Quick/Comparisons
12 | Gaming | 43 |
21 | Workstation | 53 |
11 | AI/ML |
Compare with something else?
Compare GeForce GTX 970 with another →39 |
10 | Energy Efficiency | 56 |
15 | Quick Comparison Final Score | 47 |
| GM204 | Graphics Processor | AD104 |
| 1664 | Cores | 5120 |
| 104 / 56 | TMUs / ROPs | 160 / 64 |
| 286.89 | Blender GPU | 4377.08 |
| 32 | Forza Horizon 5 | 93 |
| 35 | The Witcher 3 | 134 |
| 57 | Counter-Strike 2 | 191 |
| 39 | Far Cry 6 | 113 |
| 18 | Hogwarts Legacy | 80 |
| 31 | Call of Duty | 111 |
| 17 | Ghost of Tsushima | 72 |
| 24 | Cyberpunk 2077 | 89 |
| 44 | Tomb Raider | 145 |
| 33 | Average FPS | 114 |
| 44 | 1080p High | 138 |
| 33 | 1080p Ultra | 114 |
| 23 | 1440p Ultra | 84 |
| 12 | 4K Ultra | 49 |
| Low | Margin of Error | High |
| 28989 | GB6 Compute Score | 131037 |
| 77.1 img/sec | Background Blur | 223 img/sec |
| 38.1 img/sec | Face Detection | 145.8 img/sec |
| 1.28 Gpixels/sec | Horizon Detection | 5.87 Gpixels/sec |
| 2.08 Gpixels/sec | Edge Detection | 7.77 Gpixels/sec |
| 1.85 Gpixels/sec | Gaussian Blur | 7.2 Gpixels/sec |
| 0.17 Gpixels/sec | Feature Matching | 1.39 Gpixels/sec |
| 92.9 Gpixels/sec | Stereo Matching | 458.7 Gpixels/sec |
| 2443 FPS | Particle Physics | 14313 FPS |
| OpenCL | API | OpenCL |
| 9639 | G3D Mark Score | 19719 |
| 763 | G2D Mark | 712 |
| 71 FPS | DirectX 11 | 187 FPS |
| 41 FPS | DirectX 12 | 83 FPS |
| 4080 Ops/s | GPU Compute | 7623 Ops/s |
| 6791 | GB6 ML Single Precision | 20783 |
| 6738 | GB6 ML Half Precision | 33871 |
| 5128 | GB6 ML Quantized | 16429 |
| 1852 | Image Classification (SP) | 8316 |
| 3622 | Image Segmentation (HP) | 17301 |
| 8987 | Image Super Resolution (Q) | 27105 |
| 7074 | Face Detection (HP) | 38189 |
| 43726 | Pose Estimation (Q) | 105500 |
| 1446 | Text Classification (SP) | 3319 |
| 1576 | Machine Translation (HP) | 3845 |
| 2425 | Object Detection (SP) | 10208 |
| 12902 | Depth Estimation (Q) | 38369 |
| 86714 | Style Transfer (SP) | 196359 |
| ONNX | Framework | ONNX |
| DirectML | Backend | DirectML |
| 66 GPixel/s | Pixel Fill Rate | 99-144 GPixel/s |
| 123 GTexel/s | Texture Fill Rate | 247-360 GTexel/s |
| 3.9 TFLOPS | FLOPS (FP32) | 15.8-23 TFLOPS |
| GDDR5 | Memory Type | GDDR6 |
| 4 GB | Memory Size | 12 GB |
| 1753 MHz | Memory Clock | 2250 MHz |
| 7000 Mbps | Effective Memory Speed | 18000 Mbps |
| 256-bit | Bus | 192-bit |
| No | ECC | Yes |
| 224.4 GB/s | Memory Bandwidth | 432 GB/s |
| PCIe 3.0 x16 | Interface | PCIe 4.0 x16 |
| 148 W | TGP | 60-140 W (configurable) |
| TSMC | Manufacturing | TSMC |
| 28 nm | Fabrication Process | 5 nm |
| 398 mm² | Die Size | 294 mm² |
| 5 billion | Transistor Count | 35.8 billion |
| 12.56 MTr/mm² | Transistor Density | 121.77 MTr/mm² |
| 12 | DirectX | 12 |
| 1.3 | Vulkan | 1.3 |
| 4.6 | OpenGL | 4.6 |
| 3 | OpenCL | 3 |
| 5.2 | CUDA | 8.9 |
| No | Ray Tracing | Yes |
| No | DLSS | DLSS 3 |
| 1.2 | DisplayPort | 1.4a |
| Nvidia | Vendor | Nvidia |
| Discrete | Build | Discrete |
| September 19, 2014 | Released | March 21, 2023 |
| Desktop | Case | Laptop |
| Gaming | Purpose | Professional |
| Mid-range | Segment | Mid-range |
| Maxwell 2.0 | Architecture | Ada Lovelace |
| GM204 | GPU Codename | AD104 |
| — | Recommended CPU | -Intel Core Ultra 7 165Hor above |
| 3577 | Steel Nomad Lite Score | 13536 |
| 3666 | Time Spy | 12966 |
| — | Solar Bay | 52474 |
| 4353 | Port Royal | 7976 |
| 12102 | Fire Strike | 32977 |
| 6972 | Wild Life Extreme | 26514 |
| 47591 | Night Raid | 110011 |
| 1050 MHz | Base Clock | — |
| 1178 MHz | Boost Clock | 1545-2250 MHz (configurable) |
| 1664 | Shading Units | 5120 |
| 104 | Texture Mapping Units (TMUs) | 160 |
| 56 | Render Output Units (ROPs) | 64 |
| 13 | Compute Units (Pipelines) | 40 |
| No | Tensor Cores | 160 |
| No | Ray-tracing Cores | 40 |
| 48KB per cluster | L1 Cache | 128KB per cluster |
| 2MB shared | L2 Cache | 48MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |