GeForce RTX 4090 Laptop vs 3070 Ti Laptop
VS
Nvidia GeForce RTX 4090 Laptop
AD103 (GN21-X11)
Nvidia GeForce RTX 3070 Ti Laptop
GA104 (GN20-E6)
We compared two discrete laptop gaming GPUs: the GeForce RTX 4090 Mobile 16 GB with 76 pipelines and 9728 shaders against the RTX 3070 Ti Mobile 8 GB that utilizes 44 pipelines and 5632 shaders. Here you will find complete details about specs, efficiency, performance tests, and more.
Some GPUs come in different configurations. For greater precision, pick a specific config below
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
Key differences
Reasons to consider the GeForce RTX 4090 Laptop
- Performs significantly better (up to 2x) in 3DMark Steel Nomad Lite
- 2.4x higher maximum theoretical performance (39.7 vs 16.7 TFLOPS)
- Shows 66% higher average frame rate in modern games at QHD resolution – 123 vs 74 FPS
- Manufactured using a more efficient 5 nm process technology
Gaming Performance
Frame rate comparison across popular AAA titles at different resolutionsGames
| Forza Horizon 5 | 188 131 122 92 |
134 82 73 49 |
| The Witcher 3 | 211 176 132 67 |
153 120 87 49 |
| Counter-Strike 2 | 277 217 158 82 |
222 165 119 61 |
| Far Cry 6 | 160 144 125 79 |
111 102 81 45 |
| Hogwarts Legacy | 137 111 86 47 |
84 69 51 29 |
| CoD: Modern Warfare III | 177 165 132 84 |
103 97 68 43 |
| Ghost of Tsushima | 122 102 86 52 |
80 64 48 29 |
| Cyberpunk 2077 | 151 135 93 44 |
89 82 47 21 |
| Shadow of the Tomb Raider | 250 235 175 96 |
136 125 90 48 |
| 1080p High | 186 | 124 |
| 1080p Ultra | 157 | 101 |
| 1440p Ultra | 123 | 74 |
| 4K Ultra | 71 | 42 |
| Margin of Error | Low | Low |
Benchmarks
Graphics cards’ performance in recent benchmarking apps3D Mark
Steel Nomad Lite Score
| Time Spy | 21388 | 11329 |
| Solar Bay | 96121 | 49066 |
| Port Royal | 13734 | 6850 |
| Fire Strike | 45207 | 28615 |
| Wild Life Extreme | 42501 | 23076 |
| Night Raid | 147911 | 95157 |
GeekBench 6 OpenCL
GB6 Compute Score
| Background Blur | 264 img/sec | 173.8 img/sec |
| Face Detection | 134.3 img/sec | 96.9 img/sec |
| Horizon Detection | 9.27 Gpixels/sec | 5.69 Gpixels/sec |
| Edge Detection | 12.7 Gpixels/sec | 7.75 Gpixels/sec |
| Gaussian Blur | 13.2 Gpixels/sec | 6.27 Gpixels/sec |
| Feature Matching | 1.92 Gpixels/sec | 1.14 Gpixels/sec |
| Stereo Matching | 1060 Gpixels/sec | 480.4 Gpixels/sec |
| Particle Physics | 29225.1 FPS | 14706.2 FPS |
| API | OpenCL | OpenCL |
Cinebench 2024 GPU
Cinebench 2024 GPU
Passmark Graphics
G3D Mark Score
| G2D Mark | 988 | 768 |
| DirectX 11 | 257 FPS | 150 FPS |
| DirectX 12 | 106 FPS | 71 FPS |
| GPU Compute | 12157 Ops/s | 7276 Ops/s |
Blender
Blender GPU
Recent User Tests
GeForce RTX 4090 Laptop
| Date | Benchmark | Result |
|---|---|---|
| 📘 2026-06-09 (Horikita) | Geekbench 6 OpenCL | 213782 |
| 📘 2025-11-03 (JJ) | Cinebench 2024 | 23277 |
| 📘 2025-11-03 (JJ) | Cinebench 2024 | 22714 |
GeForce RTX 3070 Ti Laptop
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) | 11346 | 6969 |
| Image Segmentation (HP) | 24625 | 17142 |
| Image Super Resolution (Q) | 36874 | 16426 |
| Face Detection (HP) | 50088 | 31628 |
| Pose Estimation (Q) | 174844 | 74413 |
| Text Classification (SP) | 3226 | 2522 |
| Machine Translation (HP) | 4905 | 3639 |
| Object Detection (SP) | 14667 | 9002 |
| Depth Estimation (Q) | 53868 | 27790 |
| Style Transfer (SP) | 340056 | 160005 |
| Framework | ONNX | ONNX |
| Backend | DirectML | DirectML |
Specifications
Technical specifications of GeForce RTX 4090 Laptop and 3070 Ti LaptopGeneral
| Vendor | Nvidia | Nvidia |
| Build | Discrete | Discrete |
| Released | February 8, 2023 | - |
| Case | Laptop | Laptop |
| Purpose | Gaming | Gaming |
| Segment | High-end | High-end |
| Architecture | Ada Lovelace | Ampere |
| GPU Codename | AD103 (GN21-X11) | GA104 (GN20-E6) |
| Successor | - GeForce RTX 5090 Laptop | - |
| Recommended CPU | - Intel Core Ultra 9 185H or above | - Intel Core i7 12700H or above |
Laptop GPU ranking (4th and 40th place)
Graphics Processing Unit
| Base Clock | 930-1620 MHz (configurable) | 510-1035 MHz (configurable) |
| Boost Clock | 1455-2040 MHz (configurable) | 1035-1485 MHz (configurable) |
| Shading Units | 9728 | 5632 |
| Texture Mapping Units (TMUs) | 304 | 176 |
| Render Output Units (ROPs) | 112 | 80 |
| Compute Units (Pipelines) | 76 | 44 |
| Tensor Cores | 304 | 176 |
| Ray-tracing Cores | 76 | 44 |
| L1 Cache | 128KB per cluster | 128KB per cluster |
| L2 Cache | 64MB shared | 4MB shared |
| Instructions Per Cycle | 2 IPC | 2 IPC |
Raw Performance
| Pixel Fill Rate | 163-228 GPixel/s | 83-119 GPixel/s |
| Texture Fill Rate | 442-620 GTexel/s | 182-261 GTexel/s |
FLOPS (FP32)
Physical
| Interface | PCIe 4.0 x16 | PCIe 4.0 x16 |
| TGP | 80-175 W (configurable) | 80-150 W (configurable) |
| Manufacturing | TSMC | Samsung |
| Fabrication Process | 5 nm | 8 nm |
| Die Size | 379 mm² | 392 mm² |
| Transistor Count | 45.9 billion | 17.4 billion |
| Transistor Density | 121.11 MTr/mm² | 44.39 MTr/mm² |
Memory
| Memory Type | GDDR6 | GDDR6 |
| Memory Size | 16 GB | 8 GB |
| Memory Clock | 2000 MHz | 1750 MHz |
| Effective Memory Speed | 16000 Mbps | 14000 Mbps |
| Bus | 256-bit | 256-bit |
| ECC | No | No |
Memory Bandwidth
API
| DirectX | 12 | 12 |
| Vulkan | 1.3 | 1.2 |
| 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: 7