Quick/Comparisons
18 | Gaming | 9 |
26 | Workstation | 8 |
| — | AI/ML |
Compare with something else?
Compare GeForce GTX 1070 with another →37 | Energy Efficiency | 73 |
| — | Quick Comparison Final Score | 19 |
| GP104 | Graphics Processor | GP108 |
| 2048 | Cores | 256 |
| 128 / 64 | TMUs / ROPs | 16 / 16 |
| 512.01 | Blender GPU | 41.33 |
| 48 | Forza Horizon 5 | — |
| 59 | The Witcher 3 | — |
| 88 | Counter-Strike 2 | — |
| 60 | Far Cry 6 | — |
| 36 | Hogwarts Legacy | — |
| 41 | Call of Duty | — |
| 29 | Ghost of Tsushima | — |
| 39 | Cyberpunk 2077 | — |
| 67 | Tomb Raider | — |
| 52 | Average FPS | — |
| 66 | 1080p High | — |
| 52 | 1080p Ultra | — |
| 37 | 1440p Ultra | — |
| 19 | 4K Ultra | — |
| Low | Margin of Error | — |
| 48520 | GB6 Compute Score | 7785 |
| 134.6 img/sec | Background Blur | 19.4 img/sec |
| 61.4 img/sec | Face Detection | 9.09 img/sec |
| 1.98 Gpixels/sec | Horizon Detection | 0.29 Gpixels/sec |
| 3.02 Gpixels/sec | Edge Detection | 0.43 Gpixels/sec |
| 3.44 Gpixels/sec | Gaussian Blur | 0.44 Gpixels/sec |
| 0.27 Gpixels/sec | Feature Matching | 0.08 Gpixels/sec |
| 159.9 Gpixels/sec | Stereo Matching | 25 Gpixels/sec |
| 4611.7 FPS | Particle Physics | 759.2 FPS |
| OpenCL | API | OpenCL |
| 10465 | G3D Mark Score | 1818 |
| 603 | G2D Mark | 213 |
| 92 FPS | DirectX 11 | 15 FPS |
| 34 FPS | DirectX 12 | 9 FPS |
| 4591 Ops/s | GPU Compute | 762 Ops/s |
| — | GB6 ML Single Precision | 1408 |
| — | GB6 ML Half Precision | 1384 |
| — | GB6 ML Quantized | 1028 |
| — | Image Classification (SP) | 277 |
| — | Image Segmentation (HP) | 726 |
| — | Image Super Resolution (Q) | 1787 |
| — | Face Detection (HP) | 1076 |
| — | Pose Estimation (Q) | 5990 |
| — | Text Classification (SP) | 743 |
| — | Machine Translation (HP) | 970 |
| — | Object Detection (SP) | 379 |
| — | Depth Estimation (Q) | 2059 |
| — | Style Transfer (SP) | 11915 |
| — | Framework | ONNX |
| — | Backend | DirectML |
| 105 GPixel/s | Pixel Fill Rate | 24 GPixel/s |
| 211 GTexel/s | Texture Fill Rate | 24 GTexel/s |
| 6.7 TFLOPS | FLOPS (FP32) | 0.8 TFLOPS |
| GDDR5 | Memory Type | GDDR5 |
| 8 GB | Memory Size | 2 GB |
| 2002 MHz | Memory Clock | 1502 MHz |
| 8008 Mbps | Effective Memory Speed | 6000 Mbps |
| 256-bit | Bus | 64-bit |
| No | ECC | — |
| 256.3 GB/s | Memory Bandwidth | 48 GB/s |
| PCIe 3.0 x16 | Interface | PCIe 3.0 x4 |
| 120 W | TGP | 10 W |
| TSMC | Manufacturing | Samsung |
| 16 nm | Fabrication Process | 14 nm |
| 314 mm² | Die Size | 74 mm² |
| 7.2 billion | Transistor Count | 1.8 billion |
| 22.93 MTr/mm² | Transistor Density | 24.32 MTr/mm² |
| 12 | DirectX | 12 |
| 1.3 | Vulkan | 1.2 |
| 4.6 | OpenGL | 4.6 |
| 1.2 | OpenCL | 1.2 |
| 6.1 | CUDA | 6.1 |
| No | Ray Tracing | No |
| No | DLSS | No |
| — | DisplayPort | 1.4 |
| Nvidia | Vendor | Nvidia |
| Discrete | Build | Discrete |
| August 15, 2016 | Released | February 21, 2019 |
| Laptop | Case | Laptop |
| Gaming | Purpose | Gaming |
| High-end | Segment | Entry-level |
| Pascal | Architecture | Pascal |
| GP104 | GPU Codename | GP108 |
| -Intel Core i7 8750Hor above | Recommended CPU | -Intel Core i7 1065G7or above |
| 5473 | Steel Nomad Lite Score | — |
| 5540 | Time Spy | — |
| 14681 | Solar Bay | — |
| 1207 | Port Royal | — |
| 17005 | Fire Strike | — |
| 11402 | Wild Life Extreme | — |
| 56792 | Night Raid | — |
| 1442 MHz | Base Clock | 1519 MHz |
| 1645 MHz | Boost Clock | 1531 MHz |
| 2048 | Shading Units | 256 |
| 128 | Texture Mapping Units (TMUs) | 16 |
| 64 | Render Output Units (ROPs) | 16 |
| 16 | Compute Units (Pipelines) | 2 |
| No | Tensor Cores | No |
| No | Ray-tracing Cores | No |
| 48KB per cluster | L1 Cache | 48KB per cluster |
| 2MB shared | L2 Cache | 512KB shared |
| 2 IPC | Instructions Per Cycle | 2 IPC |