GeForce RTX 40 series
Nvidia consumer graphics cards using Ada Lovelace architecture
Nvidia launched this family of graphics processors in October 2022. Built on the Ada Lovelace design and TSMC 4N process, it brings third-generation ray tracing and fourth-generation Tensor Cores to gamers and creators.
- Uses TSMC 4N custom five-nanometer process
- Features third-generation RT cores and fourth-gen Tensor Cores
- RTX 4090 has 76.3 billion transistors and 450 W TDP
- Supports AV1 hardware encoding and DLSS 3 frame generation
- Super refresh launched in January 2024 with improved specs
What it is
The GeForce RTX 40 series is a line of consumer graphics processing units from Nvidia. It serves as the successor to the RTX 30 series. These chips power desktop and laptop computers for gaming and content creation. The range includes models from the entry-level RTX 4060 to the flagship RTX 4090. Nvidia announced the family in September 2022 and began sales the following month. The cards rely on the Ada Lovelace microarchitecture to deliver improved performance and efficiency.
How it came about
Nvidia revealed the series at the GPU Technology Conference in September 2022. The launch started with the high-end RTX 4090 in October 2022. The RTX 4080 followed in November. An initial plan for a 12GB RTX 4080 faced backlash over naming and specs. Nvidia cancelled that specific model and rebranded it as the RTX 4070 Ti in January 2023. Mid-range and entry-level cards arrived throughout 2023. A Super refresh updated several models in early 2024 with better specs and lower prices.
Inside the chip
These processors use the Ada Lovelace architecture. They feature third-generation RT cores for hardware-accelerated ray tracing. Fourth-generation Tensor Cores handle deep learning tasks like DLSS 3 frame generation. The design supports Shader Execution Reordering to improve efficiency. Video encoding uses dual NVENC engines with AV1 support. Display outputs include DisplayPort 1.4a and HDMI 2.1. Double-precision performance is set at one sixty-fourth of single-precision speed. NVLink support was removed from this generation.
How it was made
TSMC manufactures the chips using its 4N process node. This is a custom five-nanometer technology designed for Nvidia. Transistor counts vary by die size. The smallest AD107 die holds 18.9 billion transistors. The largest AD102 die contains 76.3 billion transistors. High-end models use GDDR6X memory from Micron. Lower-tier cards use standard GDDR6. Some mobile variants also use GDDR6. The physical cards often require large cooling solutions due to power demands.
Where it went
Gamers use these cards for real-time ray tracing and high-resolution play. The RTX 4090 targets native 4K gaming. Content creators benefit from AV1 encoding and large video memory buffers. Streamers use the dual encoders for high-quality broadcasts. The cards fit into desktop PCs and some laptops. They support modern APIs like DirectX 12 Ultimate and Vulkan 1.4. Professionals use them for AI inference and video editing tasks.
Versions and family
The lineup spans from the RTX 4050 laptop chip to the desktop RTX 4090. Desktop models include the 4060, 4070, 4080, and 4090. Super variants launched in 2024 replaced some original models. The RTX 4060 Ti comes in 8GB and 16GB memory versions. Laptop SKUs all use GDDR6 memory. Desktop cards above the 4060 Ti use GDDR6X. A revised RTX 4070 with faster GDDR6 appeared in August 2024.
Why it matters
The RTX 4090 set new benchmarks for consumer graphics performance. It enabled native 4K gaming with ray tracing enabled. The series pushed power consumption higher for top-tier models. Sales of the RTX 4080 and 4070 Ti were initially weak due to pricing. The Super refresh aimed to correct value propositions. The architecture introduced widespread AV1 encoding support for streamers. DLSS 3 frame generation became a key selling point.
Problems and reception
Reviewers praised the RTX 4090 for its extreme speed. Critics noted its high price and large physical size. The RTX 4080 faced criticism for its cost increase over previous generations. The RTX 4060 Ti 8GB model was denounced for insufficient video memory. Lack of review samples for the 16GB Ti variant caused controversy. Power efficiency received mixed comments across the range. Some users reported difficulty fitting large cards into standard cases.




Questions people ask about GeForce RTX 40 series
What architecture powers the RTX 40 series?
The series uses the Ada Lovelace microarchitecture. It includes third-generation ray tracing cores and fourth-generation Tensor Cores for AI tasks. This design enables features like DLSS 3 and Shader Execution Reordering.
Who manufactures the chips for this series?
TSMC produces the silicon using its 4N process node. This is a custom five-nanometer technology built specifically for Nvidia. The foundry handles the physical creation of the GPU dies.
Why was the RTX 4080 12GB cancelled?
Nvidia faced backlash for marketing two different GPUs under the same 4080 name. The 12GB version had significantly fewer cores and a narrower bus than the 16GB model. The company rebranded it as the RTX 4070 Ti instead.
What memory types do these cards use?
Desktop models from the RTX 4070 upwards use GDDR6X memory. The RTX 4060 and 4060 Ti use standard GDDR6. All laptop variants in the series utilize GDDR6 memory modules.
Good reading and sources
- Official websiteOfficial site
- NVIDIA Delivers Quantum Leap in Performance, Introduces New Era of Neural Rendering With GeForce RTX 40 SeriesNvidia, Article
- Why NvidiaTom's Hardware, Article
- Nvidia says it’s “unlaunching” the 12GB RTX 4080 after backlashThe Verge, Article
- Nvidia GeForce RTX 4090 review: Brutally, futuristically fastPCWorld, Article
- GeForce RTX 40 series on WikipediaEncyclopedia
- More photos on Wikimedia CommonsPhotos
- Wikidata recordOpen data
The facts on this page come from the Wikipedia article "GeForce RTX 40 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.
Like GeForce RTX 40 series
All graphics processorsMore from Nvidia
All of them
🇺🇸 United States and 1 moreApple A16 Bionic
ARM-based system on a chip for iPhone and iPadSoC · Arm · 2022