Toggle search
Toggle menu
5.1K
5.9K
229
50.6K
VR & AR Wiki
Toggle preferences menu
Toggle personal menu
Not logged in
Your IP address will be publicly visible if you make any edits.

GeForce

From VR & AR Wiki
More actions
GeForce
Information
Type Graphics processing unit
Developer NVIDIA
Release Date October 11, 1999 (GeForce 256)
Website https://www.nvidia.com/en-us/geforce/


GeForce is NVIDIA's brand of graphics processing units (GPUs) and graphics cards for consumer desktop PCs and laptops. The line began with the GeForce 256, which NVIDIA marketed in 1999 as "the world's first GPU", and the current generation is the GeForce RTX 50 Series, announced at CES on January 6, 2025.[1][2]

In PC-based virtual reality, the GPU in the user's computer renders a separate image for each eye at the headset's refresh rate, so GeForce cards have been a reference point in headset system requirements since the first consumer Oculus Rift: Oculus's 2015 recommended specification named the GeForce GTX 970.[3] NVIDIA has added VR-specific hardware and driver features to GeForce GPUs through its VRWorks technologies, ran a "GeForce GTX VR Ready" badge program for qualifying cards and PCs from 2016,[4] and extended its DLSS upscaling to VR games in 2021.[5] The brand also covers GeForce NOW, NVIDIA's cloud gaming service.[6]

Reviewed 13 October 2026. Checked every date, price, spec, quote and attribution against the cited NVIDIA, Meta, Valve, IEEE Micro and press sources, plus the March 2026 DLSS 4.5 6X release. About review dates.

History

GeForce 256 (1999)

NVIDIA announced the GeForce 256 in August 1999, a few months after its RIVA TNT2, and the card went on sale in October 1999.[7] The Centre for Computing History dates its release to October 11, 1999, and records it as the first product released under the GeForce brand. NVIDIA defined a GPU at the time as "a single-chip processor with integrated transform, lighting, triangle setup/clipping, and rendering engines" capable of processing at least 10 million polygons per second.[8] The main change was hardware transform and lighting (T&L): the geometry stage that positions vertices relative to the camera and computes scene lighting moved from the CPU to the graphics chip.[1][9]

The original card used 32 MB of SDR memory, with a 120 MHz GPU clock according to The FPS Review;[7] a version with DDR memory followed in December 1999.[9] NVIDIA's own corporate timeline describes 1999 as the year the company "invents the GPU".[10]

Programmable shaders and unified architecture (2001-2006)

A 2008 IEEE Micro paper by NVIDIA engineers Erik Lindholm, John Nickolls, Stuart Oberman and John Montrym traces the next steps. The GeForce 256 had a fixed-function floating-point vertex T&L processor and a fixed-function integer pixel pipeline, programmed through OpenGL and Microsoft's DirectX 7. In 2001 the GeForce 3 introduced what the authors call the first programmable vertex processor, executing vertex shaders, and the later GeForce FX added 32-bit floating-point pixel-fragment processors.[11]

The GeForce 8800, introduced in November 2006, was the first GPU with NVIDIA's Tesla architecture. It replaced separate vertex and pixel processors with a single array of 128 streaming-processor cores organized as 16 streaming multiprocessors, and the same cores could run general-purpose programs written in C through the CUDA programming model.[11] NVIDIA's timeline places the unveiling of the CUDA architecture in 2006.[10]

The first consumer VR headsets (2015-2016)

When Oculus published the recommended PC specification for the consumer Oculus Rift on May 15, 2015, the graphics requirement was an "NVIDIA GTX 970 / AMD 290 equivalent or greater". Oculus's Atman Binstock explained that the Rift's 2160x1200 display at 90 Hz consumes 233 million pixels per second, compared with about 124 million for a 1080p game at 60 Hz, and that with the default eye-target scale a VR game needs roughly three times the GPU power of 1080p rendering.[3]

At CES in January 2016, NVIDIA announced the GeForce GTX VR Ready badge for graphics cards and PCs meeting its VR requirements: a GeForce GTX 970 for desktops or a GeForce GTX 980 for notebooks, an Intel Core i5-4590 or better, 8 GB or more of RAM, two USB 3.0 ports, HDMI 1.3 and Windows 7 SP1 or later. PCWorld listed Alienware, Maingear, MSI, Asus, HP and Falcon Northwest among the PC makers, and EVGA and Zotac among the card makers, taking part.[4] The program is covered in more detail in NVIDIA VR Ready.

On May 6, 2016, NVIDIA introduced the GeForce GTX 1080, which it called the world's first 16 nm FinFET, Pascal-powered graphics card, with retail availability from add-in card partners expected at the end of May and pricing "expected to start at $599". Its Simultaneous Multi-Projection hardware brought two VR rendering features: Lens Matched Shading, which renders closer to the shape of the image after the headset's lens warp, and Single Pass Stereo, which lets both eyes share one geometry pass.[12]

GeForce RTX (2018-present)

On August 20, 2018, NVIDIA announced the GeForce RTX 20 Series, the first gaming GPUs on its Turing architecture, with RT cores for real-time ray tracing and Tensor Cores for neural network processing. Launch pricing started at US$499 for the RTX 2070, US$699 for the RTX 2080 and US$999 for the RTX 2080 Ti (Founders Edition cards: US$599, US$799 and US$1,199). For VR, the cards supported VirtualLink, an open standard for powering and driving next-generation headsets over a single USB-C connector.[13] VirtualLink did not catch on. Road to VR reported that the later RTX 20 Super cards dropped the USB-C port and that none of the RTX 30 Series cards announced in September 2020 appeared to include it, and described the standard as abandoned.[14]

The GeForce RTX 30 Series, on the Ampere architecture, was announced on September 1, 2020, with the RTX 3080 at US$699 (available September 17), the RTX 3090 at US$1,499 (September 24) and the RTX 3070 at US$499 (October).[15] The RTX 40 Series followed on September 20, 2022, built on the Ada Lovelace architecture and launched with DLSS 3, which generates entire frames with AI; NVIDIA set the RTX 4090's availability for October 12, 2022, from US$1,599.[16]

NVIDIA announced the Blackwell-based GeForce RTX 50 Series at CES on January 6, 2025. NVIDIA set January 30, 2025, as the launch date for the RTX 5090 (92 billion transistors) and RTX 5080, at US$1,999 and US$999, and February 2025 for the RTX 5070 Ti and RTX 5070, at US$749 and US$549. The series introduced DLSS 4 Multi Frame Generation, which NVIDIA says generates up to three frames for each rendered frame.[2] NVIDIA's RTX 50 Series page lists the RTX 5090, 5080, 5070 Ti, 5070, 5060 Ti, 5060 and 5050.[17] At CES 2026 NVIDIA announced DLSS 4.5, with a second-generation transformer model for DLSS Super Resolution that all GeForce RTX owners could apply through the NVIDIA app, and a 6X Dynamic Multi Frame Generation mode for the RTX 50 Series planned for spring 2026.[18] Dynamic Multi Frame Generation and the 6X mode became available on March 31, 2026, through an NVIDIA app update; NVIDIA describes 6X as generating five additional frames for every natively rendered frame on RTX 50 Series GPUs.[19]

Selected generations

Line Architecture Announced or introduced Notes Source
GeForce 256 August 1999 (on sale October 1999) Marketed as the first GPU; hardware T&L [7][8]
GeForce 3 2001 First programmable vertex processor, per NVIDIA engineers [11]
GeForce 8800 Tesla November 2006 Unified shader cores; CUDA computing [11]
GeForce GTX 10 Series Pascal May 2016 Simultaneous Multi-Projection, Lens Matched Shading, Single Pass Stereo [12]
GeForce RTX 20 Series Turing August 2018 RT cores, Tensor Cores, VirtualLink USB-C port, Multi-View Rendering [13][20]
GeForce RTX 30 Series Ampere September 2020 No VirtualLink port [15][14]
GeForce RTX 40 Series Ada Lovelace September 2022 DLSS 3 frame generation [16]
GeForce RTX 50 Series Blackwell January 2025 DLSS 4 Multi Frame Generation [2]

VR features

Most of NVIDIA's VR rendering work for GeForce GPUs is grouped under VRWorks, which has its own article. The features below are tied to specific GeForce generations or drivers.

Single Pass Stereo and Multi-View Rendering. Pascal's Single Pass Stereo draws at most two views in one pass, and the two views can differ only along the horizontal axis, which suits headsets whose two displays are coplanar and whose field of view is limited. Turing GPUs added Multi-View Rendering, which raises the number of projection views in a single rendering pass from two to four, each with its own projection center. NVIDIA presented it as a way to support canted displays, where the two eye cameras are no longer parallel, and very wide field-of-view headsets.[20]

Variable Rate Supersampling (VRSS). VRSS builds on Turing's variable rate shading. The driver supersamples the central region of each frame, where the user is usually looking, by up to 8x while leaving the periphery untouched, and only uses idle GPU cycles. It works without developer integration, but the game must be a DirectX 11, forward-rendered title using MSAA that NVIDIA has profiled; 24 titles were supported at launch with driver 441.87.[21] VRSS 2, released in April 2021 with driver R465, moved the supersampled region with the user's gaze using Tobii Spotlight eye tracking, a form of dynamic foveated rendering. NVIDIA named HP's Omnicept G2 headset (the HP Reverb G2 Omnicept Edition) as the first to use it.[22]

DLSS in VR. In May 2021 a GeForce Game Ready driver added DLSS support to nine games, three of which work in VR: No Man's Sky, Wrench and Into the Radius. In NVIDIA's own tests, DLSS doubled the VR performance of No Man's Sky to a stable 90 fps on the Ultra preset with an RTX 3080 and an Oculus Quest 2, and improved Wrench by 80% in VR mode. DLSS requires a GeForce RTX card.[5]

Headset requirements

PC VR headset makers usually state their GPU requirements as a GeForce model (and an AMD equivalent). Examples from official sources:

Headset or program GeForce requirement as published Source
Oculus Rift (recommended specification, 2015) NVIDIA GTX 970 / AMD 290 equivalent or greater [3]
GeForce GTX VR Ready badge (2016) GeForce GTX 970 (desktop) or GTX 980 (notebook) [4]
Valve Index NVIDIA GeForce GTX 970+ or AMD RX480+ [23]
Meta Horizon Link (Meta Quest Link) Recommended: Nvidia RTX 20 Series or AMD Radeon RX 6000 Series; the RTX 2050 is not supported [24]

GeForce NOW

GeForce NOW is NVIDIA's service for streaming PC games from its servers. It left beta on February 4, 2020, with a free tier limited to one-hour sessions, a paid tier at US$5 a month with priority access and sessions of up to six hours, and more than 1,000 games, including titles users already owned on stores such as Steam.[6] At CES 2025, NVIDIA announced that GeForce NOW would come to the Apple Vision Pro, Meta Quest 3 and Meta Quest 3S and Pico headsets through their web browsers, starting with version 2.0.70. Road to VR noted that the announcement covered ordinary games shown on a large virtual screen and did not appear to include VR games.[25]

See also

References

  1. ↑ 1.0 1.1 Vivek Singh (2024-10-11). "Game-Changer: How the World's First GPU Leveled Up Gaming and Ignited the AI Era". NVIDIA Blog. NVIDIA. https://blogs.nvidia.com/blog/first-gpu-gaming-ai/. Retrieved 2026-10-13.
  2. ↑ 2.0 2.1 2.2 "NVIDIA Blackwell GeForce RTX 50 Series Opens New World of AI Computer Graphics". NVIDIA Newsroom. NVIDIA. 2025-01-06. https://nvidianews.nvidia.com/news/nvidia-blackwell-geforce-rtx-50-series-opens-new-world-of-ai-computer-graphics. Retrieved 2026-10-13.
  3. ↑ 3.0 3.1 3.2 Atman Binstock (2015-05-15). "Powering the Rift". Meta Horizon Developers blog. Meta. https://developers.meta.com/horizon/blog/powering-the-rift/. Retrieved 2026-10-13.
  4. ↑ 4.0 4.1 4.2 Gordon Mah Ung (2016-01-04). "Want VR? Nvidia says look for the GTX VR Ready badge". PCWorld. https://www.pcworld.com/article/3018897/hardware/want-vr-nvidia-says-look-for-gtx-vr-badge.html. Retrieved 2026-10-13.
  5. ↑ 5.0 5.1 Theo Binns (2021-05-18). "VR games now have access to Nvidia DLSS, starting with No Man's Sky". PCGamesN. https://www.pcgamesn.com/nvidia/vr-dlss-support. Retrieved 2026-10-13.
  6. ↑ 6.0 6.1 Ben Reeves (2020-02-04). "Nvidia GeForce Now Streaming Service Exits Beta". Game Informer. https://www.gameinformer.com/2020/02/04/nvidia-geforce-now-streaming-service-exits-beta. Retrieved 2026-10-13.
  7. ↑ 7.0 7.1 7.2 Tsing Mui (2024-09-02). "NVIDIA Launched the World's First GPU 25 Years Ago". The FPS Review. https://www.thefpsreview.com/2024/09/02/nvidia-launched-the-worlds-first-gpu-25-years-ago/. Retrieved 2026-10-13.
  8. ↑ 8.0 8.1 "Nvidia releases GeForce 256". Centre for Computing History. https://www.computinghistory.org.uk/det/32766/Nvidia-releases-GeForce-256/. Retrieved 2026-10-13.
  9. ↑ 9.0 9.1 Ben Hardwidge (2024-10-11). "Remembering the Nvidia GeForce 256: the first PC gaming GPU turns 25 today". PCGamesN. https://www.pcgamesn.com/nvidia/geforce-256. Retrieved 2026-10-13.
  10. ↑ 10.0 10.1 "Corporate Timeline". NVIDIA. NVIDIA. https://www.nvidia.com/en-us/about-nvidia/corporate-timeline/. Retrieved 2026-10-13.
  11. ↑ 11.0 11.1 11.2 11.3 Erik Lindholm, John Nickolls, Stuart Oberman, John Montrym (2008). "NVIDIA Tesla: A Unified Graphics and Computing Architecture". IEEE Micro, vol. 28, no. 2. https://doi.org/10.1109/MM.2008.31. Retrieved 2026-10-13.
  12. ↑ 12.0 12.1 Andrew Burnes (2016-05-06). "Introducing The GeForce GTX 1080: Gaming Perfected". GeForce News. NVIDIA. https://www.nvidia.com/en-us/geforce/news/geforce-gtx-1080/. Retrieved 2026-10-13.
  13. ↑ 13.0 13.1 "10 Years in the Making: NVIDIA Brings Real-Time Ray Tracing to Gamers with GeForce RTX". NVIDIA Newsroom. NVIDIA. 2018-08-20. https://nvidianews.nvidia.com/news/10-years-in-the-making-nvidia-brings-real-time-ray-tracing-to-gamers-with-geforce-rtx. Retrieved 2026-10-13.
  14. ↑ 15.0 15.1 "NVIDIA Delivers Greatest-Ever Generational Leap with GeForce RTX 30 Series GPUs". NVIDIA Newsroom. NVIDIA. 2020-09-01. https://nvidianews.nvidia.com/news/nvidia-delivers-greatest-ever-generational-leap-in-performance-with-geforce-rtx-30-series-gpus. Retrieved 2026-10-13.
  15. ↑ 16.0 16.1 "NVIDIA Delivers Quantum Leap in Performance, Introduces New Era of Neural Rendering With GeForce RTX 40 Series". NVIDIA Newsroom. NVIDIA. 2022-09-20. https://nvidianews.nvidia.com/news/nvidia-delivers-quantum-leap-in-performance-introduces-new-era-of-neural-rendering-with-geforce-rtx-40-series. Retrieved 2026-10-13.
  16. ↑ "GeForce RTX 50 Series Graphics Cards". NVIDIA. NVIDIA. https://www.nvidia.com/en-us/geforce/graphics-cards/50-series/. Retrieved 2026-10-13.
  17. ↑ Andrew Burnes (2026-01-05). "GeForce @ CES 2026: NVIDIA DLSS 4.5 Announced, 250+ DLSS Multi Frame Generation Games & Apps Available Now, G-SYNC Pulsar Gaming Monitors Available This Week, And Much More". GeForce News. NVIDIA. https://www.nvidia.com/en-us/geforce/news/ces-2026-nvidia-geforce-rtx-announcements/. Retrieved 2026-10-13.
  18. ↑ Andrew Burnes (2026-03-31). "DLSS 4.5 Dynamic Multi Frame Generation & Multi Frame Generation 6X Available Now". GeForce News. NVIDIA. https://www.nvidia.com/en-us/geforce/news/dlss-4-5-dynamic-multi-frame-generation-6x-mode-released/. Retrieved 2026-10-13.
  19. ↑ 20.0 20.1 Swaroop Bhonde, Mahalakshmi Shanmugam (2018-09-24). "Turing Multi-View Rendering in VRWorks". NVIDIA Technical Blog. NVIDIA. https://developer.nvidia.com/blog/turing-multi-view-rendering-vrworks/. Retrieved 2026-10-13.
  20. ↑ DeepChand Palswamy, Swaroop Bhonde (2020-01-10). "NVIDIA VRSS, a Zero-Effort Way to Improve Your VR Image Quality". NVIDIA Technical Blog. NVIDIA. https://developer.nvidia.com/blog/nvidia-vrss-a-zero-effort-way-to-improve-your-vr-image-quality. Retrieved 2026-10-13.
  21. ↑ DeepChand Palswamy, Peter Pang (2021-04-12). "Delivering Dynamic Foveated Rendering with NVIDIA VRSS 2". NVIDIA Technical Blog. NVIDIA. https://developer.nvidia.com/blog/delivering-dynamic-foveated-rendering-with-nvidia-vrss-2/. Retrieved 2026-10-13.
  22. ↑ "Valve Index". Steam. Valve. https://store.steampowered.com/valveindex. Retrieved 2026-10-13.
  23. ↑ Scott Hayden (2025-01-07). "NVIDIA Announces GeForce Now Cloud Gaming Support for Quest 3/S, Vision Pro & Pico Headsets". Road to VR. https://www.roadtovr.com/nvidia-cloud-gaming-quest-3-vision-pro-pico/. Retrieved 2026-10-13.
Retrieved from "https://vrarwiki.com/index.php?title=GeForce&oldid=50231"