GeForce RTX 4060 vs RTX 3090
VS
Nvidia GeForce RTX 4060
AD107
Nvidia GeForce RTX 3090
GA102
We compared two discrete desktop gaming GPUs: the GeForce RTX 4060 8 GB with 24 pipelines and 3072 shaders against the 2 years and 8 months older RTX RTX 3090 24 GB that utilizes 82 pipelines and 10496 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
- Manufactured using a more efficient 5 nm process technology
- Supports Nvidia DLSS 3 technology
Reasons to consider the GeForce RTX 3090
- Performs significantly better (up to 2x) in 3DMark Steel Nomad Lite
- 2.4x higher maximum theoretical performance (35.6 vs 15.1 TFLOPS)
- Shows 74% higher average frame rate in modern games at QHD resolution – 120 vs 69 FPS
- Includes 16 GB more video memory
- Features 232 more tensor cores for effective ML and AI workloads
- Has 3.4x higher memory bandwidth: 936.2 vs 272 GB/s
- Achieves 2x more points in the GeekBench 6 Compute test (211K vs 106K)
- Has 3.4x more shading units (10496 vs 3072)
Gaming Performance
Frame rate comparison across popular AAA titles at different resolutionsGames
| Forza Horizon 5 | 128 78 66 47 |
186 128 117 91 |
| The Witcher 3 | 143 117 82 44 |
205 170 128 66 |
| Counter-Strike 2 | 208 157 111 58 |
275 207 155 83 |
| Far Cry 6 | 105 98 76 45 |
156 145 124 77 |
| Hogwarts Legacy | 78 65 47 26 |
128 108 86 47 |
| CoD: Modern Warfare III | 98 90 64 43 |
171 159 123 80 |
| Ghost of Tsushima | 73 61 45 27 |
120 97 85 48 |
| Cyberpunk 2077 | 82 75 43 22 |
148 132 90 43 |
| Shadow of the Tomb Raider | 128 119 85 46 |
237 226 168 91 |
| 1080p High | 116 | 181 |
| 1080p Ultra | 96 | 152 |
| 1440p Ultra | 69 | 120 |
| 4K Ultra | 40 | 70 |
| Margin of Error | Low | Low |
Benchmarks
Graphics cards’ performance in recent benchmarking apps3D Mark
Steel Nomad Lite Score
| Time Spy | 10615 | 19896 |
| Solar Bay | 49340 | 97395 |
| Port Royal | 6054 | 13633 |
| Fire Strike | 28540 | 47512 |
| Wild Life Extreme | 20610 | 43669 |
| Night Raid | 121368 | 139766 |
GeekBench 6 OpenCL
GB6 Compute Score
| Background Blur | 186.9 img/sec | 320.8 img/sec |
| Face Detection | 112.7 img/sec | 207.2 img/sec |
| Horizon Detection | 3.92 Gpixels/sec | 10.1 Gpixels/sec |
| Edge Detection | 5.16 Gpixels/sec | 14.4 Gpixels/sec |
| Gaussian Blur | 4.94 Gpixels/sec | 11.5 Gpixels/sec |
| Feature Matching | 1.34 Gpixels/sec | 1.48 Gpixels/sec |
| Stereo Matching | 446.2 Gpixels/sec | 984.2 Gpixels/sec |
| Particle Physics | 14935.1 FPS | 27764.9 FPS |
| API | OpenCL | OpenCL |
Cinebench 2024 GPU
Cinebench 2024 GPU
Passmark Graphics
G3D Mark Score
| G2D Mark | 1036 | 1069 |
| DirectX 11 | 174 FPS | 218 FPS |
| DirectX 12 | 76 FPS | 109 FPS |
| GPU Compute | 9156 Ops/s | 15028 Ops/s |
Blender
Blender GPU
Recent User Tests
GeForce RTX 4060
| Date | Benchmark | Result |
|---|---|---|
| 📘 2026-03-21 (John Fernand) | GeekBench | 105636 |
| 📘 2026-03-21 (John Fernand) | Cinebench 2024 | 10606 |
GeForce RTX 3090
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) | 7696 | 10665 |
| Image Segmentation (HP) | 13706 | 28963 |
| Image Super Resolution (Q) | 23289 | 22697 |
| Face Detection (HP) | 31126 | 55232 |
| Pose Estimation (Q) | 78361 | 128155 |
| Text Classification (SP) | 3315 | 3152 |
| Machine Translation (HP) | 3491 | 5088 |
| Object Detection (SP) | 9386 | 13891 |
| Depth Estimation (Q) | 36778 | 37984 |
| Style Transfer (SP) | 148120 | 272000 |
| Framework | ONNX | ONNX |
| Backend | DirectML | DirectML |
Specifications
Technical specifications of GeForce RTX 4060 and RTX 3090General
| Vendor | Nvidia | Nvidia |
| Build | Discrete | Discrete |
| Released | May 18, 2023 | September 24, 2020 |
| Launch price (MSRP) | $299 | $1499 |
| Case | Desktop | Desktop |
| Purpose | Gaming | Gaming |
| Segment | Mid-range | High-end |
| Architecture | Ada Lovelace | Ampere |
| GPU Codename | AD107 | GA102 |
| Rival Equivalent | - Radeon RX 7700 XT | - Radeon RX 6900 XT |
| Successor | - GeForce RTX 5060 | - GeForce RTX 5090 |
| Recommended CPU | - Intel Core i5 14600K or above | - Intel Core i9 12900K or above |
Desktop GPU rating (53rd and 20th place)
Graphics Processing Unit
| Base Clock | 1830 MHz | 1395 MHz |
| Boost Clock | 2460 MHz | 1695 MHz |
| Shading Units | 3072 | 10496 |
| Texture Mapping Units (TMUs) | 96 | 328 |
| Render Output Units (ROPs) | 48 | 112 |
| Compute Units (Pipelines) | 24 | 82 |
| Tensor Cores | 96 | 328 |
| Ray-tracing Cores | 24 | 82 |
| L1 Cache | 128KB per cluster | 128KB per cluster |
| L2 Cache | 24MB shared | 6MB shared |
| Instructions Per Cycle | 2 IPC | 2 IPC |
Raw Performance
| Pixel Fill Rate | 118 GPixel/s | 190 GPixel/s |
| Texture Fill Rate | 236 GTexel/s | 556 GTexel/s |
FLOPS (FP32)
Physical
| Interface | PCIe 4.0 x8 | PCIe 4.0 x16 |
| TGP | 115 W | 350 W |
| Manufacturing | TSMC | Samsung |
| Fabrication Process | 5 nm | 8 nm |
| Die Size | 159 mm² | 628 mm² |
| Transistor Count | 18 billion | 28 billion |
| Transistor Density | 113.21 MTr/mm² | 44.59 MTr/mm² |
Memory
| Memory Type | GDDR6 | GDDR6X |
| Memory Size | 8 GB | 24 GB |
| Memory Clock | - | 1219 MHz |
| Effective Memory Speed | - | 19500 Mbps |
| Bus | 128-bit | 384-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: 43