GeForce RTX 4060 Ti 8GB vs 3070 Ti
VS
We compared two discrete desktop gaming GPUs: the GeForce RTX 4060 Ti 8GB with 34 pipelines and 4352 shaders against the 1 year and 11 months older RTX 3070 Ti that utilizes 48 pipelines and 6144 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 RTX 4060 Ti 8GB
- Manufactured using a more efficient 5 nm process technology
- Supports Nvidia DLSS 3 technology
Reasons to consider the GeForce RTX 3070 Ti
- Performs slightly better (up to 13%) in 3DMark Steel Nomad Lite
- Shows 10% higher average frame rate in modern games at QHD resolution – 95 vs 86 FPS
- Has 2.1x higher memory bandwidth: 608.3 vs 288 GB/s
- Features 56 more tensor cores for effective ML and AI workloads
- Has 41% more shading units (6144 vs 4352)
Gaming Performance
Frame rate comparison across popular AAA titles at different resolutionsGames
| Forza Horizon 5 | 150 97 84 61 |
157 101 93 66 |
| The Witcher 3 | 167 137 98 52 |
177 149 104 57 |
| Counter-Strike 2 | 239 187 136 76 |
255 203 142 83 |
| Far Cry 6 | 129 119 95 54 |
136 127 106 61 |
| Hogwarts Legacy | 97 80 59 34 |
104 85 69 38 |
| CoD: Modern Warfare III | 121 112 80 53 |
129 121 92 58 |
| Ghost of Tsushima | 86 74 56 33 |
96 78 62 34 |
| Cyberpunk 2077 | 102 91 57 26 |
113 100 65 32 |
| Shadow of the Tomb Raider | 164 146 109 58 |
176 167 124 67 |
| 1080p High | 139 | 149 |
| 1080p Ultra | 116 | 126 |
| 1440p Ultra | 86 | 95 |
| 4K Ultra | 50 | 55 |
| Margin of Error | Low | Low |
Benchmarks
Graphics cards’ performance in recent benchmarking apps3D Mark
Steel Nomad Lite Score
| Time Spy | 13479 | 14881 |
| Solar Bay | 62428 | 69901 |
| Port Royal | 8103 | 8884 |
| Fire Strike | 34401 | 37242 |
| Wild Life Extreme | 25599 | 31912 |
| Night Raid | 136255 | 134068 |
GeekBench 6 OpenCL
GB6 Compute Score
| Background Blur | 198.3 img/sec | 197 img/sec |
| Face Detection | 126.2 img/sec | 128.1 img/sec |
| Horizon Detection | 4.35 Gpixels/sec | 6.56 Gpixels/sec |
| Edge Detection | 5.69 Gpixels/sec | 8.81 Gpixels/sec |
| Gaussian Blur | 6.72 Gpixels/sec | 6.47 Gpixels/sec |
| Feature Matching | 1.53 Gpixels/sec | 1.26 Gpixels/sec |
| Stereo Matching | 656.3 Gpixels/sec | 609.3 Gpixels/sec |
| Particle Physics | 19149.2 FPS | 18205 FPS |
| API | OpenCL | OpenCL |
Cinebench 2024 GPU
Cinebench 2024 GPU
Passmark Graphics
G3D Mark Score
| G2D Mark | 1095 | 1057 |
| DirectX 11 | 202 FPS | 192 FPS |
| DirectX 12 | 86 FPS | 90 FPS |
| GPU Compute | 12016 Ops/s | 11384 Ops/s |
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) | 9359 | 8615 |
| Image Segmentation (HP) | 21463 | 24659 |
| Image Super Resolution (Q) | 26023 | 20631 |
| Face Detection (HP) | 42533 | 43834 |
| Pose Estimation (Q) | 99698 | 98347 |
| Text Classification (SP) | 3181 | 3162 |
| Machine Translation (HP) | 4956 | 4911 |
| Object Detection (SP) | 11009 | 11498 |
| Depth Estimation (Q) | 43518 | 34719 |
| Style Transfer (SP) | 203667 | 196833 |
| Framework | ONNX | ONNX |
| Backend | DirectML | DirectML |
Specifications
Technical specifications of GeForce RTX 4060 Ti 8GB and 3070 TiGeneral
| Vendor | Nvidia | Nvidia |
| Build | Discrete | Discrete |
| Released | May 18, 2023 | June 10, 2021 |
| Launch price (MSRP) | $399 | $599 |
| Case | Desktop | Desktop |
| Purpose | Gaming | Gaming |
| Segment | Mid-range | Mid-range |
| Architecture | Ada Lovelace | Ampere |
| GPU Codename | AD106 | GA104 |
| Rival Equivalent | - Radeon RX 7700 XT | - Radeon RX 7800 XT |
| Successor | - GeForce RTX 5060 Ti (8GB) | - GeForce RTX 5070 Ti |
| Recommended CPU | - Intel Core i5 14600K or above | - Intel Core i7 12700K or above |
Desktop GPU rating (42nd and 36th place)
Graphics Processing Unit
| Base Clock | 2310 MHz | 1575 MHz |
| Boost Clock | 2535 MHz | 1770 MHz |
| Shading Units | 4352 | 6144 |
| Texture Mapping Units (TMUs) | 136 | 192 |
| Render Output Units (ROPs) | 48 | 96 |
| Compute Units (Pipelines) | 34 | 48 |
| Tensor Cores | 136 | 192 |
| Ray-tracing Cores | 34 | 48 |
| L1 Cache | 128KB per cluster | 128KB per cluster |
| L2 Cache | 32MB shared | 4MB shared |
| Instructions Per Cycle | 2 IPC | 2 IPC |
Raw Performance
| Pixel Fill Rate | 122 GPixel/s | 170 GPixel/s |
| Texture Fill Rate | 345 GTexel/s | 340 GTexel/s |
FLOPS (FP32)
Physical
| Interface | PCIe 4.0 x8 | PCIe 4.0 x16 |
| TGP | 160 W | 290 W |
| Manufacturing | TSMC | Samsung |
| Fabrication Process | 5 nm | 8 nm |
| Die Size | 188 mm² | 392 mm² |
| Transistor Count | 22 billion | 17 billion |
| Transistor Density | 117.02 MTr/mm² | 43.37 MTr/mm² |
Memory
| Memory Type | GDDR6 | GDDR6X |
| Memory Size | 8 GB | 8 GB |
| Memory Clock | 2250 MHz | 1188 MHz |
| Effective Memory Speed | 18000 Mbps | 19000 Mbps |
| Bus | 128-bit | 256-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 | 8.9 | 8.6 |
| Ray Tracing | Yes | Yes |
| DLSS | DLSS 3 | DLSS 2 |
| DisplayPort | 1.4a | 1.4a |
Cast your vote
Total votes: 171