Product Compare
NVIDIA GeForce RTX 4090 vs AMD Radeon AI PRO R9700S
| GPU | GPU | |
|---|---|---|
| Graphics Card | ||
| Release Date | Sep 2022 | Dec 2025 |
| Generation | GeForce 40 | Radeon Pro Navi |
| Type | Desktop | Desktop |
| Bus Interface | PCIe 4.0 x16 | PCIe 5.0 x16 |
| Clock Speeds | ||
| Base Clock | 2235 MHz | 1660 MHz |
| Boost Clock | 2520 MHz | 2920 MHz |
| Memory Clock | 1313 MHz | 2518 MHz |
| Memory | ||
| Memory Size | 24GB | 32GB |
| Memory Type | GDDR6X | GDDR6 |
| Memory Bus | 384bit | 256bit |
| Bandwidth | 1.01TB/s | 644.6GB/s |
| Render Config | ||
| SM Count | 128 | |
| Shading Units | 16384 | 4096 |
| TMUs | 512 | 256 |
| ROPs | 176 | 128 |
| Tensor Cores | 512 | |
| RT Cores | 128 | 64 |
| L1 Cache | 128 KB (per SM) | |
| L2 Cache | 72 MB | 8 MB |
| Compute Units | 64 | |
| L3 Cache | 64 MB | |
| Theoretical Performance | ||
| Pixel Rate | 443.5 GPixel/s | 373.8 GPixel/s |
| Texture Rate | 1290.2 GTexel/s | 747.5 GTexel/s |
| FP16 (half) | 82.58 TFLOPS | 95.68 TFLOPS |
| FP32 (float) | 82.58 TFLOPS | 47.84 TFLOPS |
| FP64 (double) | 1290.2 GFLOPS | 1495.0 GFLOPS |
| Graphics Processor | ||
| GPU Name | AD102 | Navi 48 |
| GPU Variant | AD102-300-A1 | Navi 48 XTW |
| Architecture | Ada Lovelace | RDNA 4.0 |
| Foundry | TSMC | TSMC |
| Process Size | 5 nm | 4 nm |
| Transistors | 76.3 billion | 53.9 billion |
| Die Size | 609 mm² | 357 mm² |
| Board Design | ||
| TDP | 450W | 300W |
| Suggested PSU | 850 W | 700 W |
| Outputs | 1x HDMI 2.13x DisplayPort 1.4a | 4x DisplayPort 2.1a |
| Power Connectors | 1x 16-pin | 1x 16-pin |
| Graphics Features | ||
| DirectX | 12 Ultimate (12_2) | 12 Ultimate (12_2) |
| OpenGL | 4.6 | 4.6 |
| OpenCL | 3.0 | 2.2 |
| Vulkan | 1.4 | 1.4 |
| CUDA | 8.9 | |
| Shader Model | 6.8 | 6.8 |