Product Compare
NVIDIA GeForce GTX 1650 TU106 vs NVIDIA H200 NVL
| GPU | GPU | |
|---|---|---|
| Graphics Card | ||
| Release Date | Jun 2020 | Nov 2024 |
| Generation | GeForce 16 | Server Hopper |
| Type | Desktop | AI GPU |
| Bus Interface | PCIe 3.0 x16 | PCIe 5.0 x16 |
| Clock Speeds | ||
| Base Clock | 1410 MHz | 1365 MHz |
| Boost Clock | 1590 MHz | 1785 MHz |
| Memory Clock | 1500 MHz | 1593 MHz |
| Memory | ||
| Memory Size | 4GB | 141GB |
| Memory Type | GDDR6 | HBM3e |
| Memory Bus | 128bit | 6144bit |
| Bandwidth | 192.0GB/s | 4.89TB/s |
| Render Config | ||
| SM Count | 14 | 132 |
| Shading Units | 896 | 16896 |
| TMUs | 56 | 528 |
| ROPs | 32 | 24 |
| L1 Cache | 64 KB (per SM) | 256 KB (per SM) |
| L2 Cache | 1024 KB | 50 MB |
| Tensor Cores | 528 | |
| Theoretical Performance | ||
| Pixel Rate | 50.88 GPixel/s | 42.84 GPixel/s |
| Texture Rate | 89.04 GTexel/s | 942.5 GTexel/s |
| FP16 (half) | 5.699 TFLOPS | 241.3 TFLOPS |
| FP32 (float) | 2.849 TFLOPS | 60.32 TFLOPS |
| FP64 (double) | 89.04 GFLOPS | 30.16 TFLOPS |
| Graphics Processor | ||
| GPU Name | TU106 | GH100 |
| GPU Variant | TU106-125-KAB-A1 | |
| Architecture | Turing | Hopper |
| Foundry | TSMC | TSMC |
| Process Size | 12 nm | 5 nm |
| Transistors | 10.8 billion | 80 billion |
| Die Size | 445 mm² | 814 mm² |
| Board Design | ||
| TDP | 90W | 600W |
| Suggested PSU | 250 W | 1000 W |
| Outputs | 1x DVI1x HDMI 2.01x DisplayPort 1.4a | No outputs |
| Power Connectors | 1x 6-pin | 8-pin EPS |
| Graphics Features | ||
| DirectX | 12 Ultimate (12_2) | N/A |
| OpenGL | 4.6 | N/A |
| OpenCL | 3.0 | 3.0 |
| Vulkan | 1.3 | N/A |
| CUDA | 7.5 | 9.0 |
| Shader Model | 6.6 | N/A |