Quick/Comparisons
The GeForce RTX 5080 takes the lead in gaming performance, boasting a 7-point advantage over the GeForce RTX 4080.
The GeForce RTX 5080 outperforms the GeForce RTX 4080 in content creation, with a 14-point difference in workstation performance.
Although the GeForce RTX 5080 has a slight edge, the score gap is only 4 points, making it a close call, and we consider it a tie.
80 | Gaming | 87 |
71 | Workstation | 85 |
66 | AI/ML |
Compare with something else?
Compare GeForce RTX 4080 with another →70 |
43 | Energy Efficiency | 49 |
75 | Quick Comparison Final Score | 84 |
| AD103 | Graphics Processor | GB203-400 |
| 9728 | Cores | 10752 |
| 304 / 112 | TMUs / ROPs | 336 / 112 |
| 8710.73 | Blender GPU | 9138.54 |
| — | Forza Horizon 5 | 185 |
| — | The Witcher 3 | 237 |
| — | Counter-Strike 2 | 237 |
| — | Far Cry 6 | 176 |
| — | Hogwarts Legacy | 154 |
| — | Call of Duty | 241 |
| — | Ghost of Tsushima | 137 |
| — | Cyberpunk 2077 | 190 |
| — | Tomb Raider | 270 |
| — | Average FPS | 203 |
| — | 1080p High | 239 |
| — | 1080p Ultra | 203 |
| — | 1440p Ultra | 164 |
| — | 4K Ultra | 97 |
| — | Margin of Error | Low |
| 217712 | GB6 Compute Score | 297747 |
| 250.4 img/sec | Background Blur | 319.6 img/sec |
| 190 img/sec | Face Detection | 225 img/sec |
| 9.56 Gpixels/sec | Horizon Detection | 13.6 Gpixels/sec |
| 12.8 Gpixels/sec | Edge Detection | 19.5 Gpixels/sec |
| 15.3 Gpixels/sec | Gaussian Blur | 20.4 Gpixels/sec |
| 1.97 Gpixels/sec | Feature Matching | 2.33 Gpixels/sec |
| 1130 Gpixels/sec | Stereo Matching | 1750 Gpixels/sec |
| 28795.5 FPS | Particle Physics | 43676.6 FPS |
| OpenCL | API | OpenCL |
| 34430 | G3D Mark Score | 35695 |
| 1250 | G2D Mark | 1407 |
| 313 FPS | DirectX 11 | 327 FPS |
| 132 FPS | DirectX 12 | 144 FPS |
| 20406 Ops/s | GPU Compute | 19397 Ops/s |
| 36799 | GB6 ML Single Precision | 38994 |
| 53331 | GB6 ML Half Precision | 66609 |
| 28104 | GB6 ML Quantized | 29206 |
| 13466 | Image Classification (SP) | 15441 |
| 34767 | Image Segmentation (HP) | 47173 |
| 42687 | Image Super Resolution (Q) | 43223 |
| 70927 | Face Detection (HP) | 92863 |
| 239350 | Pose Estimation (Q) | 202202 |
| 3606 | Text Classification (SP) | 3701 |
| 6519 | Machine Translation (HP) | 7905 |
| 18153 | Object Detection (SP) | 20955 |
| 65667 | Depth Estimation (Q) | 69068 |
| 414848 | Style Transfer (SP) | 349205 |
| ONNX | Framework | ONNX |
| DirectML | Backend | DirectML |
| 281 GPixel/s | Pixel Fill Rate | 293 GPixel/s |
| 762 GTexel/s | Texture Fill Rate | 879 GTexel/s |
| 48.7 TFLOPS | FLOPS (FP32) | 56.3 TFLOPS |
| GDDR6X | Memory Type | GDDR7 |
| 16 GB | Memory Size | 16 GB |
| 1400 MHz | Memory Clock | 1875 MHz |
| 22400 Mbps | Effective Memory Speed | 30000 Mbps |
| 256-bit | Bus | 256-bit |
| No | ECC | No |
| 716.8 GB/s | Memory Bandwidth | 960 GB/s |
| PCIe 4.0 x16 | Interface | PCIe 5.0 x16 |
| 320 W | TGP | 360 W |
| TSMC | Manufacturing | TSMC |
| 5 nm | Fabrication Process | 4 nm |
| 379 mm² | Die Size | 378 mm² |
| 45 billion | Transistor Count | 45.6 billion |
| 118.73 MTr/mm² | Transistor Density | 120.63 MTr/mm² |
| 12 | DirectX | 12.2 |
| 1.3 | Vulkan | 1.4 |
| 4.6 | OpenGL | 4.6 |
| 3 | OpenCL | 3 |
| 8.9 | CUDA | 12 |
| Yes | Ray Tracing | Yes |
| DLSS 3 | DLSS | DLSS 4 |
| 1.4a | DisplayPort | 2.1b |
| Nvidia | Vendor | Nvidia |
| Discrete | Build | Discrete |
| September 20, 2022 | Released | January 7, 2025 |
| Desktop | Case | Desktop |
| Gaming | Purpose | Gaming |
| High-end | Segment | High-end |
| Ada Lovelace | Architecture | Blackwell 2.0 |
| AD103 | GPU Codename | GB203-400 |
| -Intel Core i7 14700Kor above | Recommended CPU | -Intel Core Ultra 7 265Kor above |
| 30172 | Steel Nomad Lite Score | 36200 |
| 28161 | Time Spy | 32989 |
| 136931 | Solar Bay | 160901 |
| 17982 | Port Royal | 23257 |
| 64633 | Fire Strike | 75518 |
| 59111 | Wild Life Extreme | 69544 |
| 184963 | Night Raid | 200077 |
| 2205 MHz | Base Clock | 2295 MHz |
| 2505 MHz | Boost Clock | 2617 MHz |
| 9728 | Shading Units | 10752 |
| 304 | Texture Mapping Units (TMUs) | 336 |
| 112 | Render Output Units (ROPs) | 112 |
| 76 | Compute Units (Pipelines) | 84 |
| 304 | Tensor Cores | 336 |
| 76 | Ray-tracing Cores | 84 |
| 128KB per cluster | L1 Cache | 128KB per cluster |
| 64MB shared | L2 Cache | 64MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |