The L40S is a professional graphics card by NVIDIA, launched on October 13th, 2022. Built on the 5 nm process, and based on the AD102 graphics processor, in its AD102-???-A1 variant, the card supports DirectX 12 Ultimate. The AD102 graphics processor is a large chip with a die area of 609 mm² and 76,300 million transistors. Unlike the fully unlocked TITAN Ada, which uses the same GPU but has all 18432 shaders enabled, NVIDIA has disabled some shading units on the L40S to reach the product's target shader count. It features 18176 shading units, 568 texture mapping units, and 192 ROPs. Also included are 568 tensor cores which help improve the speed of machine learning applications. The card also has 142 raytracing acceleration cores. NVIDIA has paired 48 GB GDDR6 memory with the L40S, which are connected using a 384-bit memory interface. The GPU is operating at a frequency of 1110 MHz, which can be boosted up to 2520 MHz, memory is running at 2250 MHz (18 Gbps effective).
Being a dual-slot card, the NVIDIA L40S draws power from 1x 16-pin power connector, with power draw rated at 300 W maximum. Display outputs include: 1x HDMI 2.1, 3x DisplayPort 1.4a. L40S 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
- GPU Name
- AD102
- GPU Variant
- AD102-???-A1
- Architecture
- Ada Lovelace
- Foundry
- TSMC
- Process Size
- 5 nm
- Transistors
- 76,300 million
- Density
- 125.3M / mm²
- Die Size
- 609 mm²
Clock Speeds
- Base Clock
- 1110 MHz
- Boost Clock
- 2520 MHz
- Memory Clock
- 2250 MHz
18 Gbps effective
Memory
- Memory Size
- 48 GB
- Memory Type
- GDDR6
- Memory Bus
- 384 bit
- Bandwidth
- 864.0 GB/s
Render Config
- Shading Units 18176
- TMUs 568
- ROPs 192
- SM Count 142
- Tensor Cores 568
- RT Cores 142
- L1 Cache 128 KB (per SM)
- L2 Cache 48 MB
Theoretical Performance
- Pixel Rate
- 483.8 GPixel/s
- Texture Rate
- 1,431 GTexel/s
- FP16 (half)
- 91.61 TFLOPS (1:1)
- FP32 (float)
- 91.61 TFLOPS
- FP64 (double)
- 1,431 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
- 1x HDMI 2.1
3x DisplayPort 1.4a
- Power Connectors 1x 16-pin
- Board Number PG133 SKU 250
Graphics Features
- DirectX 12 Ultimate (12_2)
- OpenGL 4.6
- OpenCL 3.0
- Vulkan 1.3
- CUDA 8.9
- Shader Model 6.7
Card Notes
GeForce Now RTX 4080 simulator |
AD102 GPU Notes
Ray Tracing Cores: 3rd Gen Tensor Cores: 4th Gen NVENC: 8th Gen NVDEC: 5th Gen PureVideo HD: VP12 VDPAU: Feature Set L |