Product Compare
NVIDIA GeForce RTX 4070 Ti SUPER AD102 vs NVIDIA B200
| GPU | GPU | |
|---|---|---|
| Graphics Card | ||
| Release Date | Jun 2024 | Jan 2024 |
| Generation | GeForce 40 | Server Blackwell |
| Type | Desktop | Desktop |
| Bus Interface | PCIe 4.0 x16 | PCIe 5.0 x16 |
| Clock Speeds | ||
| Base Clock | 2340 MHz | 700 MHz |
| Boost Clock | 2610 MHz | 1965 MHz |
| Memory Clock | 1313 MHz | 2000 MHz |
| Memory | ||
| Memory Size | 16GB | 90GB |
| Memory Type | GDDR6X | HBM3e |
| Memory Bus | 256bit | 4096bit |
| Bandwidth | 672.3GB/s | 4.10TB/s |
| Render Config | ||
| SM Count | 66 | 148 |
| Shading Units | 8448 | 18944 |
| TMUs | 264 | 592 |
| ROPs | 96 | 24 |
| Tensor Cores | 264 | 592 |
| RT Cores | 66 | |
| L1 Cache | 128 KB (per SM) | 256 KB (per SM) |
| L2 Cache | 48 MB | 50 MB |
| Theoretical Performance | ||
| Pixel Rate | 250.6 GPixel/s | 47.16 GPixel/s |
| Texture Rate | 689.0 GTexel/s | 1163.3 GTexel/s |
| FP16 (half) | 44.10 TFLOPS | 1191.2 TFLOPS |
| FP32 (float) | 44.10 TFLOPS | 74.45 TFLOPS |
| FP64 (double) | 689.0 GFLOPS | 37.22 TFLOPS |
| Graphics Processor | ||
| GPU Name | AD102 | GB100 |
| GPU Variant | AD102-175-KEF-A1 | |
| Architecture | Ada Lovelace | Blackwell |
| Foundry | TSMC | TSMC |
| Process Size | 5 nm | 5 nm |
| Transistors | 76.3 billion | 104 billion |
| Die Size | 609 mm² | Unknown |
| Board Design | ||
| TDP | 285W | 1000W |
| Suggested PSU | 600 W | 1400 W |
| Outputs | 1x HDMI 2.13x DisplayPort 1.4a | No outputs |
| Power Connectors | 1x 16-pin | |
| 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 | 8.9 | 10.0 |
| Shader Model | 6.8 | N/A |