Quick/Comparisons
The GeForce RTX 4080 SUPER takes the lead in gaming performance, thanks to its higher score of 78, outpacing the Radeon RX 7900 GRE's 64.
The GeForce RTX 4080 SUPER excels in content creation with a score of 76, surpassing the Radeon RX 7900 GRE's 60, indicating better performance in this area.
The GeForce RTX 4080 SUPER leads in AI and machine learning with a score of 66, outperforming the Radeon RX 7900 GRE's 59, thanks to its more efficient architecture.
78 | Gaming | 64 |
76 | Workstation | 60 |
66 | AI/ML | 59 |
46 | Energy Efficiency | 44 |
75 | Quick Comparison Final Score | 62 |
| AD103 | Graphics Processor | Navi 31 |
| 10240 | Cores | 5120 |
| 320 / 112 | TMUs / ROPs | 320 / 160 |
| 8420.48 | Blender GPU | 2808.96 |
| 171 | Forza Horizon 5 | 143 |
| 210 | The Witcher 3 | 201 |
| 223 | Counter-Strike 2 | 217 |
| 166 | Far Cry 6 | 155 |
| 137 | Hogwarts Legacy | 109 |
| 216 | Call of Duty | 190 |
| 122 | Ghost of Tsushima | 116 |
| 177 | Cyberpunk 2077 | 138 |
| 265 | Tomb Raider | 221 |
| 187 | Average FPS | 166 |
| 221 | 1080p High | 198 |
| 187 | 1080p Ultra | 166 |
| 149 | 1440p Ultra | 128 |
| 87 | 4K Ultra | 74 |
| Low | Margin of Error | Low |
| 250691 | GB6 Compute Score | 172224 |
| 279.5 img/sec | Background Blur | 352.4 img/sec |
| 220.1 img/sec | Face Detection | 229 img/sec |
| 11 Gpixels/sec | Horizon Detection | 6.42 Gpixels/sec |
| 15.2 Gpixels/sec | Edge Detection | 8.86 Gpixels/sec |
| 16.2 Gpixels/sec | Gaussian Blur | 7.89 Gpixels/sec |
| 2.08 Gpixels/sec | Feature Matching | 1.5 Gpixels/sec |
| 1370 Gpixels/sec | Stereo Matching | 662.8 Gpixels/sec |
| 37199.7 FPS | Particle Physics | 24012 FPS |
| OpenCL | API | OpenCL |
| 34252 | G3D Mark Score | 27367 |
| 1278 | G2D Mark | 1207 |
| 300 FPS | DirectX 11 | 300 FPS |
| 134 FPS | DirectX 12 | 110 FPS |
| 19639 Ops/s | GPU Compute | 15171 Ops/s |
| 36883 | GB6 ML Single Precision | 34533 |
| 54286 | GB6 ML Half Precision | 42224 |
| 28474 | GB6 ML Quantized | 26568 |
| 12453 | Image Classification (SP) | 15095 |
| 34467 | Image Segmentation (HP) | 21206 |
| 43534 | Image Super Resolution (Q) | 33962 |
| 70300 | Face Detection (HP) | 57614 |
| 242331 | Pose Estimation (Q) | 214748 |
| 3767 | Text Classification (SP) | 4325 |
| 6614 | Machine Translation (HP) | 6598 |
| 18301 | Object Detection (SP) | 16608 |
| 66477 | Depth Estimation (Q) | 57731 |
| 417906 | Style Transfer (SP) | 394987 |
| ONNX | Framework | ONNX |
| DirectML | Backend | DirectML |
| 286 GPixel/s | Pixel Fill Rate | 359 GPixel/s |
| 816 GTexel/s | Texture Fill Rate | 718 GTexel/s |
| 52.2 TFLOPS | FLOPS (FP32) | 46 TFLOPS |
| GDDR6X | Memory Type | GDDR6 |
| 16 GB | Memory Size | 16 GB |
| 1438 MHz | Memory Clock | 2250 MHz |
| 23000 Mbps | Effective Memory Speed | 18000 Mbps |
| 256-bit | Bus | 256-bit |
| No | ECC | No |
| 736.3 GB/s | Memory Bandwidth | 576 GB/s |
| PCIe 4.0 x16 | Interface | PCIe 4.0 x16 |
| 320 W | TGP | 260 W |
| TSMC | Manufacturing | TSMC |
| 5 nm | Fabrication Process | 5 nm |
| 379 mm² | Die Size | 529 mm² |
| 45 billion | Transistor Count | 57 billion |
| 118.73 MTr/mm² | Transistor Density | 107.75 MTr/mm² |
| 12 | DirectX | 12 |
| 1.3 | Vulkan | 1.3 |
| 4.6 | OpenGL | 4.6 |
| 3 | OpenCL | 2.2 |
| 8.9 | CUDA | — |
| Yes | Ray Tracing | Yes |
| DLSS 3 | DLSS | No |
| 1.4a | DisplayPort | 2.1 |
| Nvidia | Vendor | Amd |
| Discrete | Build | Discrete |
| January 31, 2024 | Released | — |
| Desktop | Case | Desktop |
| Gaming | Purpose | Gaming |
| High-end | Segment | High-end |
| Ada Lovelace | Architecture | RDNA 3.0 |
| AD103 | GPU Codename | Navi 31 |
| -Intel Core i7 14700Kor above | Recommended CPU | — |
| 30711 | Steel Nomad Lite Score | 21359 |
| 28279 | Time Spy | 22537 |
| 138998 | Solar Bay | 90757 |
| 18348 | Port Royal | 12461 |
| 63714 | Fire Strike | 56212 |
| 60015 | Wild Life Extreme | 41740 |
| 180412 | Night Raid | 169038 |
| 2295 MHz | Base Clock | 1287 MHz |
| 2550 MHz | Boost Clock | 2245 MHz |
| 10240 | Shading Units | 5120 |
| 320 | Texture Mapping Units (TMUs) | 320 |
| 112 | Render Output Units (ROPs) | 160 |
| 80 | Compute Units (Pipelines) | — |
| 320 | Tensor Cores | — |
| 80 | Ray-tracing Cores | 80 |
| 128KB per cluster | L1 Cache | 256KB per cluster |
| 64MB shared | L2 Cache | 6MB shared |
| 2 IPC | Instructions Per Cycle | 4 IPC |
Compare with something else?
Compare GeForce RTX 4080 with another →