GeForce RTX 4070 vs 2080 SUPER
VS
Nvidia GeForce RTX 4070
AD104
We compared two discrete desktop gaming GPUs: the GeForce RTX 4070 12 GB with 46 pipelines and 5888 shaders against the 3 years and 10 months older RTX 2080 SUPER 8 GB that utilizes 48 pipelines and 3072 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 4070
- Performs significantly better (up to 56%) in 3DMark Steel Nomad Lite
- 2.6x higher maximum theoretical performance (29.1 vs 11.2 TFLOPS)
- Manufactured using a more efficient 5 nm process technology
- Shows 41% higher average frame rate in modern games at QHD resolution – 106 vs 75 FPS
- Includes 4 GB more video memory
- Supports Nvidia DLSS 3 technology
- Achieves 36% more points in the GeekBench 6 Compute test (167K vs 122K)
- Has 92% more shading units (5888 vs 3072)
Reasons to consider the GeForce RTX 2080 SUPER
- Features 200 more tensor cores for effective ML and AI workloads
Gaming Performance
Frame rate comparison across popular AAA titles at different resolutionsGames
| Forza Horizon 5 | 172 115 102 79 |
134 85 72 51 |
| The Witcher 3 | 194 159 114 61 |
148 123 90 50 |
| Counter-Strike 2 | 272 205 149 81 |
223 169 122 64 |
| Far Cry 6 | 149 133 115 68 |
115 103 82 47 |
| Hogwarts Legacy | 117 97 77 39 |
87 69 53 26 |
| CoD: Modern Warfare III | 149 145 107 69 |
108 97 69 45 |
| Ghost of Tsushima | 106 88 72 43 |
77 64 48 28 |
| Cyberpunk 2077 | 129 119 75 37 |
89 82 49 24 |
| Shadow of the Tomb Raider | 211 190 145 80 |
140 130 94 48 |
| 1080p High | 167 | 125 |
| 1080p Ultra | 139 | 102 |
| 1440p Ultra | 106 | 75 |
| 4K Ultra | 62 | 43 |
| Margin of Error | Low | Low |
Benchmarks
Graphics cards’ performance in recent benchmarking apps3D Mark
Steel Nomad Lite Score
| Time Spy | 17799 | 11621 |
| Solar Bay | 82861 | 52087 |
| Port Royal | 11156 | 7001 |
| Fire Strike | 44067 | 28789 |
| Wild Life Extreme | 35472 | 23794 |
| Night Raid | 153385 | 108237 |
GeekBench 6 OpenCL
GB6 Compute Score
| Background Blur | 233 img/sec | 197.1 img/sec |
| Face Detection | 141.7 img/sec | 110.7 img/sec |
| Horizon Detection | 6.62 Gpixels/sec | 5.03 Gpixels/sec |
| Edge Detection | 9.19 Gpixels/sec | 6.5 Gpixels/sec |
| Gaussian Blur | 9.2 Gpixels/sec | 7.05 Gpixels/sec |
| Feature Matching | 1.78 Gpixels/sec | 1.28 Gpixels/sec |
| Stereo Matching | 845.4 Gpixels/sec | 576.8 Gpixels/sec |
| Particle Physics | 25043.9 FPS | 15732.9 FPS |
| API | OpenCL | OpenCL |
Cinebench 2024 GPU
Cinebench 2024 GPU
Passmark Graphics
G3D Mark Score
| G2D Mark | 1165 | 917 |
| DirectX 11 | 243 FPS | 168 FPS |
| DirectX 12 | 103 FPS | 75 FPS |
| GPU Compute | 14561 Ops/s | 8159 Ops/s |
Blender
Blender GPU
Recent User Tests
GeForce RTX 4070
| Date | Benchmark | Result |
|---|---|---|
| 📘 2025-12-14 (kashitsu) | Cinebench 2024 | 18051 |
GeForce RTX 2080 SUPER
No benchmark results yet
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) | 10246 | 6780 |
| Image Segmentation (HP) | 26153 | 15029 |
| Image Super Resolution (Q) | 32054 | 15410 |
| Face Detection (HP) | 51337 | 30849 |
| Pose Estimation (Q) | 124617 | 71857 |
| Text Classification (SP) | 3703 | 1983 |
| Machine Translation (HP) | 5484 | 3417 |
| Object Detection (SP) | 13134 | 8214 |
| Depth Estimation (Q) | 50548 | 26815 |
| Style Transfer (SP) | 263477 | 148497 |
| Framework | ONNX | ONNX |
| Backend | DirectML | DirectML |
Specifications
Technical specifications of GeForce RTX 4070 and 2080 SUPERGeneral
| Vendor | Nvidia | Nvidia |
| Build | Discrete | Discrete |
| Released | April 13, 2023 | July 23, 2019 |
| Launch price (MSRP) | $599 | $699 |
| Case | Desktop | Desktop |
| Purpose | Gaming | Gaming |
| Segment | Mid-range | High-end |
| Architecture | Ada Lovelace | Turing |
| GPU Codename | AD104 | TU104 |
| Rival Equivalent | - Radeon RX 7900 XT | - |
| Successor | - GeForce RTX 5070 | - |
| Recommended CPU | - Intel Core i7 14700K or above | - Intel Core i9 9900K or above |
Desktop GPU rating (30th and 48th place)
Graphics Processing Unit
| Base Clock | 1920 MHz | 1650 MHz |
| Boost Clock | 2475 MHz | 1815 MHz |
| Shading Units | 5888 | 3072 |
| Texture Mapping Units (TMUs) | 184 | 192 |
| Render Output Units (ROPs) | 64 | 64 |
| Compute Units (Pipelines) | 46 | 48 |
| Tensor Cores | 184 | 384 |
| Ray-tracing Cores | 46 | 48 |
| L1 Cache | 128KB per cluster | 64KB per cluster |
| L2 Cache | 36MB shared | 4MB shared |
| Instructions Per Cycle | 2 IPC | 2 IPC |
Raw Performance
| Pixel Fill Rate | 158 GPixel/s | 116 GPixel/s |
| Texture Fill Rate | 455 GTexel/s | 348 GTexel/s |
FLOPS (FP32)
Physical
| Interface | PCIe 4.0 x16 | PCIe 3.0 x16 |
| TGP | 200 W | 250 W |
| Manufacturing | TSMC | TSMC |
| Fabrication Process | 5 nm | 12 nm |
| Die Size | 294 mm² | 545 mm² |
| Transistor Count | 35 billion | 13 billion |
| Transistor Density | 119.05 MTr/mm² | 23.85 MTr/mm² |
Memory
| Memory Type | GDDR6X | GDDR6 |
| Memory Size | 12 GB | 8 GB |
| Memory Clock | 1313 MHz | 1937 MHz |
| Effective Memory Speed | 21000 Mbps | 15500 Mbps |
| Bus | 192-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 | 7.5 |
| Ray Tracing | Yes | Yes |
| DLSS | DLSS 3 | DLSS 2 |
| DisplayPort | 1.4a | 1.4a |
Cast your vote
Total votes: 5