Apple II. Later that yr, Motorola launched the MC6847 video show generator, which was integrated into a number of first-generation private computer systems, together with the Tandy TRS-80.
An analogous resolution got here from Commodore’s MOS Technology subsidiary with the VIC (Video Interface Chip), which delivered graphics output for Commodore house computer systems produced between 1980 and 1983.
In November of the following yr, LSI launched the ANTIC (Alphanumeric Television Interface Controller) and the CTIA/GTIA co-processor (Color or Graphics Television Interface Adapter) in the Atari 400. The ANTIC processed 2D show directions utilizing Direct Memory Access (DMA).
Like many video co-processors of the time, it was succesful of producing playfield graphics (comparable to background, title screens, and scoring shows), whereas the CTIA was liable for producing colours and movable objects. Yamaha and Texas Instruments supplied related built-in circuits (ICs) to a variety of early house pc producers.
The subsequent vital steps in the graphics evolution occurred primarily in skilled fields. Intel’s 82720 graphics chip served as the basis for the $1,000 iSBX 275 Video Graphics Controller Multimode Board. This board was succesful of displaying eight-color knowledge at a decision of 256×256 or monochrome at 512×512.
Its 32KB of show reminiscence enabled it to attract traces, arcs, circles, rectangles, and character bitmaps. The chip additionally featured capabilities for zooming, display partitioning, and scrolling.
Silicon Graphics (SGI) quickly adopted with their IRIS Graphics for workstations. This was a GR1.x graphics board that allowed for the addition of separate daughter boards for enhanced coloration choices, geometry processing, Z-buffering, and Overlay/Underlay functionalities.
Intel’s $1,000 iSBX 275 Video Graphics Controller Multimode Board was succesful of displaying eight coloration knowledge at a decision of 256×256 (or monochrome at 512×512).
Industrial and army 3D virtualization applied sciences have been comparatively superior at the moment. Companies like IBM, General Electric, and Martin Marietta (which might later purchase GE’s aerospace division in 1992), together with a large number of army contractors, expertise institutes, and NASA, undertook numerous tasks that required this expertise for army and house simulations.
The U.S. Navy developed a flight simulator in 1951 utilizing 3D virtualization expertise from MIT’s Whirlwind pc. In addition to protection contractors, there have been corporations that bridged the hole between army markets {and professional} graphics sectors.
Evans & Sutherland, who would later present skilled graphics card collection comparable to the Freedom and REALimage, additionally developed graphics for the CT5 flight simulator. This $20 million package deal was powered by a DEC PDP-11 mainframe. Ivan Sutherland, the firm’s co-founder, created a pc program in 1961 referred to as Sketchpad, which allowed the drawing geometric shapes and displaying on a CRT in real-time utilizing a light-weight pen. This was the progenitor of the fashionable Graphical User Interface (GUI).
In the much less esoteric subject of private computing, Chips and Technologies launched the 82C43x collection of EGA (Extended Graphics Adapter). These adapters supplied much-needed competitors to IBM’s choices and have been generally discovered in lots of PC/AT clones round 1985. That yr was additionally noteworthy for the Commodore Amiga, which was launched with the OCS chipset. This chipset consisted of three important part chips – Agnus, Denise, and Paula – permitting some graphics and audio processing to be unbiased of the CPU.
In August of 1985, three Hong Kong immigrants, Kwok Yuan Ho, Lee Lau, and Benny Lau, established Array Technology Inc in Canada. By the finish of the yr, the firm had been renamed ATI Technologies Inc.
ATI launched their first product the following yr: the OEM Color Emulation Card. It was designed for outputting monochrome inexperienced, amber, or white phosphor textual content in opposition to a black background to a TTL monitor by way of a 9-pin DE-9 connector. The card was geared up with a minimal of 16KB of reminiscence and contributed considerably to ATI’s CAD$10 million in gross sales throughout the firm’s first yr, largely due to a contract to provide round 7,000 chips per week to Commodore Computers.
ATI’s Color Emulation Card got here with a minimal 16KB of reminiscence and was liable for a big half of the firm’s CAD$10 million in gross sales the first yr of operation.
The introduction of coloration displays and the absence of a regular amongst the numerous rivals led to the formation of the Video Electronics Standards Association (VESA), of which ATI was a founding member alongside NEC and 6 different graphics adapter producers.
In 1987, ATI expanded its product line for OEMs with the Graphics Solution Plus collection. This collection utilized IBM’s PC/XT ISA 8-bit bus for Intel 8086/8088-based IBM PCs. The chip supported MDA, CGA, and EGA graphics modes by dip switches. Essentially, it was a clone of the Plantronics Colorplus board however with capability for 64KB of reminiscence. Paradise Systems’ PEGA1, 1a, and 2a (256KB), launched in 1987, have been additionally clones of Plantronics boards.
The EGA Wonder collection 1 to 4, launched in March for $399, featured 256KB of DRAM and was appropriate with CGA, EGA, and MDA emulation, supporting resolutions as much as 640×350 with 16 colours. Extended EGA was out there for collection 2, 3, and 4.
The high-end section included the EGA Wonder 800, which supplied 16-color VGA emulation and supported a decision of 800×600, and the VGA Improved Performance (VIP) card. The VIP card was basically an EGA Wonder with a digital-to-analog converter (DAC) added to offer restricted VGA compatibility. The VIP card was priced at $449, with a further $99 for the Compaq enlargement module.
ATI was not alone in capitalizing on the rising shopper curiosity in private computing. Numerous corporations and merchandise emerged throughout that interval. New entrants included Trident, SiS, Tamerack, Realtek, Oak Technology, LSI’s G-2 Inc., Hualon, Cornerstone Imaging, and Winbond – all established between 1986 and 1987. Meanwhile, corporations like AMD, Western Digital/Paradise Systems, Intergraph, Cirrus Logic, Texas Instruments, Gemini, and Genoa started producing their first graphics merchandise round this time.
ATI’s Wonder collection continued to obtain vital updates over the following years. In 1988, the Small Wonder Graphics Solution, that includes a recreation controller port and composite output choices (for CGA and MDA emulation), was launched, together with the EGA Wonder 480 and 800+ with Extended EGA and 16-bit VGA help. The VGA Wonder and Wonder 16, with added VGA and SVGA help, have been additionally launched.
A Wonder 16 geared up with 256KB of reminiscence retailed for $499, whereas a variant with 512KB was priced at $699. In 1989, the VGA Wonder/Wonder 16 collection was up to date, together with the lower-cost VGA Edge 16 (Wonder 1024 collection). New options included a bus mouse port and help for the VESA Feature Connector, a gold-fingered connector much like a shortened knowledge bus slot connector, used to hyperlink by way of a ribbon cable to a different video controller to bypass a congested knowledge bus.
The Wonder collection continued to maneuver apace in 1991. The Wonder XL card added VESA 32K coloration compatibility and a Sierra RAMDAC, growing the most show decision to 640×480 @ 72Hz or 800×600 @ 60Hz. Prices diversified from $249 (256KB) to $349 (512KB), with a 1MB RAM possibility out there for $399. A lower-cost model, the VGA Charger, based mostly on the earlier yr’s Basic-16, was additionally launched.
The Mach collection was launched with the Mach8 in May of that yr. It was offered both as a chip or a board, permitting for the offloading of restricted 2D drawing operations, comparable to line-drawing, color-filling, and bitmap mixture (BitBLT), by way of a programming interface (API).
ATI additionally launched a variation of the Wonder XL that integrated a Creative Sound Blaster 1.5 chip on an prolonged PCB. Known as the VGA Stereo-F/X, it was succesful of simulating stereo sound from Sound Blaster mono recordsdata at a top quality corresponding to FM radio.
Graphics boards like the ATI VGAWonder GT supplied a 2D + 3D possibility, combining the Mach8 with the graphics core (28800-2) of the VGA Wonder+ for 3D duties. The success of the Wonder and Mach8 collection helped ATI surpass the CAD$100 million gross sales milestone for the yr, largely pushed by the adoption of Windows 3.0 and the elevated 2D workloads it facilitated.
S3 Graphics was shaped in early 1989 and produced its first 2D accelerator chip and a graphics card 18 months later, the S3 911 (or 86C911). Key specs for the latter included 1MB of VRAM and 16-bit coloration help.
The S3 911 was shortly succeeded by the 924 later that yr – basically a revised 911 with 24-bit coloration help – and was additional up to date the following yr with the 928, which added 32-bit coloration help, and the 801 and 805 accelerators. The 801 utilized an ISA interface, whereas the 805 used VLB (VESA Local Bus). From the introduction of the 911 to the emergence of the 3D accelerator, the market was inundated with 2D GUI designs based mostly on S3’s authentic structure, notably from Tseng Labs, Cirrus Logic, Trident, IIT, ATI’s Mach32, and Matrox’s MAGIC RGB.
In January 1992, Silicon Graphics (SGI) launched OpenGL 1.0, a multi-platform, vendor-agnostic software programming interface (API) for each 2D and 3D graphics.
Microsoft was growing a rival API of their very own referred to as Direct3D and did not precisely break a sweat ensuring OpenGL ran in addition to it may below Windows.
OpenGL advanced from SGI’s proprietary API, referred to as IRIS GL (Integrated Raster Imaging System Graphical Library). This growth was an initiative to separate non-graphical performance from IRIS and allow the API to run on non-SGI methods, particularly as rival distributors started rising on the horizon with their proprietary APIs. Initially, OpenGL focused the skilled UNIX-based markets. However, as a consequence of its developer-friendly help for extension implementation, it was shortly adopted for 3D gaming.
Meanwhile, Microsoft was growing its personal rival API, Direct3D, and did not precisely break a sweat ensuring OpenGL ran in addition to it may below the new Windows working methods. Tensions peaked a couple of years later when John Carmack of id Software, whose launch of Doom had revolutionized PC gaming, ported Quake to make use of OpenGL on Windows. He brazenly criticized Direct3D, highlighting the rising competitors between these two graphics APIs.
Microsoft’s intransigence intensified after they refused to license the Mini-Client Driver (MCD) for OpenGL on Windows 95. The MCD would have allowed distributors to pick particular options for {hardware} acceleration. In response, SGI developed the Installable Client Driver (ICD), which not solely supplied the identical capabilities but additionally improved upon them. Unlike the MCD, which solely coated rasterization, the ICD added lighting and rework performance (T&L).
During the rise of OpenGL, which first gained prominence in the workstation market, Microsoft was specializing in the burgeoning gaming market with plans for its proprietary API. In February 1995, they acquired RenderMorphics, whose Reality Lab API was gaining recognition amongst builders, and this turned the basis for Direct3D.
Around the identical time, 3Dfx’s Brian Hook was growing the Glide API, which might change into the dominant gaming API. This was partly as a consequence of Microsoft’s involvement in the Talisman mission (a tile-based rendering ecosystem), which diverted assets from DirectX.
As Direct3D turned extensively out there, bolstered by the widespread adoption of Windows, proprietary APIs like S3d (S3), Matrox Simple Interface, Creative Graphics Library, C Interface (ATI), SGL (PowerVR), NVLIB (Nvidia), RRedline (Rendition), and Glide began dropping favor with builders.
This decline was exacerbated by the incontrovertible fact that some of these proprietary APIs have been tied to board producers who have been below growing stress to reinforce their quickly increasing characteristic lists. These enhancements included larger display resolutions, better coloration depth (progressing from 16-bit to 24 after which 32 bits), and picture high quality enhancements comparable to anti-aliasing. All these options demanded elevated bandwidth, better graphics effectivity, and quicker product growth cycles.
By 1993, market volatility had already pressured a quantity of graphics corporations to withdraw from the enterprise, or to be absorbed by rivals.
The yr 1993 ushered in a flurry of new graphics rivals, most notably Nvidia, based in January of that yr by Jen-Hsun Huang, Curtis Priem, and Chris Malachowsky. Huang was beforehand the Director of Coreware at LSI, whereas Priem and Malachowsky each hailed from Sun Microsystems, the place they’d developed the SunSPARC-based GX graphics structure.
Soon becoming a member of the fray have been fellow newcomers Dynamic Pictures, ARK Logic, and Rendition. The market’s volatility had already led to the exit of a number of graphics corporations or their absorption by rivals. Among these have been Tamerack, Gemini Technology, Genoa Systems, Hualon, Headland Technology (acquired by SPEA), Acer, Motorola, and Acumos (bought by Cirrus Logic).
However, one firm persistently gaining energy was ATI. As a forerunner to the All-In-Wonder collection, ATI introduced the 68890 PC TV decoder chip in late November, which debuted in the Video-It! card. This chip was succesful of capturing video at 320×240 at 15 fps or 160×120 at 30 fps. It additionally featured real-time compression/decompression due to the onboard Intel i750PD VCP (Video Compression Processor) and will talk with the graphics board by way of the knowledge bus, eliminating the want for dongles, ports, and ribbon cables. The Video-It! was priced at $399, whereas a much less feature-rich mannequin named Video-Basic rounded out the product line.
Five months later, in March, ATI belatedly launched a 64-bit accelerator: the Mach64. The monetary yr proved difficult for ATI, with a CAD$2.7 million loss as the firm struggled amid fierce competitors. Rival boards included the S3 Vision 968, which was fashionable amongst many board distributors, and the Trio64, which secured OEM contracts with Dell (Dimension XPS), Compaq (Presario 7170/7180), AT&T (Globalyst), HP (Vectra VE 4), and DEC (Venturis/Celebris).
Released in 1995, the Mach64 achieved a number of notable firsts. It was the first graphics adapter out there for each PC and Mac computer systems, as demonstrated by the Xclaim ($450 and $650, relying on onboard reminiscence). Along with S3’s Trio, it supplied full-motion video playback acceleration.
The Mach64 additionally ushered in ATI’s entry into skilled graphics playing cards with the 3D Pro Turbo and 3D Pro Turbo+PC2TV, priced at $599 for the 2MB possibility and $899 for the 4MB variant.
The following month witnessed the emergence of a expertise startup, 3DLabs, born from DuPont’s Pixel graphics division buying its subsidiary. They introduced the GLINT 300SX processor, succesful of OpenGL rendering, fragment processing, and rasterization.
Initially focusing on the skilled market as a consequence of their excessive costs, the Fujitsu Sapphire2SX 4MB retailed for $1,600-$2,000, whereas the 8MB ELSA GLoria 8 was priced between $2,600 and $2,850. However, the 300SX was supposed for the gaming market.
S3 gave the impression to be in every single place at the moment. The high-end OEM market was dominated by the firm’s Trio64 chipsets, which built-in a DAC, graphics controller, and clock synthesizer right into a single chip.
The 1995 Gaming GLINT 300SX, that includes a decreased 2MB of reminiscence (1MB for textures and Z-buffer and 1MB for the body buffer), had an possibility to extend VRAM for Direct3D compatibility for a further $50 over the $349 base worth. Despite its capabilities, the card struggled in an already saturated market, however 3DLabs was already growing its successor in the Permedia collection.
S3 gave the impression to be in every single place at the moment. The high-end OEM marked was dominated by the firm’s Trio64 chipsets that built-in DAC, a graphics controller, and clock synthesiser right into a single chip. They additionally utilized a unified body buffer and supported {hardware} video overlay (a devoted portion of graphics reminiscence for rendering video as the software requires). The Trio64 and its 32-bit reminiscence bus sibling, the Trio32, have been out there as OEM models and standalone playing cards from distributors comparable to Diamond, ELSA, Sparkle, STB, Orchid, Hercules and Number Nine. Diamond Multimedia’s costs ranged from $169 for a ViRGE based mostly card, to $569 for a Trio64+ based mostly Diamond Stealth64 Video with 4MB of VRAM.
The mainstream market additionally noticed choices from Trident, a long-time OEM provider of fundamental 2D graphics adapters, who had just lately added the 9680 chip to their lineup. This chip shared many options with the Trio64, with boards usually priced round $170-200. They supplied acceptable 3D efficiency in that worth vary, with good video playback functionality.
Other newcomers in the mainstream market included Weitek’s Power Player 9130 and Alliance Semiconductor’s ProMovement 6410 (typically branded as the Alaris Matinee or FIS’s OptiViewPro). Both confirmed wonderful scaling with CPU velocity, with the latter combining a powerful scaling engine with anti-blocking circuitry for smoother video playback, outperforming earlier chips like the ATI Mach64, Matrox MGA 2064W, and S3 Vision968.
Nvidia launched their first graphics chip, the NV1, in May, and it turned the first business graphics processor succesful of 3D rendering, video acceleration, and built-in GUI acceleration.
They partnered with ST Microelectronics to provide the chip utilizing their 500nm course of. ST Microelectronics additionally promoted their model of the chip, the STG2000. Although it wasn’t an enormous success, it did symbolize Nvidia’s first monetary return. Unfortunately for Nvidia, simply as the first vendor boards started transport (notably the Diamond Edge 3D) in September, Microsoft finalized and launched DirectX 1.0.
The D3D graphics API was based mostly on rendering triangular polygons, whereas the NV1 used quad texture mapping. Limited D3D compatibility was added by way of drivers to wrap triangles as quadratic surfaces, however the lack of video games tailor-made for the NV1 doomed the card as a jack of all trades, grasp of none.
Most of the video games for the NV1 have been ported from the Sega Saturn. A 4MB NV1 with built-in Saturn ports (two per enlargement bracket linked to the card by way of a ribbon cable) retailed for round $450 in September 1995.
Microsoft’s late adjustments and the launch of the DirectX SDK left board producers unable to instantly entry {hardware} for digital video playback, leading to performance points for nearly all discrete graphics playing cards in Windows 95. By distinction, drivers below Win 3.1 from numerous corporations have been usually faultless.
ATI introduced their first 3D accelerator chip, the 3D Rage (also referred to as the Mach 64 GT), in November 1995. The first public demonstration of it was at the E3 online game convention held in Los Angeles in May the following yr, and the card turned out there a month later. The 3D Rage mixed the 2D core of the Mach64 with 3D capabilities.
Late revisions to the DirectX specification meant that the 3D Rage had compatibility issues with many video games utilizing the API, primarily as a consequence of the lack of depth buffering. With an onboard 2MB EDO RAM body buffer, 3D modality was restricted to 640x480x16-bit or 400x300x32-bit. Attempting 32-bit coloration at 600×480 usually resulted in on-screen coloration corruption, and 2D decision peaked at 1280×1024. While gaming efficiency was mediocre, its full-screen MPEG playback potential at the very least considerably balanced the characteristic set.
The 3D efficiency race was successfully over earlier than it had began, with the 3Dfx Voodoo Graphics annihilating all competitors.
ATI reworked the chip, and in September, the Rage II launched. It rectified the D3DX points of the first chip and added MPEG2 playback help. Initial playing cards nonetheless shipped with 2MB of reminiscence, hampering efficiency and inflicting points with perspective/geometry rework. As the collection expanded to incorporate the Rage II+DVD and 3D Xpression+, reminiscence capability choices grew to 8MB.
While ATI was first to market with a 3D graphics resolution, it wasn’t lengthy earlier than different rivals with completely different approaches to 3D implementation arrived on the scene, particularly 3Dfx, Rendition, and VideoLogic.
In the race to launch new merchandise into the market, 3Dfx received over Rendition and VideoLogic. The efficiency race was over earlier than it had began, with the 3Dfx Voodoo Graphics successfully annihilating all competitors.
This is the first article on our History of the GPU collection. If you loved this, hold studying as we take a stroll down reminiscence lane to the heyday of 3Dfx, Rendition, Matrox, and a younger firm referred to as Nvidia.
TechSpot’s History of the GPU
The fashionable graphics processor has change into one of the largest, most complicated, and costliest elements present in nearly any computing system. From the early VGA days to the fashionable GPU. The historical past and evolution of the chip that got here to dominate gaming, and later AI and compute.
1976 – 1995
1995 – 1999
2000 – 2006
2006 – 2013
2013 – 2020