Report an Error
NVIDIA A40 PCIe
- Graphics Processor
- GA102
- Cores
- 10752
- TMUs
- 336
- ROPs
- 112
- Memory Size
- 48 GB
- Memory Type
- GDDR6
- Bus Width
- 384 bit
GPU
Rear
Recommended Gaming Resolutions:
- 1920x1080
- 2560x1440
- 3840x2160
The A40 PCIe is a professional graphics card by NVIDIA, launched on October 5th, 2020. Built on the 8 nm process, and based on the GA102 graphics processor, the card supports DirectX 12 Ultimate. The GA102 graphics processor is a large chip with a die area of 628 mm² and 28,300 million transistors. It features 10752 shading units, 336 texture mapping units, and 112 ROPs. Also included are 336 tensor cores which help improve the speed of machine learning applications. The card also has 84 raytracing acceleration cores. NVIDIA has paired 48 GB GDDR6 memory with the A40 PCIe, which are connected using a 384-bit memory interface. The GPU is operating at a frequency of 1305 MHz, which can be boosted up to 1740 MHz, memory is running at 1812 MHz (14.5 Gbps effective).
Being a dual-slot card, the NVIDIA A40 PCIe draws power from an 8-pin EPS power connector, with power draw rated at 300 W maximum. Display outputs include: 3x DisplayPort 1.4a. A40 PCIe is connected to the rest of the system using a PCI-Express 4.0 x16 interface. The card measures 267 mm in length, 111 mm in width, and features a dual-slot cooling solution.
Being a dual-slot card, the NVIDIA A40 PCIe draws power from an 8-pin EPS power connector, with power draw rated at 300 W maximum. Display outputs include: 3x DisplayPort 1.4a. A40 PCIe is connected to the rest of the system using a PCI-Express 4.0 x16 interface. The card measures 267 mm in length, 111 mm in width, and features a dual-slot cooling solution.
Graphics Processor
Graphics Card
- Release Date
- Oct 5th, 2020
- Availability
- 2021
- Generation
-
Tesla Ampere
(Axx)
- Predecessor
- Tesla Turing
- Successor
- Tesla Ada
- Production
- Active
- Bus Interface
- PCIe 4.0 x16
Relative Performance
Based on TPU review data: "Performance Summary" at 1920x1080, 4K for 2080 Ti and faster.
Performance estimated based on architecture, shader count and clocks.
Clock Speeds
- Base Clock
- 1305 MHz
- Boost Clock
- 1740 MHz
- Memory Clock
-
1812 MHz
14.5 Gbps effective
Memory
- Memory Size
- 48 GB
- Memory Type
- GDDR6
- Memory Bus
- 384 bit
- Bandwidth
- 695.8 GB/s
Render Config
- Shading Units
- 10752
- TMUs
- 336
- ROPs
- 112
- SM Count
- 84
- Tensor Cores
- 336
- RT Cores
- 84
- L1 Cache
- 128 KB (per SM)
- L2 Cache
- 6 MB
Theoretical Performance
- Pixel Rate
- 194.9 GPixel/s
- Texture Rate
- 584.6 GTexel/s
- FP16 (half)
- 37.42 TFLOPS (1:1)
- FP32 (float)
- 37.42 TFLOPS
- FP64 (double)
- 584.6 GFLOPS (1:64)
Board Design
- Slot Width
- Dual-slot
- Length
- 267 mm
10.5 inches
- Width
- 111 mm
4.4 inches
- TDP
- 300 W
- Suggested PSU
- 700 W
- Outputs
- 3x DisplayPort 1.4a
- Power Connectors
- 8-pin EPS
- Board Number
- PG133 SKU 200
Graphics Features
- DirectX
- 12 Ultimate (12_2)
- OpenGL
- 4.6
- OpenCL
- 3.0
- Vulkan
- 1.3
- CUDA
- 8.6
- Shader Model
- 6.8
GA102 GPU Notes
| Ray Tracing Cores: 2nd Gen Tensor Cores: 3rd Gen NVENC: 7th Gen NVDEC: 5th Gen PureVideo HD: VP11 VDPAU: Feature Set K Latest Drivers: Windows 7 / 8 / 8.1 (x64): GeForce Release 474.89 Windows 10 / 11 (x64): GeForce Release: Latest Quadro Release: Latest Data Center Release: Latest |
Retail boards based on this design (4)
| Name | GPU Clock | Boost Clock | Memory Clock | Other Changes |
|---|---|---|---|---|
| 1305 MHz | 1740 MHz | 1812 MHz | 24 GB | |
| 1305 MHz | 1740 MHz | 1812 MHz | 2 GB | |
| 1305 MHz | 1740 MHz | 1812 MHz | 4 GB | |
| 1305 MHz | 1740 MHz | 1812 MHz | 6 GB |
Sep 12th, 2024 14:00 EDT
change timezone
Latest GPU Drivers
New Forum Posts
- Next Gen GPU's will be even more expensive (245)
- TPU's Nostalgic Hardware Club (18958)
- G.Skill or Corsair (38)
- which versions of 3080 are known to have coil whine? (23)
- Technical Issues - TPU Main Site & Forum (2024) (226)
- EK seems to be having major issues (297)
- Apple Vision Pro seems very dissapointing (46)
- Free Games Thread (4044)
- Can you help me make a comprehensive efficiency graph for CPUs? (40)
- getting errors installing on 24h2 IoT Enterprise (8)
Popular Reviews
- ASUS GeForce RTX 4090 Matrix Platinum Review - The RTX 4090 Ti
- ASUS GeForce RTX 4090 STRIX OC Review
- MSI GeForce RTX 4090 Gaming X Trio Review
- ADATA Legend 970 Pro 2 TB Review
- MSI GeForce RTX 4090 Suprim X Review
- MOONDROP COSMO Open-Back Planar Magnetic Headphones Review
- MSI GeForce RTX 4090 Suprim Liquid X Review
- Gigabyte GeForce RTX 4090 Gaming OC Review
- NVIDIA GeForce RTX 4090 PCI-Express Scaling with Core i9-13900K
- Colorful GeForce RTX 4090 Vulcan OC-V Review
Controversial News Posts
- AMD Confirms Retreat from the Enthusiast GPU Segment, to Focus on Gaining Market-Share (241)
- AnandTech Shuts Down, an Icon of Tech News and Reviews Rides into the Sunset (151)
- AMD Ryzen Branch Prediction Optimizations Now Available to Windows 11 23H2 (130)
- Sony Reveals the PlayStation 5 Pro, Launches November 7th (129)
- "Sinkclose" Vulnerability Affects Every AMD CPU Dating Back to 2006 (120)
- AMD Ryzen 5 7600X3D Launched in the US as a MicroCenter-exclusive for $300, Part of a Bundle (116)
- Report: Intel Could Spin Out Foundry Business or Cancel Some Expansion Plans to Control Losses (113)
- AMD Readies Ryzen 5 5500X3D Socket AM4 Processor with 3D V-Cache (112)