Product Compare
NVIDIA Jetson AGX Orin 32 GB vs NVIDIA GRID K520Q
| GPU | GPU | |
|---|---|---|
| Graphics Card | ||
| Release Date | Feb 2023 | Jul 2014 |
| Generation | Tegra | GRID |
| Type | Professional | Professional |
| Bus Interface | PCIe 4.0 x4 | PCIe 3.0 x16 |
| Clock Speeds | ||
| Memory Clock | 800 MHz | 1250 MHz |
| Memory | ||
| Memory Size | 32GB | 4GB |
| Memory Type | LPDDR5 | GDDR5 |
| Memory Bus | 256bit | 256bit |
| Bandwidth | 204.8GB/s | 160.0GB/s |
| Render Config | ||
| SM Count | 14 | |
| Shading Units | 1792 | 1536 |
| TMUs | 56 | 128 |
| ROPs | 24 | 32 |
| Tensor Cores | 56 | |
| L1 Cache | 128 KB (per SM) | 16 KB (per SMX) |
| L2 Cache | 6 MB | 512 KB |
| Theoretical Performance | ||
| Pixel Rate | 22.32 GPixel/s | 23.84 GPixel/s |
| Texture Rate | 52.08 GTexel/s | 95.36 GTexel/s |
| FP16 (half) | 6.666 TFLOPS | |
| FP32 (float) | 3.333 TFLOPS | 2.289 TFLOPS |
| FP64 (double) | 1.667 TFLOPS | 95.36 GFLOPS |
| Graphics Processor | ||
| GPU Name | GA10B | GK104 |
| Architecture | Ampere | Kepler |
| Foundry | Samsung | TSMC |
| Process Size | 8 nm | 28 nm |
| Transistors | Unknown | 3.54 billion |
| Die Size | 200 mm² | 294 mm² |
| Board Design | ||
| TDP | 40W | 225W |
| Outputs | Portable Device Dependent | No outputs |
| Suggested PSU | 550 W | |
| Graphics Features | ||
| DirectX | 12 Ultimate (12_2) | 12 (11_0) |
| OpenGL | 4.6 | 4.6 |
| OpenCL | 3.0 | 3.0 |
| Vulkan | 1.4 | 1.1 |
| CUDA | 8.7 | 3.0 |
| Shader Model | 6.8 | 5.1 |