Quick/Comparisons
19 | Gaming | 43 |
27 | Workstation | 44 |
13 | AI/ML |
Compare with something else?
Compare GeForce GTX 980 with another →35 |
1 | Energy Efficiency | 35 |
21 | Quick Comparison Final Score | 42 |
| GM200 | Graphics Processor | GA104 |
| 2816 | Cores | 5888 |
| 176 / 96 | TMUs / ROPs | 184 / 96 |
| 536.2 | Blender GPU | 3368.42 |
| 52 | Forza Horizon 5 | 99 |
| 64 | The Witcher 3 | 143 |
| 95 | Counter-Strike 2 | 194 |
| 63 | Far Cry 6 | 121 |
| 37 | Hogwarts Legacy | 80 |
| 47 | Call of Duty | 113 |
| 33 | Ghost of Tsushima | 74 |
| 44 | Cyberpunk 2077 | 94 |
| 69 | Tomb Raider | 153 |
| 56 | Average FPS | 119 |
| 71 | 1080p High | 141 |
| 56 | 1080p Ultra | 119 |
| 39 | 1440p Ultra | 88 |
| 20 | 4K Ultra | 50 |
| Low | Margin of Error | Low |
| 49950 | GB6 Compute Score | 126195 |
| 134.8 img/sec | Background Blur | 161.6 img/sec |
| 59.8 img/sec | Face Detection | 117.2 img/sec |
| 2.37 Gpixels/sec | Horizon Detection | 5.89 Gpixels/sec |
| 3.79 Gpixels/sec | Edge Detection | 8.05 Gpixels/sec |
| 3.46 Gpixels/sec | Gaussian Blur | 6.65 Gpixels/sec |
| 0.22 Gpixels/sec | Feature Matching | 1.25 Gpixels/sec |
| 191.9 Gpixels/sec | Stereo Matching | 550.6 Gpixels/sec |
| 4183.6 FPS | Particle Physics | 17613.1 FPS |
| OpenCL | API | OpenCL |
| 13642 | G3D Mark Score | 22128 |
| 850 | G2D Mark | 1005 |
| 106 FPS | DirectX 11 | 182 FPS |
| 55 FPS | DirectX 12 | 84 FPS |
| 6100 Ops/s | GPU Compute | 11062 Ops/s |
| 7711 | GB6 ML Single Precision | 19789 |
| 7464 | GB6 ML Half Precision | 33013 |
| 5338 | GB6 ML Quantized | 12443 |
| 3173 | Image Classification (SP) | 7624 |
| 4916 | Image Segmentation (HP) | 23420 |
| 7407 | Image Super Resolution (Q) | 17898 |
| 8201 | Face Detection (HP) | 40535 |
| 27221 | Pose Estimation (Q) | 83624 |
| 1675 | Text Classification (SP) | 2724 |
| 2607 | Machine Translation (HP) | 4481 |
| 4123 | Object Detection (SP) | 10304 |
| 11660 | Depth Estimation (Q) | 30418 |
| 55570 | Style Transfer (SP) | 180089 |
| ONNX | Framework | ONNX |
| DirectML | Backend | DirectML |
| 103 GPixel/s | Pixel Fill Rate | 166 GPixel/s |
| 189 GTexel/s | Texture Fill Rate | 317 GTexel/s |
| 6.1 TFLOPS | FLOPS (FP32) | 20.3 TFLOPS |
| GDDR5 | Memory Type | GDDR6 |
| 6 GB | Memory Size | 8 GB |
| 1753 MHz | Memory Clock | 1750 MHz |
| 7000 Mbps | Effective Memory Speed | 14000 Mbps |
| 384-bit | Bus | 256-bit |
| No | ECC | No |
| 336.6 GB/s | Memory Bandwidth | 448 GB/s |
| PCIe 3.0 x16 | Interface | PCIe 4.0 x16 |
| 250 W | TGP | 220 W |
| TSMC | Manufacturing | Samsung |
| 28 nm | Fabrication Process | 8 nm |
| 601 mm² | Die Size | 392 mm² |
| 8 billion | Transistor Count | 17 billion |
| 13.31 MTr/mm² | Transistor Density | 43.37 MTr/mm² |
| 12 | DirectX | 12 |
| 1.3 | Vulkan | 1.3 |
| 4.6 | OpenGL | 4.6 |
| 3 | OpenCL | 3 |
| 5.2 | CUDA | 8.6 |
| No | Ray Tracing | Yes |
| No | DLSS | DLSS 2 |
| 1.2 | DisplayPort | 1.4a |
| Nvidia | Vendor | Nvidia |
| Discrete | Build | Discrete |
| June 2, 2015 | Released | October 22, 2020 |
| Desktop | Case | Desktop |
| Gaming | Purpose | Gaming |
| High-end | Segment | Mid-range |
| Maxwell 2.0 | Architecture | Ampere |
| GM200 | GPU Codename | GA104 |
| — | Recommended CPU | -Intel Core i7 12700Kor above |
| 5945 | Steel Nomad Lite Score | 14153 |
| 5855 | Time Spy | 13643 |
| — | Solar Bay | 64426 |
| 3250 | Port Royal | 8311 |
| 19504 | Fire Strike | 34054 |
| 11830 | Wild Life Extreme | 28486 |
| 70105 | Night Raid | 122039 |
| 1000 MHz | Base Clock | 1500 MHz |
| 1076 MHz | Boost Clock | 1725 MHz |
| 2816 | Shading Units | 5888 |
| 176 | Texture Mapping Units (TMUs) | 184 |
| 96 | Render Output Units (ROPs) | 96 |
| 22 | Compute Units (Pipelines) | 46 |
| No | Tensor Cores | 184 |
| No | Ray-tracing Cores | 46 |
| 48KB per cluster | L1 Cache | 128KB per cluster |
| 3MB shared | L2 Cache | 4MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |