GeForce GTX 970 vs GTX 1060
Nvidia GeForce GTX 970
GM204
Nvidia GeForce GTX 1060
GP106
We compared two discrete desktop gaming GPUs: the GeForce GTX 970 4 GB with 13 pipelines and 1664 shaders against the 1 year and 10 months newer GTX 1060 6 GB that utilizes 10 pipelines and 1280 shaders. Here you will find complete details about specs, efficiency, performance tests, and more.
Review
Gaming
Performance in DirectX, OpenCL, and Vulkan games
Workstation
Perf. in 3D modeling, video editing and rendering apps
AI/ML
Capabilities for machine learning and AI-related tasks
Energy Efficiency
Power consumption efficiency in different scenarios
NanoReview Final Score
Overall video card score
Value for money (Beta)
Key differences
Reasons to consider the GeForce GTX 970
- Has 30% more shading units (1664 vs 1280)
Reasons to consider the GeForce GTX 1060
- Performs slightly better (up to 14%) in 3DMark Steel Nomad Lite
- Manufactured using a more efficient 16 nm process technology
- Includes 2 GB more video memory
- Achieves 35% more points in the GeekBench 6 Compute test (39K vs 28K)
Gaming Performance
Frame rate comparison across popular AAA titles at different resolutionsGames
| Forza Horizon 5 | 58 33 26 16 |
68 37 29 16 |
| The Witcher 3 | 46 35 22 11 |
54 40 26 13 |
| Counter-Strike 2 | 77 56 37 16 |
96 63 42 18 |
| Far Cry 6 | 47 43 29 13 |
50 45 33 15 |
| Hogwarts Legacy | 25 22 15 - |
32 22 18 9 |
| CoD: Modern Warfare III | 38 32 21 15 |
38 32 22 13 |
| Ghost of Tsushima | 24 16 14 - |
29 20 16 9 |
| Cyberpunk 2077 | 27 21 13 - |
30 26 19 - |
| Shadow of the Tomb Raider | 49 44 27 14 |
57 49 34 18 |
| 1080p High | 43 | 50 |
| 1080p Ultra | 34 | 37 |
| 1440p Ultra | 23 | 27 |
| 4K Ultra | 12 | 13 |
| Margin of Error | Low | Low |
Benchmarks
Graphics cards’ performance in recent benchmarking apps3D Mark
Steel Nomad Lite Score
| Time Spy | 3665 | 4203 |
| Solar Bay | - | 11278 |
| Port Royal | 4353 | 844 |
| Fire Strike | 12101 | 13132 |
| Wild Life Extreme | 6976 | 8510 |
| Night Raid | 47583 | 54686 |
GeekBench 6 OpenCL
GB6 Compute Score
| Background Blur | 77.1 img/sec | 100.2 img/sec |
| Face Detection | 38.1 img/sec | 48.1 img/sec |
| Horizon Detection | 1.28 Gpixels/sec | 1.54 Gpixels/sec |
| Edge Detection | 2.08 Gpixels/sec | 2.25 Gpixels/sec |
| Gaussian Blur | 1.85 Gpixels/sec | 2.58 Gpixels/sec |
| Feature Matching | 0.17 Gpixels/sec | 0.25 Gpixels/sec |
| Stereo Matching | 92.9 Gpixels/sec | 136.9 Gpixels/sec |
| Particle Physics | 2443 FPS | 4220.5 FPS |
| API | OpenCL | OpenCL |
Cinebench 2024 GPU
Cinebench 2024 GPU
Passmark Graphics
G3D Mark Score
| G2D Mark | 763 | 737 |
| DirectX 11 | 71 FPS | 72 FPS |
| DirectX 12 | 41 FPS | 37 FPS |
| GPU Compute | 4079 Ops/s | 4246 Ops/s |
Blender
Blender GPU
Artificial Intelligence Tests
Performance in machine learning and artificial intelligence tasksGeekBench 6 ML
GB6 ML Single Precision
GB6 ML Half Precision
GB6 ML Quantized
| Image Classification (SP) | 1852 | 1462 |
| Image Segmentation (HP) | 3622 | 3872 |
| Image Super Resolution (Q) | 8987 | 8136 |
| Face Detection (HP) | 7074 | 6356 |
| Pose Estimation (Q) | 43726 | 33181 |
| Text Classification (SP) | 1446 | 1806 |
| Machine Translation (HP) | 1576 | 3149 |
| Object Detection (SP) | 2425 | 2096 |
| Depth Estimation (Q) | 12902 | 10515 |
| Style Transfer (SP) | 86714 | 67831 |
| Framework | ONNX | ONNX |
| Backend | DirectML | DirectML |
Specifications
Technical specifications of GeForce GTX 970 and GTX 1060General
| Vendor | Nvidia | Nvidia |
| Build | Discrete | Discrete |
| Released | September 19, 2014 | July 19, 2016 |
| Launch price (MSRP) | $329 | $299 |
| Case | Desktop | Desktop |
| Purpose | Gaming | Gaming |
| Segment | Mid-range | Mid-range |
| Architecture | Maxwell 2.0 | Pascal |
| GPU Codename | GM204 | GP106 |
| Successor | - GeForce RTX 5070 | - GeForce RTX 5060 |
| Recommended CPU | - | - Intel Core i7 7700 or above |
Desktop GPU rating (86th and 83rd place)
Graphics Processing Unit
| Base Clock | 1050 MHz | 1506 MHz |
| Boost Clock | 1178 MHz | 1709 MHz |
| Shading Units | 1664 | 1280 |
| Texture Mapping Units (TMUs) | 104 | 80 |
| Render Output Units (ROPs) | 56 | 48 |
| Compute Units (Pipelines) | 13 | 10 |
| Tensor Cores | No | No |
| Ray-tracing Cores | No | No |
| L1 Cache | 48KB per cluster | 48KB per cluster |
| L2 Cache | 2MB shared | 1536KB shared |
| Instructions Per Cycle | 2 IPC | 2 IPC |
Raw Performance
| Pixel Fill Rate | 66 GPixel/s | 82 GPixel/s |
| Texture Fill Rate | 123 GTexel/s | 137 GTexel/s |
FLOPS (FP32)
Physical
| Interface | PCIe 3.0 x16 | PCIe 3.0 x16 |
| TGP | 148 W | 120 W |
| Manufacturing | TSMC | TSMC |
| Fabrication Process | 28 nm | 16 nm |
| Die Size | 398 mm² | 200 mm² |
| Transistor Count | 5 billion | 4 billion |
| Transistor Density | 12.56 MTr/mm² | 20 MTr/mm² |
Memory
| Memory Type | GDDR5 | GDDR5 |
| Memory Size | 4 GB | 6 GB |
| Memory Clock | 1753 MHz | 2002 MHz |
| Effective Memory Speed | 7000 Mbps | 8000 Mbps |
| Bus | 256-bit | 192-bit |
| ECC | No | No |
Memory Bandwidth
API
| DirectX | 12 | 12 |
| Vulkan | 1.3 | 1.3 |
| OpenGL | 4.6 | 4.6 |
| OpenCL | 3.0 | 3.0 |
| CUDA | 5.2 | 6.1 |
| Ray Tracing | No | No |
| DLSS | No | No |
| DisplayPort | 1.2 | 1.4a |
Further details
Notes on GTX 1060:
- There is also 3GB memory version, which is about 10% slower.
Cast your vote
Total votes: 35