Nvidia Corporation is a multinational company specializing in the development of graphic processing units and integrated circuit technologies for workstations, personal computers and mobile devices. Headquartered in Santa Clara, California, the company has become one of the leading providers of integrated circuits (CI), such as graphics processing units (GPUs) and chipsets used in graphics cards for consoles and motherboards for PC.
Nvidia expanded its presence in the video game industry with its SHIELD Portable, SHIELD Tablet and SHIELD Android TV. Since 2014, Nvidia has changed to be a company focused on four markets: gaming, professional visualization, data centers and auto. NVIDIA produces GPUs including the GeForce series for video games, the NVIDIA Quadro series of computer-aided design and digital content creation at workstations, and the nForce integrated circuit series for motherboards.
Jen-Hsun Huang, Chris Malachowsky and Curtis Priem founded the company in January 1993 and placed the headquarters in California in April of 1993.
In 1995 NVIDIA launches its first product, NVIDIA NV1. This is a PCI card which was marketed under the name of Diamond Edge 3D based on texture maps of quadratic surfaces.
The GeForce name that he used for his graphic models was created through a contest in California to name the next generation of NV20 graphics cards.
In 2012, tablets and smartphones with the Tegra 3 processor, created by NVIDIA, appear on the market.
In 2013 Nvidia decided to license its graphic technology so that manufacturers of all types of devices can take advantage of their patent catalog, following the same model as ARM Holdings.
For fiscal year 2018, Nvidia reported revenues of US $3.047 billion, with an annual income of US $9.714 billion, an increase of 40.6% over the previous fiscal cycle. Nvidia's shares were trading at more than $245 per share, and its market capitalization was valued at more than $120.6 billion in September 2018.
- RIVA 128, RIVA 128ZX
- VANTA LT, RIVA TNT, RIVA TNT 2
From the GeForce series, chipsets take care of virtually the entire graphic process, constituting what NVIDIA named GPU (Graphic Processing Unit).
- GeForce Series
- GeForce 256
- GeForce 2: MX 200, MX 400, GTS, Pro, Ti and Ultra.
- GeForce 3: Ti 200 and Ti 500.
- GeForce 4: MX 420, MX 440, MX 460, Ti 4200, Ti 4400, Ti 4600 and Ti 4800.
- GeForce FX (5): (DirectX 8.0, 8.1 or lately 9.0b per hardware) composed of FX 5950 Ultra, FX 5900, FX 5800, FX 5700, FX 5600, FX 5500, FX 5300 and FX 5200 models.
- GeForce 6: (DirectX 9.0c by hardware) composed of 6800 Ultra, 6800 GT, 6800, 6600 GT, 6600, 6500, 6200, 6150 and 6100 (AGP) models.
- GeForce 7: (DirectX 9.0c by hardware) composed of 7950 GX2, 7950 GT, 7900 GTX, 7900 GTO, 7900 GT, 7900 GS, 7800 GTX, 7800 GT, 7800 GS, 7600 GT, 7600 GS, 7300 GT, 7300 GS, 7300 LE, 7100 GS and 7025 GT.
- GeForce 8: (DirectX 10.0 by hardware) composed of 8800 Ultra, 8800 GTX, 8800 GTS, 8800 GT, 8800 GS, 8600 GTS, 8600 GT, 8500 GT and 8400 GS models.
- GeForce 9: (DirectX 10.0 by hardware) composed of 9300 GS, 9400 GT, 9500 GT, 9600 GSO, 9600 GSO 512, 9600 GT Green Edition, 9600 GT, 9800 GT Green Edition, 9800 GT, 9800 GTX, 9800 GTX + and 9800 GX2.
- GeForce 200: (DirectX 10 and 10.1 hardware) composed of the GT 220, GT 240, GTS 240, GTS 250, GTX 260, GTX 275, GTX 280, GTX 285 and GTX 295 models.
- GeForce 300: (DirectX 10.1 by hardware and DirectX 11 by Software) It is the GT 200 series but improved, lower consumption, less manufacturing size (Less than 55 nanometers of the GT 200).
- GeForce 400: (DirectX 11.0 by hardware) Architecture of code name "Fermi", temporarily composed of the GT 420, GT 430, GT 440, GTS 450, GTX 460, GTX 465, GTX 470 and GTX 480.
- GeForce 500: GT 520, GTX 550 Ti, GTX 560, GTX 560 Ti, GTX 570, GTX 580 and GTX 590.
- GeForce 600: (Kepler codename architecture, except for Fermi-based 610, 620 and 630 models) GT 610, GT 620, GT 630, GT 640, GTX 650, GTX 650 Ti, GTX 650 Ti BOOST, GTX 660, GTX 660 Ti, GTX 670, GTX 680, GTX 690.
- GeForce 700: (improved Kepler architecture, except 750 and 750 Ti models based on Maxwell architecture) GT 730, GTX 750, GTX 750Ti, GTX 760, GTX 770, GTX 780, GTX 780 Ti, GTX TITAN, GTX TITAN Black and GTX TITAN Z.
- GeForce 900: (Maxwell codename architecture) GTX 950, GTX 960, GTX 970, GTX 980, GTX 980Ti and GTX TITAN X.
- GeForce 10: (Pascal codename architecture) GT 1030, GTX 1050, GTX 1050 Ti, GTX 1060, GTX 1070, GTX 1070 Ti, GTX 1080, GTX 1080Ti, TITAN X (Pascal Architecture) and TITAN Xp.
- GeForce 20: (Turing codename architecture) GTX 1650, GTX 1660, GTX 1660Ti, RTX 2060, RTX 2070, RTX 2080, RTX 2080 Ti and TITAN RTX.
Graphic chips for professional and scientific use
Chipset for motherboards
- nForce 6 Series
- nForce 7 Series
- 790i SLI
- 790i Ultra SLI
- 980th SLI for AMD (Release)
Chipsets for mobile devices
- NVIDIA GoForce - series of graphics processors created especially for mobile devices (PDA, smartphones and mobile phones). They include nPower technology for efficient energy use.
- GoForce 2150 - Supports mobile phones with camera up to 1.3 megapixels, 2D graphic acceleration.
- GoForce 3000 - Low cost version of the GoForce 4000 with some cropped features.
- GoForce 4000 - Camera support up to 3.0 megapixels with recording and playback of videos in MPEG-4/H.263.
- GoForce 4500 - 2D and 3D video acceleration with programmable pixel shaders and geometry
- GoForce 4800 - Camera support up to 3.0 megapixels with 2D and 3D video acceleration with programmable pixel shader.
- GoForce 6100 - Camera support up to 10 megapixels, 2D and 3D acceleration, and H.264 codec.
- NVIDIA Tegra - is a system on a chip for mobile devices such as smartphones, tablets, personal digital assistants, mobile devices and the Internet. Tegra integrates the architecture of ARM, GPU, northbridge, southbridge, and memory controller processors in a single package. The series emphasizes low power consumption and high performance for audio and video playback. This chip is used by portable devices such as the NVIDIA Shield or the Nintendo Switch.
Open source software support
Nvidia does not publish the documentation of its hardware instead, Nvidia provides its own binary drivers for X.Org and a thin open source library for the interface between the Linux, FreeBSD or Solaris cores and its proprietary software. Nvidia also provided, but stopped supporting, an obfuscated open source driver that only supports 2D acceleration.
The proprietary nature of Nvidia drivers has generated a lot of dissatisfaction within the free software community. Linus Torvalds has officially declared that Nvidia "has been the worst problem we have had with hardware manufacturers."
Nvidia drivers are known to have artificially imposed restrictions, such as limiting the number of monitors that can be used at the same time.