Product Compare
NVIDIA P102 101 vs NVIDIA L40 CNX
| GPU | GPU | |
|---|---|---|
| Graphics Card | ||
| Release Date | Jan 2018 | Oct 2022 |
| Generation | Mining GPUs | Tesla Ada |
| Type | Desktop | Professional |
| Bus Interface | PCIe 3.0 x4 | PCIe 4.0 x16 |
| Clock Speeds | ||
| Base Clock | 1557 MHz | 1005 MHz |
| Boost Clock | 1670 MHz | 2475 MHz |
| Memory Clock | 2002 MHz | 2250 MHz |
| Memory | ||
| Memory Size | 10GB | 24GB |
| Memory Type | GDDR5 | GDDR6 |
| Memory Bus | 320bit | 384bit |
| Bandwidth | 320.3GB/s | 864.0GB/s |
| Render Config | ||
| SM Count | 25 | 142 |
| Shading Units | 3200 | 18176 |
| TMUs | 200 | 568 |
| ROPs | 80 | 192 |
| L1 Cache | 48 KB (per SM) | 128 KB (per SM) |
| L2 Cache | 0 MB | 48 MB |
| Tensor Cores | 568 | |
| RT Cores | 142 | |
| Theoretical Performance | ||
| Pixel Rate | 133.6 GPixel/s | 475.2 GPixel/s |
| Texture Rate | 334.0 GTexel/s | 1406 GTexel/s |
| FP16 (half) | 167.0 GFLOPS | 89.97 TFLOPS |
| FP32 (float) | 10.69 TFLOPS | 89.97 TFLOPS |
| FP64 (double) | 334.0 GFLOPS | 1406 GFLOPS |
| Graphics Processor | ||
| GPU Name | GP102 | AD102 |
| GPU Variant | GP102-101-A1 | AD102-???-A1 |
| Architecture | Pascal | Ada Lovelace |
| Foundry | TSMC | TSMC |
| Process Size | 16 nm | 4 nm |
| Transistors | 11.8 billion | 76.3 billion |
| Die Size | 471 mm² | 608 mm² |
| Board Design | ||
| TDP | 250W | 300W |
| Suggested PSU | 600 W | 700 W |
| Outputs | No outputs | 1x HDMI 2.13x DisplayPort 1.4a |
| Power Connectors | 2x 8-pin | 1x 16-pin |
| Graphics Features | ||
| DirectX | 12 (12_1) | 12 Ultimate (12_2) |
| OpenGL | 4.6 | 4.6 |
| OpenCL | 3.0 | 3.0 |
| Vulkan | 1.3 | 1.3 |
| CUDA | 6.1 | 8.9 |
| Shader Model | 6.4 | 6.6 |