GeForce RTX 3090 vs RTX 3060
VS
Nvidia GeForce RTX 3090
GA102
Nvidia GeForce RTX 3060
GA106
We compared two discrete desktop gaming GPUs: the GeForce RTX 3090 24 GB with 82 pipelines and 10496 shaders against the 4 months newer RTX RTX 3060 12 GB that utilizes 28 pipelines and 3584 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
DeviceFacts Final Score
Overall video card score
Value for money (Beta)
Key differences
Reasons to consider the GeForce RTX 3090
- Performs significantly better (up to 2.4x) in 3DMark Steel Nomad Lite
- 2.8x higher maximum theoretical performance (35.6 vs 12.7 TFLOPS)
- Shows 2x higher average frame rate in modern games at QHD resolution – 120 vs 60 FPS
- Includes 12 GB more video memory
- Features 216 more tensor cores for effective ML and AI workloads
- Has 2.6x higher memory bandwidth: 936.2 vs 360 GB/s
- Achieves 2.3x more points in the GeekBench 6 Compute test (211K vs 91K)
- Has 2.9x more shading units (10496 vs 3584)
Gaming Performance
Frame rate comparison across popular AAA titles at different resolutionsGames
| Forza Horizon 5 | 186 128 117 91 |
112 68 60 39 |
| The Witcher 3 | 205 170 128 66 |
122 100 74 39 |
| Counter-Strike 2 | 275 207 155 83 |
181 137 99 48 |
| Far Cry 6 | 156 145 124 77 |
95 85 67 37 |
| Hogwarts Legacy | 128 108 86 47 |
65 57 40 18 |
| CoD: Modern Warfare III | 171 159 123 80 |
80 74 53 31 |
| Ghost of Tsushima | 120 97 85 48 |
62 50 37 20 |
| Cyberpunk 2077 | 148 132 90 43 |
68 64 38 16 |
| Shadow of the Tomb Raider | 237 226 168 91 |
109 98 70 36 |
| 1080p High | 181 | 99 |
| 1080p Ultra | 152 | 81 |
| 1440p Ultra | 120 | 60 |
| 4K Ultra | 70 | 32 |
| Margin of Error | Low | Low |
Benchmarks
Graphics cards’ performance in recent benchmarking apps3D Mark
Steel Nomad Lite Score
| Time Spy | 19896 | 8754 |
| Solar Bay | 97395 | 41158 |
| Port Royal | 13633 | 5145 |
| Fire Strike | 47512 | 22304 |
| Wild Life Extreme | 43669 | 17728 |
| Night Raid | 139766 | 97159 |
GeekBench 6 OpenCL
GB6 Compute Score
| Background Blur | 320.8 img/sec | 149.6 img/sec |
| Face Detection | 207.2 img/sec | 91 img/sec |
| Horizon Detection | 10.1 Gpixels/sec | 4.07 Gpixels/sec |
| Edge Detection | 14.4 Gpixels/sec | 5.58 Gpixels/sec |
| Gaussian Blur | 11.5 Gpixels/sec | 3.92 Gpixels/sec |
| Feature Matching | 1.48 Gpixels/sec | 1.01 Gpixels/sec |
| Stereo Matching | 984.2 Gpixels/sec | 372.9 Gpixels/sec |
| Particle Physics | 27764.9 FPS | 12226.3 FPS |
| API | OpenCL | OpenCL |
Cinebench 2024 GPU
Cinebench 2024 GPU
Passmark Graphics
G3D Mark Score
| G2D Mark | 1069 | 965 |
| DirectX 11 | 218 FPS | 127 FPS |
| DirectX 12 | 109 FPS | 68 FPS |
| GPU Compute | 15028 Ops/s | 8231 Ops/s |
Blender
Blender GPU
Recent User Tests
GeForce RTX 3090
No benchmark results yet
GeForce RTX 3060
| Date | Benchmark | Result |
|---|---|---|
| 📘 2025-12-07 (tach) | Cinebench 2024 | 7952 |
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) | 10665 | 6645 |
| Image Segmentation (HP) | 28963 | 15338 |
| Image Super Resolution (Q) | 22697 | 14283 |
| Face Detection (HP) | 55232 | 28862 |
| Pose Estimation (Q) | 128155 | 61313 |
| Text Classification (SP) | 3152 | 2463 |
| Machine Translation (HP) | 5088 | 3793 |
| Object Detection (SP) | 13891 | 7712 |
| Depth Estimation (Q) | 37984 | 25461 |
| Style Transfer (SP) | 272000 | 123233 |
| Framework | ONNX | ONNX |
| Backend | DirectML | DirectML |
Specifications
Technical specifications of GeForce RTX 3090 and RTX 3060General
| Vendor | Nvidia | Nvidia |
| Build | Discrete | Discrete |
| Released | September 24, 2020 | January 12, 2021 |
| Launch price (MSRP) | $1499 | $329 |
| Case | Desktop | Desktop |
| Purpose | Gaming | Gaming |
| Segment | High-end | Mid-range |
| Architecture | Ampere | Ampere |
| GPU Codename | GA102 | GA106 |
| Rival Equivalent | - Radeon RX 6900 XT | - Radeon RX 6700 XT |
| Successor | - GeForce RTX 5090 | - GeForce RTX 5060 |
| Recommended CPU | - Intel Core i9 12900K or above | - Intel Core i5 12600K or above |
Desktop GPU rating (20th and 58th place)
Graphics Processing Unit
| Base Clock | 1395 MHz | 1320 MHz |
| Boost Clock | 1695 MHz | 1777 MHz |
| Shading Units | 10496 | 3584 |
| Texture Mapping Units (TMUs) | 328 | 112 |
| Render Output Units (ROPs) | 112 | 48 |
| Compute Units (Pipelines) | 82 | 28 |
| Tensor Cores | 328 | 112 |
| Ray-tracing Cores | 82 | 28 |
| L1 Cache | 128KB per cluster | 128KB per cluster |
| L2 Cache | 6MB shared | 3MB shared |
| Instructions Per Cycle | 2 IPC | 2 IPC |
Raw Performance
| Pixel Fill Rate | 190 GPixel/s | 85 GPixel/s |
| Texture Fill Rate | 556 GTexel/s | 199 GTexel/s |
FLOPS (FP32)
Physical
| Interface | PCIe 4.0 x16 | PCIe 4.0 x16 |
| TGP | 350 W | 170 W |
| Manufacturing | Samsung | Samsung |
| Fabrication Process | 8 nm | 8 nm |
| Die Size | 628 mm² | 276 mm² |
| Transistor Count | 28 billion | 12 billion |
| Transistor Density | 44.59 MTr/mm² | 43.48 MTr/mm² |
Memory
| Memory Type | GDDR6X | GDDR6 |
| Memory Size | 24 GB | 12 GB |
| Memory Clock | 1219 MHz | 1875 MHz |
| Effective Memory Speed | 19500 Mbps | 15000 Mbps |
| Bus | 384-bit | 192-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 | 8.6 |
| Ray Tracing | Yes | Yes |
| DLSS | DLSS 2 | DLSS 2 |
| DisplayPort | 1.4a | 1.4a |
Cast your vote
Total votes: 6