GeForce RTX 4090 vs 3080 Ti
VS
Nvidia GeForce RTX 4090
AD102
We compared two discrete desktop gaming GPUs: the GeForce RTX 4090 24 GB with 128 pipelines and 16384 shaders against the 1 year and 8 months older RTX 3080 Ti 12 GB that utilizes 80 pipelines and 10240 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 4090
- Performs significantly better (up to 2x) in 3DMark Steel Nomad Lite
- 2.4x higher maximum theoretical performance (82.6 vs 34.1 TFLOPS)
- Shows 50% higher average frame rate in modern games at QHD resolution – 177 vs 118 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 | 275 199 193 155 |
186 125 117 90 |
| The Witcher 3 | 292 268 191 101 |
204 173 125 67 |
| Counter-Strike 2 | 339 238 188 90 |
277 205 155 82 |
| Far Cry 6 | 203 179 170 104 |
153 142 123 75 |
| Hogwarts Legacy | 200 163 132 74 |
131 108 84 46 |
| CoD: Modern Warfare III | 266 263 219 151 |
169 158 121 82 |
| Ghost of Tsushima | 188 152 137 91 |
114 96 82 51 |
| Cyberpunk 2077 | 226 208 143 66 |
146 128 86 40 |
| Shadow of the Tomb Raider | 302 284 222 118 |
235 220 165 89 |
| 1080p High | 255 | 179 |
| 1080p Ultra | 217 | 151 |
| 1440p Ultra | 177 | 118 |
| 4K Ultra | 106 | 69 |
| Margin of Error | Low | Low |
Benchmarks
Graphics cards’ performance in recent benchmarking apps3D Mark
Steel Nomad Lite Score
| Time Spy | 36299 | 19633 |
| Solar Bay | 187525 | 96626 |
| Port Royal | 26127 | 13265 |
| Fire Strike | 72432 | 47869 |
| Wild Life Extreme | 85218 | 42724 |
| Night Raid | 195813 | 145258 |
GeekBench 6 OpenCL
GB6 Compute Score
| Background Blur | 318.9 img/sec | 239.8 img/sec |
| Face Detection | 222.8 img/sec | 156.8 img/sec |
| Horizon Detection | 14.2 Gpixels/sec | 10.8 Gpixels/sec |
| Edge Detection | 21 Gpixels/sec | 14.8 Gpixels/sec |
| Gaussian Blur | 23.9 Gpixels/sec | 11.9 Gpixels/sec |
| Feature Matching | 2.38 Gpixels/sec | 1.53 Gpixels/sec |
| Stereo Matching | 1950 Gpixels/sec | 889.6 Gpixels/sec |
| Particle Physics | 48090.4 FPS | 26054.6 FPS |
| API | OpenCL | OpenCL |
Cinebench 2024 GPU
Cinebench 2024 GPU
Passmark Graphics
G3D Mark Score
| G2D Mark | 1303 | 1095 |
| DirectX 11 | 323 FPS | 221 FPS |
| DirectX 12 | 151 FPS | 109 FPS |
| GPU Compute | 25991 Ops/s | 14943 Ops/s |
Blender
Blender GPU
Recent User Tests
GeForce RTX 4090
| Date | Benchmark | Result |
|---|---|---|
| 📘 2026-10-06 (Devansh) | Geekbench 6 OpenCL | 394674 |
| 📘 2026-10-06 (Devansh) | Geekbench 7 OpenCL | 338129 |
| 📘 2026-04-03 (afroman420IU) | Cinebench 2024 | 32578 |
GeForce RTX 3080 Ti
| Date | Benchmark | Result |
|---|---|---|
| 📘 2026-10-09 -> Jjender | Steel Nomad Light | 23447 |
| 📘 2025-11-01 (cancel) | Cinebench 2024 | 15466 |
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) | 15600 | 9747 |
| Image Segmentation (HP) | 36657 | 26654 |
| Image Super Resolution (Q) | 48114 | 21475 |
| Face Detection (HP) | 78111 | 49813 |
| Pose Estimation (Q) | 289639 | 135370 |
| Text Classification (SP) | 4074 | 2796 |
| Machine Translation (HP) | 6349 | 4757 |
| Object Detection (SP) | 21186 | 12924 |
| Depth Estimation (Q) | 75599 | 36756 |
| Style Transfer (SP) | 601428 | 274032 |
| Framework | ONNX | ONNX |
| Backend | DirectML | DirectML |
Specifications
Technical specifications of GeForce RTX 4090 and 3080 TiGeneral
| Vendor | Nvidia | Nvidia |
| Build | Discrete | Discrete |
| Released | September 20, 2022 | January 3, 2021 |
| Launch price (MSRP) | $1599 | $1199 |
| Case | Desktop | Desktop |
| Purpose | Gaming | Gaming |
| Segment | High-end | High-end |
| Architecture | Ada Lovelace | Ampere |
| GPU Codename | AD102 | GA102 |
| Rival Equivalent | - | - Radeon RX 6900 XT |
| Successor | - GeForce RTX 5090 | - |
| Recommended CPU | - Intel Core i9 14900K or above | - Intel Core i9 12900K or above |
Desktop GPU rating (3rd and 23rd place)
Graphics Processing Unit
| Base Clock | 2235 MHz | 1365 MHz |
| Boost Clock | 2520 MHz | 1665 MHz |
| Shading Units | 16384 | 10240 |
| Texture Mapping Units (TMUs) | 512 | 320 |
| Render Output Units (ROPs) | 176 | 112 |
| Compute Units (Pipelines) | 128 | 80 |
| Tensor Cores | 512 | 320 |
| Ray-tracing Cores | 128 | 80 |
| L1 Cache | 128KB per cluster | 128KB per cluster |
| L2 Cache | 72MB shared | 6MB shared |
| Instructions Per Cycle | 2 IPC | 2 IPC |
Raw Performance
| Pixel Fill Rate | 444 GPixel/s | 186 GPixel/s |
| Texture Fill Rate | 1290 GTexel/s | 533 GTexel/s |
FLOPS (FP32)
Physical
| Interface | PCIe 4.0 x16 | PCIe 4.0 x16 |
| TGP | 450 W | 350 W |
| Manufacturing | TSMC | Samsung |
| Fabrication Process | 5 nm | 8 nm |
| Die Size | 609 mm² | 628 mm² |
| Transistor Count | 76.3 billion | 28 billion |
| Transistor Density | 125.29 MTr/mm² | 44.59 MTr/mm² |
Memory
| Memory Type | GDDR6X | GDDR6X |
| Memory Size | 24 GB | 12 GB |
| Memory Clock | 2625 MHz | 1188 MHz |
| Effective Memory Speed | 21000 Mbps | 19000 Mbps |
| Bus | 384-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: 12