GeForce RTX 4070 vs RTX 3080
VS
Nvidia GeForce RTX 4070
AD104
Nvidia GeForce RTX 3080
GA102
We compared two discrete desktop gaming GPUs: the GeForce RTX 4070 12 GB with 46 pipelines and 5888 shaders against the 2 years and 8 months older RTX RTX 3080 10 GB that utilizes 68 pipelines and 8704 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
- Manufactured using a more efficient 5 nm process technology
- Includes 2 GB more video memory
- Supports Nvidia DLSS 3 technology
Reasons to consider the GeForce RTX 3080
- Features 88 more tensor cores for effective ML and AI workloads
- Has 51% higher memory bandwidth: 760.3 vs 504 GB/s
- Has 48% more shading units (8704 vs 5888)
Gaming Performance
Frame rate comparison across popular AAA titles at different resolutionsGames
| Forza Horizon 5 | 172 115 102 79 |
170 114 104 80 |
| The Witcher 3 | 194 159 114 61 |
189 165 116 63 |
| Counter-Strike 2 | 272 205 149 81 |
274 206 149 82 |
| Far Cry 6 | 149 133 115 68 |
145 138 114 72 |
| Hogwarts Legacy | 117 97 77 39 |
119 98 78 43 |
| CoD: Modern Warfare III | 149 145 107 69 |
152 145 110 73 |
| Ghost of Tsushima | 106 88 72 43 |
109 89 74 44 |
| Cyberpunk 2077 | 129 119 75 37 |
135 117 79 36 |
| Shadow of the Tomb Raider | 211 190 145 80 |
210 195 151 82 |
| 1080p High | 167 | 167 |
| 1080p Ultra | 139 | 141 |
| 1440p Ultra | 106 | 108 |
| 4K Ultra | 62 | 64 |
| Margin of Error | Low | Low |
Benchmarks
Graphics cards’ performance in recent benchmarking apps3D Mark
Steel Nomad Lite Score
| Time Spy | 17799 | 17664 |
| Solar Bay | 82861 | 85898 |
| Port Royal | 11156 | 11597 |
| Fire Strike | 44067 | 42472 |
| Wild Life Extreme | 35472 | 37900 |
| Night Raid | 153385 | 133004 |
GeekBench 6 OpenCL
GB6 Compute Score
| Background Blur | 233 img/sec | 217.3 img/sec |
| Face Detection | 141.7 img/sec | 141.9 img/sec |
| Horizon Detection | 6.62 Gpixels/sec | 8.19 Gpixels/sec |
| Edge Detection | 9.19 Gpixels/sec | 11 Gpixels/sec |
| Gaussian Blur | 9.2 Gpixels/sec | 8.43 Gpixels/sec |
| Feature Matching | 1.78 Gpixels/sec | 1.29 Gpixels/sec |
| Stereo Matching | 845.4 Gpixels/sec | 780.7 Gpixels/sec |
| Particle Physics | 25043.9 FPS | 22538.5 FPS |
| API | OpenCL | OpenCL |
Cinebench 2024 GPU
Cinebench 2024 GPU
Passmark Graphics
G3D Mark Score
| G2D Mark | 1165 | 1062 |
| DirectX 11 | 243 FPS | 207 FPS |
| DirectX 12 | 103 FPS | 99 FPS |
| GPU Compute | 14570 Ops/s | 14058 Ops/s |
Blender
Blender GPU
Recent User Tests
GeForce RTX 4070
| Date | Benchmark | Result |
|---|---|---|
| 📘 2025-12-14 (kashitsu) | Cinebench 2024 | 18051 |
GeForce RTX 3080
| Date | Benchmark | Result |
|---|---|---|
| 📘 2026-08-06 (Henry) | Cinebench 2024 | 50695 |
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 | 9439 |
| Image Segmentation (HP) | 26153 | 25471 |
| Image Super Resolution (Q) | 32054 | 19892 |
| Face Detection (HP) | 51337 | 46956 |
| Pose Estimation (Q) | 124617 | 104704 |
| Text Classification (SP) | 3703 | 2712 |
| Machine Translation (HP) | 5484 | 4828 |
| Object Detection (SP) | 13134 | 12153 |
| Depth Estimation (Q) | 50548 | 36822 |
| Style Transfer (SP) | 263477 | 246005 |
| Framework | ONNX | ONNX |
| Backend | DirectML | DirectML |
Specifications
Technical specifications of GeForce RTX 4070 and RTX 3080General
| Vendor | Nvidia | Nvidia |
| Build | Discrete | Discrete |
| Released | April 13, 2023 | September 1, 2020 |
| Launch price (MSRP) | $599 | $699 |
| Case | Desktop | Desktop |
| Purpose | Gaming | Gaming |
| Segment | Mid-range | High-end |
| Architecture | Ada Lovelace | Ampere |
| GPU Codename | AD104 | GA102 |
| Rival Equivalent | - Radeon RX 7900 XT | - Radeon RX 6800 XT |
| Successor | - GeForce RTX 5070 | - GeForce RTX 5080 |
| Recommended CPU | - Intel Core i7 14700K or above | - Intel Core i7 12700K or above |
Desktop GPU rating (28th and 29th place)
Graphics Processing Unit
| Base Clock | 1920 MHz | 1440 MHz |
| Boost Clock | 2475 MHz | 1710 MHz |
| Shading Units | 5888 | 8704 |
| Texture Mapping Units (TMUs) | 184 | 272 |
| Render Output Units (ROPs) | 64 | 96 |
| Compute Units (Pipelines) | 46 | 68 |
| Tensor Cores | 184 | 272 |
| Ray-tracing Cores | 46 | 68 |
| L1 Cache | 128KB per cluster | 128KB per cluster |
| L2 Cache | 36MB shared | 5MB shared |
| Instructions Per Cycle | 2 IPC | 2 IPC |
Raw Performance
| Pixel Fill Rate | 158 GPixel/s | 164 GPixel/s |
| Texture Fill Rate | 455 GTexel/s | 465 GTexel/s |
FLOPS (FP32)
Physical
| Interface | PCIe 4.0 x16 | PCIe 4.0 x16 |
| TGP | 200 W | 320 W |
| Manufacturing | TSMC | Samsung |
| Fabrication Process | 5 nm | 8 nm |
| Die Size | 294 mm² | 628 mm² |
| Transistor Count | 35 billion | 28 billion |
| Transistor Density | 119.05 MTr/mm² | 44.59 MTr/mm² |
Memory
| Memory Type | GDDR6X | GDDR6X |
| Memory Size | 12 GB | 10 GB |
| Memory Clock | 1313 MHz | 1188 MHz |
| Effective Memory Speed | 21000 Mbps | 19000 Mbps |
| Bus | 192-bit | 320-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: 360