GeForce RTX 20 series
First consumer GPUs with hardware ray tracing
Nvidia launched this graphics card family in September 2018 using the Turing microarchitecture. It introduced dedicated cores for real-time ray tracing and AI upscaling, marking a major shift in consumer visual computing.
- First consumer GPUs with hardware ray tracing
- Uses TSMC 12 nm FinFET process
- Introduced Tensor Cores for AI upscaling
- Launched in September 2018
- Up to 18.6 billion transistors on TU102 die
What it is
The GeForce RTX 20 series is a line of graphics processing units created by Nvidia. Released in late 2018, these cards brought the Turing microarchitecture to the mainstream market. They were the first consumer products to include hardware specifically designed for real-time ray tracing. This technology allows for realistic lighting effects like reflections and shadows. The series also introduced Tensor Cores for artificial intelligence tasks such as image upscaling. Nvidia manufactured the chips using a 12 nanometre process from TSMC. The lineup skipped an entry-level tier, leaving that segment to the older GTX 16 series.
How it came about
Nvidia teased the new architecture at SIGGRAPH in early 2018 before officially announcing the cards at Gamescom that August. The initial launch included high-end models like the RTX 2080 and 2080 Ti. Shipping began in September 2018. Early reviews praised performance but criticized the high prices and limited software support for ray tracing. In response to market feedback, Nvidia released the Super refresh in July 2019 with better specifications at similar price points. The Titan RTX arrived in December 2018 as the most powerful variant. Supply constraints affected availability throughout the initial release period.
Inside the chip
The Turing architecture introduced two new types of specialized cores. RT cores handle ray tracing calculations by accelerating bounding volume hierarchy tests. This allows the GPU to simulate light paths efficiently. Tensor Cores perform deep learning operations, enabling features like DLSS. This technique uses AI to upscale lower-resolution images, improving frame rates. The design supports concurrent integer and floating-point operations via dedicated INT cores. Memory subsystems use GDDR6 chips for higher bandwidth. The hybrid rendering model combines traditional rasterization with ray traced data for enhanced visual fidelity without replacing existing pipelines entirely.
How it was made
TSMC produced the primary dies using its 12 nm FinFET process, which Nvidia marketed as an optimized 14 nm node. The largest chip, the TU102, contains approximately 18.6 billion transistors. Smaller variants like the TU106 have fewer transistors. Initially, the cards used GDDR6 memory modules from Micron Technology. Due to reported faults in early units, Nvidia switched to Samsung memory chips by November 2018. The mobile RTX 2050 later used a different die built on Samsung’s 8N process. The desktop cards connect via a PCIe 3.0 interface.
Where it went
These graphics cards target PC gaming and professional visualization tasks. They power desktop computers and laptops from various manufacturers. The hardware enables real-time ray tracing in supported games, providing realistic lighting and reflections. Developers use the RTX platform with tools like OptiX and Vulkan for creative applications. The Tensor Cores assist in AI-driven image enhancement and denoising. During the global semiconductor shortage, Nvidia reintroduced some models like the RTX 2060 to meet demand from gamers and cryptocurrency miners. The cards also support high-resolution displays up to 8K HDR.
Versions and family
The series includes several tiers ranging from the RTX 2060 to the enthusiast-grade RTX 2080 Ti and Titan RTX. A Super refresh in 2019 updated the 2060, 2070, and 2080 models with improved specs. The RTX 2050 for laptops uses the newer Ampere architecture instead of Turing. Nvidia also released a 12 GB version of the RTX 2060 in late 2021 due to supply issues with newer generations. The GTX 16 series served as the budget alternative, sharing the Turing architecture but lacking RT and Tensor cores.
Why it matters
This series established hardware-accelerated ray tracing as a standard feature in consumer graphics. It pushed game developers to adopt new rendering techniques for more photorealistic visuals. The introduction of DLSS demonstrated how AI could enhance gaming performance, a concept that became central to future GPU designs. Despite initial criticism regarding price and software readiness, the technology matured rapidly. The RTX branding became synonymous with advanced graphical features, influencing competitor strategies and setting the stage for subsequent generations like the RTX 30 series.
Problems and reception
Reviewers noted significant performance gains over the previous generation but highlighted the premium cost. Critics pointed out that few titles initially utilized ray tracing capabilities. Supply shortages led to tight availability and delayed purchases for many consumers. The loudness and heat output of some cooling solutions drew complaints. However, the RTX 2080 Ti was widely recognized as the fastest gaming GPU available at its launch. The Super refresh was seen as a necessary correction to pricing and performance balance.




Questions people ask about GeForce RTX 20 series
What makes the RTX 20 series different from previous cards?
It was the first consumer line to include dedicated RT cores for real-time ray tracing and Tensor Cores for AI tasks. This allowed for realistic lighting effects and features like DLSS, which were not possible on earlier hardware.
Who manufactured the chips in this series?
Nvidia designed the GPUs, while TSMC manufactured the main desktop dies using a 12 nm FinFET process. Some laptop models, like the RTX 2050, used chips made by Samsung on their 8N process.
Why did Nvidia release a Super version of these cards?
The Super refresh launched in July 2019 to offer better specifications at similar prices. It addressed consumer concerns about value and performance, providing a modest upgrade over the original non-Super models.
What type of memory do these graphics cards use?
The series uses GDDR6 memory for high bandwidth. Launch models featured chips from Micron, but Nvidia switched to Samsung memory modules in November 2018 following reports of faults with the initial supply.
Good reading and sources
- Official websiteOfficial site
- Nvidia’s new RTX 2080, 2080 Ti video cards ship on Sept. 20, starting at $699Ars Technica, Article
- Nvidia announces RTX 2000 GPU series with ‘6 times more performance’ and ray tracingThe Verge, Article
- Checking your browserAnandTech, Article
- GeForce RTX 20 series on WikipediaEncyclopedia
- More photos on Wikimedia CommonsPhotos
- Wikidata recordOpen data
The facts on this page come from the Wikipedia article "GeForce RTX 20 series", written by its contributors and used under the CC BY-SA 4.0 licence, and from its Wikidata record (CC0). EveryChip rewrote the text in its own words, with AI assistance, and compares every page with the article to make sure no passage is copied. It may be reused under the same CC BY-SA 4.0 licence with a link back.
It may contain mistakes, and figures change over time. Nothing here is investment or buying advice. If something is wrong, tell us. Last refreshed 2026-11-14.