| Revision as of 19:42, 22 September 2022 edit42.190.191.119 (talk) (edit summary removed)Tags: Undo Reverted section blanking← Previous edit | Revision as of 19:44, 22 September 2022 edit undoFfffrr (talk | contribs)Extended confirmed users99,459 edits Importing Wikidata short description: "Family of graphics processing units developed by Nvidia"Tag: Shortdesc helperNext edit → | ||
| Line 1: | Line 1: | ||
| {{Short description|Family of graphics processing units developed by Nvidia}} | |||
| {{Primary sources|date=September 2022}} | {{Primary sources|date=September 2022}} | ||
| {{Infobox graphics processing unit | {{Infobox graphics processing unit | ||
Revision as of 19:44, 22 September 2022
Family of graphics processing units developed by Nvidia | This article relies excessively on references to primary sources. Please improve this article by adding secondary or tertiary sources. Find sources: "GeForce 40 series" – news · newspapers · books · scholar · JSTOR (September 2022) (Learn how and when to remove this message) |
| Release date | October 12, 2022; 2 years ago (2022-10-12) |
|---|---|
| Codename | AD10x |
| Architecture | Ada Lovelace |
| Models | GeForce RTX series |
| Transistors |
|
| Fabrication process | TSMC 4N |
| Cards | |
| High-end | GeForce RTX 4080 (12 GB) |
| Enthusiast |
|
| API support | |
| DirectX | Direct3D 12.0 Ultimate (feature level 12_2) |
| OpenCL | OpenCL 3.0 |
| OpenGL | OpenGL 4.6 |
| Vulkan | Vulkan 1.3 |
| History | |
| Predecessor | GeForce 30 series |
The GeForce 40 series is a family of graphics processing units developed by Nvidia, succeeding the GeForce 30 series. The series was announced on September 20, 2022 at GTC 2022 GeForce Beyond A Special Broadcast at GTC event and are expected to start shipping on October 12, 2022. The cards are based on the Ada Lovelace architecture and feature hardware-accelerated raytracing (RTX) with Nvidia's third-generation RT cores and fourth-generation Tensor Cores.
Details
Architectural highlights of the Ada Lovelace architecture include the following:
- CUDA Compute Capability 8.9
- TSMC 4N process (custom designed for NVIDIA) – not to be confused with N4
- Fourth-generation Tensor Cores with FP8, FP16, bfloat16, TensorFloat-32 (TF32) and sparsity acceleration
- Third-generation Ray Tracing Cores, along with concurrent ray tracing, shading and compute
- Shader Execution Reordering
- NVENC with 8K 10-bit 60FPS AV1 fixed function hardware encoding
- Dropped NVLink support
Products
| Model | Launch | Launch MSRP (USD) |
Code name(s) |
Transistors (billion) | Die size (mm) | Core config |
SM count |
L2 cache (MB) |
Clock speeds | Fillrate | Memory | Processing power (TFLOPS) | TDP (watts) | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Core clock (MHz) |
Memory (GT/s) | Pixel (Gpx/s) |
Texture (Gtex/s) |
Size (GB) |
Bandwidth (GB/s) |
Type | Bus width (bit) |
Half precision (boost) |
Single precision (boost) |
Double precision (boost) |
Tensor compute (non-sparse) | ||||||||||
| GeForce RTX 4080 (12 GB) |
Nov 2022 | $899 | AD104-400 | 450 | 7680 |
60 | 48 | 2310 (2610) |
21 | 12 | 504 | GDDR6X | 192 | 285 | |||||||
| GeForce RTX 4080 (16 GB) |
Nov 2022 | $1199 | AD103-300 | 9728 |
76 | 2210 (2510) |
23 | 16 | 736 | 256 | 320 | ||||||||||
| GeForce RTX 4090 |
Oct 12, 2022 | $1599 | AD102-300 | 76.3 | 608.4 | 16384 512:384:128:512 |
144 | 96 | 2230 (2520) |
21 | 24 | 1008 | 384 | 450 | |||||||
- Shader Processors : Texture mapping units : Render output units : Ray tracing cores : Tensor Cores
- The number of Streaming multi-processors on the GPU.
- Core boost values (if available) are stated below the base value inside (brackets).
- Pixel fillrate is calculated as the lowest of three numbers: number of ROPs multiplied by the base core clock speed, number of rasterizers multiplied by the number of fragments they can generate per rasterizer multiplied by the base core clock speed, and the number of streaming multiprocessors multiplied by the number of fragments per clock that they can output multiplied by the base clock rate.
- Texture fillrate is calculated as the number of TMUs multiplied by the base core clock speed.
See also
- GeForce 10 series
- GeForce 16 series
- GeForce 20 series
- GeForce 30 series
- Quadro
- List of Nvidia graphics processing units
Notes
References
- ^ "NVIDIA Delivers Quantum Leap in Performance, Introduces New Era of Neural Rendering With GeForce RTX 40 Series". NVIDIA Newsroom.
- "NVIDIA Ada Lovelace Architecture". NVIDIA.
- "I.7. Compute Capability 9.x". docs.nvidia.com.
- "Creativity At The Speed of Light: GeForce RTX 40 Series Graphics Cards Unleash Up To 2X Performance in 3D Rendering, AI, and Video Exports For Gamers and Creators". NVIDIA.
- https://developer.nvidia.com/nvidia-video-codec-sdk
- "Jensen Confirms: NVLink Support in Ada Lovelace is Gone". TechPowerUp. 2022-09-21.
- ^ "GeForce RTX 40 Series". NVIDIA. Retrieved September 21, 2022.
- ^ "GeForce RTX 4080". NVIDIA. Retrieved September 21, 2022.
- "NVIDIA GeForce RTX 4080 12 GB Specs". TechPowerUp. Retrieved September 21, 2022.
- Mujtaba, Hassan (2022-09-20). "NVIDIA GeForce RTX 4090 "Ada Lovelace" Graphics Card Launched – 16384 Cores, 24 GB GDDR6X, 4X Faster Than RTX 3090 at $1599 US". Wccftech. Retrieved 2022-09-20.
- "NVIDIA GeForce RTX 4080 16 GB Specs". TechPowerUp. Retrieved September 21, 2022.
- "GeForce RTX 4090". NVIDIA. Retrieved September 21, 2022.
- "NVIDIA GeForce RTX 4090 Specs". TechPowerUp. Retrieved September 21, 2022.
External links
- Official website
- GeForce RTX 40 Series Graphics Cards: Up To 4X Faster, Powered By 3rd Gen RTX Architecture & NVIDIA DLSS 3
Media related to Nvidia GeForce 40 series video cards at Wikimedia Commons
| Nvidia | |||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| |||||||||||||||||||
| |||||||||||||||||||
| |||||||||||||||||||
| |||||||||||||||||||