Quick/Comparisons
12 | Gaming | 7 |
21 | Workstation | 13 |
11 | AI/ML |
Compare with something else?
Compare GeForce GTX 970 with another →8 |
10 | Energy Efficiency | 93 |
15 | Quick Comparison Final Score | 23 |
| GM204 | Graphics Processor | Phoenix |
| 1664 | Cores | 256 |
| 104 / 56 | TMUs / ROPs | 16 / 8 |
| 286.89 | Blender GPU | 97.23 |
| 32 | Forza Horizon 5 | 14 |
| 35 | The Witcher 3 | 12 |
| 57 | Counter-Strike 2 | 20 |
| 39 | Far Cry 6 | 17 |
| 18 | Hogwarts Legacy | - |
| 31 | Call of Duty | 19 |
| 17 | Ghost of Tsushima | 10 |
| 24 | Cyberpunk 2077 | 10 |
| 44 | Tomb Raider | 18 |
| 33 | Average FPS | 14 |
| 44 | 1080p High | 17 |
| 33 | 1080p Ultra | 14 |
| 23 | 1440p Ultra | 10 |
| 12 | 4K Ultra | - |
| Low | Margin of Error | High |
| 28989 | GB6 Compute Score | 12774 |
| 77.1 img/sec | Background Blur | 31.6 img/sec |
| 38.1 img/sec | Face Detection | 15.3 img/sec |
| 1.28 Gpixels/sec | Horizon Detection | 0.44 Gpixels/sec |
| 2.08 Gpixels/sec | Edge Detection | 0.56 Gpixels/sec |
| 1.85 Gpixels/sec | Gaussian Blur | 0.47 Gpixels/sec |
| 0.17 Gpixels/sec | Feature Matching | 0.17 Gpixels/sec |
| 92.9 Gpixels/sec | Stereo Matching | 47 Gpixels/sec |
| 2443 FPS | Particle Physics | 1815.3 FPS |
| OpenCL | API | OpenCL |
| 9639 | G3D Mark Score | 2981 |
| 763 | G2D Mark | 786 |
| 71 FPS | DirectX 11 | 25 FPS |
| 41 FPS | DirectX 12 | 16 FPS |
| 4080 Ops/s | GPU Compute | 1643 Ops/s |
| 6791 | GB6 ML Single Precision | 4462 |
| 6738 | GB6 ML Half Precision | 5298 |
| 5128 | GB6 ML Quantized | 3419 |
| 1852 | Image Classification (SP) | 1608 |
| 3622 | Image Segmentation (HP) | 2046 |
| 8987 | Image Super Resolution (Q) | 5074 |
| 7074 | Face Detection (HP) | 5435 |
| 43726 | Pose Estimation (Q) | 16992 |
| 1446 | Text Classification (SP) | 1469 |
| 1576 | Machine Translation (HP) | 1922 |
| 2425 | Object Detection (SP) | 1516 |
| 12902 | Depth Estimation (Q) | 7572 |
| 86714 | Style Transfer (SP) | 30182 |
| ONNX | Framework | ONNX |
| DirectML | Backend | DirectML |
| 66 GPixel/s | Pixel Fill Rate | 20-22 GPixel/s |
| 123 GTexel/s | Texture Fill Rate | 40-45 GTexel/s |
| 3.9 TFLOPS | FLOPS (FP32) | 1.3-1.4 TFLOPS |
| GDDR5 | Memory Type | System Shared |
| 4 GB | Memory Size | — |
| 1753 MHz | Memory Clock | — |
| 7000 Mbps | Effective Memory Speed | — |
| 256-bit | Bus | — |
| No | ECC | No |
| 224.4 GB/s | Memory Bandwidth | — |
| PCIe 3.0 x16 | Interface | PCIe 4.0 x8 |
| 148 W | TGP | 15 W |
| TSMC | Manufacturing | TSMC |
| 28 nm | Fabrication Process | 4 nm |
| 398 mm² | Die Size | 178 mm² |
| 5 billion | Transistor Count | 25.4 billion |
| 12.56 MTr/mm² | Transistor Density | 142.7 MTr/mm² |
| 12 | DirectX | 12 |
| 1.3 | Vulkan | 1.3 |
| 4.6 | OpenGL | 4.6 |
| 3 | OpenCL | 2.1 |
| 5.2 | CUDA | — |
| No | Ray Tracing | Yes |
| No | DLSS | No |
| 1.2 | DisplayPort | 2.1 |
| Nvidia | Vendor | Amd |
| Discrete | Build | Integrated |
| September 19, 2014 | Released | January 4, 2023 |
| Desktop | Case | Laptop |
| Gaming | Purpose | Generic |
| Mid-range | Segment | Entry-level |
| Maxwell 2.0 | Architecture | RDNA 3.0 |
| GM204 | GPU Codename | Phoenix |
| — | Recommended CPU | -AMD Ryzen 5 8540Uor above |
| 3577 | Steel Nomad Lite Score | 1405 |
| 3666 | Time Spy | 1613 |
| — | Solar Bay | 4770 |
| 4353 | Port Royal | 722 |
| 12102 | Fire Strike | 5043 |
| 6972 | Wild Life Extreme | 3115 |
| 47591 | Night Raid | 21243 |
| 1050 MHz | Base Clock | 800 MHz |
| 1178 MHz | Boost Clock | 2500-2800 MHz (configurable) |
| 1664 | Shading Units | 256 |
| 104 | Texture Mapping Units (TMUs) | 16 |
| 56 | Render Output Units (ROPs) | 8 |
| 13 | Compute Units (Pipelines) | 4 |
| No | Tensor Cores | — |
| No | Ray-tracing Cores | 4 |
| 48KB per cluster | L1 Cache | 128KB per cluster |
| 2MB shared | L2 Cache | 2MB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |