GeForce RTX 3070 Ti vs 2080 Ti
VS
We compared two discrete desktop gaming GPUs: the GeForce RTX 3070 Ti 8 GB with 48 pipelines and 6144 shaders against the 2 years and 9 months older RTX 2080 Ti 11 GB that utilizes 68 pipelines and 4352 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 3070 Ti
- 62% higher maximum theoretical performance (21.7 vs 13.4 TFLOPS)
- Manufactured using a more efficient 8 nm process technology
- Has 41% more shading units (6144 vs 4352)
Reasons to consider the GeForce RTX 2080 Ti
- Includes 3 GB more video memory
- Features 352 more tensor cores for effective ML and AI workloads
Gaming Performance
Frame rate comparison across popular AAA titles at different resolutionsGames
| Forza Horizon 5 | 157 101 93 66 |
150 101 89 65 |
| The Witcher 3 | 177 149 104 57 |
175 142 104 55 |
| Counter-Strike 2 | 255 203 142 83 |
253 193 140 81 |
| Far Cry 6 | 136 127 106 61 |
132 122 101 61 |
| Hogwarts Legacy | 104 85 69 38 |
101 85 65 36 |
| CoD: Modern Warfare III | 129 121 92 58 |
124 117 85 56 |
| Ghost of Tsushima | 96 78 62 34 |
92 79 61 36 |
| Cyberpunk 2077 | 113 100 65 32 |
112 96 63 30 |
| Shadow of the Tomb Raider | 176 167 124 67 |
171 161 120 63 |
| 1080p High | 149 | 146 |
| 1080p Ultra | 126 | 122 |
| 1440p Ultra | 95 | 92 |
| 4K Ultra | 55 | 54 |
| Margin of Error | Low | Low |
Benchmarks
Graphics cards’ performance in recent benchmarking apps3D Mark
Steel Nomad Lite Score
| Time Spy | 14881 | 14655 |
| Solar Bay | 69901 | 64022 |
| Port Royal | 8884 | 9109 |
| Fire Strike | 37242 | 35666 |
| Wild Life Extreme | 31912 | 29561 |
| Night Raid | 134068 | 115542 |
GeekBench 6 OpenCL
GB6 Compute Score
| Background Blur | 197 img/sec | 211 img/sec |
| Face Detection | 128.1 img/sec | 141.3 img/sec |
| Horizon Detection | 6.56 Gpixels/sec | 6.24 Gpixels/sec |
| Edge Detection | 8.81 Gpixels/sec | 8.1 Gpixels/sec |
| Gaussian Blur | 6.47 Gpixels/sec | 9.22 Gpixels/sec |
| Feature Matching | 1.26 Gpixels/sec | 1.46 Gpixels/sec |
| Stereo Matching | 609.3 Gpixels/sec | 627.3 Gpixels/sec |
| Particle Physics | 18205 FPS | 16980.5 FPS |
| API | OpenCL | OpenCL |
Cinebench 2024 GPU
Cinebench 2024 GPU
Passmark Graphics
G3D Mark Score
| G2D Mark | 1057 | 926 |
| DirectX 11 | 192 FPS | 193 FPS |
| DirectX 12 | 90 FPS | 83 FPS |
| GPU Compute | 11384 Ops/s | 9831 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) | 8615 | 8213 |
| Image Segmentation (HP) | 24659 | 19891 |
| Image Super Resolution (Q) | 20631 | 17598 |
| Face Detection (HP) | 43834 | 39785 |
| Pose Estimation (Q) | 98347 | 84003 |
| Text Classification (SP) | 3162 | 2062 |
| Machine Translation (HP) | 4911 | 4192 |
| Object Detection (SP) | 11498 | 10444 |
| Depth Estimation (Q) | 34719 | 31856 |
| Style Transfer (SP) | 196833 | 207773 |
| Framework | ONNX | ONNX |
| Backend | DirectML | DirectML |
Specifications
Technical specifications of GeForce RTX 3070 Ti and 2080 TiGeneral
| Vendor | Nvidia | Nvidia |
| Build | Discrete | Discrete |
| Released | June 10, 2021 | September 27, 2018 |
| Launch price (MSRP) | $599 | $999 |
| Case | Desktop | Desktop |
| Purpose | Gaming | Gaming |
| Segment | Mid-range | High-end |
| Architecture | Ampere | Turing |
| GPU Codename | GA104 | TU102 |
| Rival Equivalent | - Radeon RX 7800 XT | - |
| Successor | - GeForce RTX 5070 Ti | - GeForce RTX 3080 Ti |
| Recommended CPU | - Intel Core i7 12700K or above | - Intel Core i9 9900K or above |
Desktop GPU rating (36th and 40th place)
Graphics Processing Unit
| Base Clock | 1575 MHz | 1350 MHz |
| Boost Clock | 1770 MHz | 1545 MHz |
| Shading Units | 6144 | 4352 |
| Texture Mapping Units (TMUs) | 192 | 272 |
| Render Output Units (ROPs) | 96 | 88 |
| Compute Units (Pipelines) | 48 | 68 |
| Tensor Cores | 192 | 544 |
| Ray-tracing Cores | 48 | 68 |
| L1 Cache | 128KB per cluster | 64KB per cluster |
| L2 Cache | 4MB shared | 5632KB shared |
| Instructions Per Cycle | 2 IPC | 2 IPC |
Raw Performance
| Pixel Fill Rate | 170 GPixel/s | 136 GPixel/s |
| Texture Fill Rate | 340 GTexel/s | 420 GTexel/s |
FLOPS (FP32)
Physical
| Interface | PCIe 4.0 x16 | PCIe 3.0 x16 |
| TGP | 290 W | 250 W |
| Manufacturing | Samsung | TSMC |
| Fabrication Process | 8 nm | 12 nm |
| Die Size | 392 mm² | 754 mm² |
| Transistor Count | 17 billion | 18 billion |
| Transistor Density | 43.37 MTr/mm² | 23.87 MTr/mm² |
Memory
| Memory Type | GDDR6X | GDDR6 |
| Memory Size | 8 GB | 11 GB |
| Memory Clock | 1188 MHz | 1750 MHz |
| Effective Memory Speed | 19000 Mbps | 14000 Mbps |
| Bus | 256-bit | 352-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 | 8.6 | 7.5 |
| Ray Tracing | Yes | Yes |
| DLSS | DLSS 2 | DLSS 2 |
| DisplayPort | 1.4a | 1.4a |
Cast your vote
Total votes: 98