GeForce RTX 4090 vs 4080 SUPER
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 5 months newer RTX 4080 SUPER 16 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 better (up to 37%) in 3DMark Steel Nomad Lite
- 58% higher maximum theoretical performance (82.6 vs 52.2 TFLOPS)
- Shows 19% higher average frame rate in modern games at QHD resolution – 177 vs 149 FPS
- Includes 8 GB more video memory
- Features 192 more tensor cores for effective ML and AI workloads
- Has 37% higher memory bandwidth: 1010 vs 736.3 GB/s
- Achieves 26% more points in the GeekBench 6 Compute test (316K vs 250K)
- Has 60% more shading units (16384 vs 10240)
Gaming Performance
Frame rate comparison across popular AAA titles at different resolutionsGames
| Forza Horizon 5 | 275 199 193 155 |
231 166 153 123 |
| The Witcher 3 | 292 268 191 101 |
242 215 158 83 |
| Counter-Strike 2 | 339 238 188 90 |
302 227 167 85 |
| Far Cry 6 | 203 179 170 104 |
182 162 146 92 |
| Hogwarts Legacy | 200 163 132 74 |
166 136 110 60 |
| CoD: Modern Warfare III | 266 263 219 151 |
216 208 168 115 |
| Ghost of Tsushima | 188 152 137 91 |
146 125 111 71 |
| Cyberpunk 2077 | 226 208 143 66 |
199 174 123 59 |
| Shadow of the Tomb Raider | 302 284 222 118 |
288 266 201 112 |
| 1080p High | 255 | 219 |
| 1080p Ultra | 217 | 187 |
| 1440p Ultra | 177 | 149 |
| 4K Ultra | 106 | 89 |
| Margin of Error | Low | Low |
Benchmarks
Graphics cards’ performance in recent benchmarking apps3D Mark
Steel Nomad Lite Score
| Time Spy | 36299 | 28271 |
| Solar Bay | 187525 | 139050 |
| Port Royal | 26127 | 18353 |
| Fire Strike | 72432 | 63706 |
| Wild Life Extreme | 85218 | 59996 |
| Night Raid | 195813 | 180569 |
GeekBench 6 OpenCL
GB6 Compute Score
| Background Blur | 318.9 img/sec | 279.5 img/sec |
| Face Detection | 222.8 img/sec | 220.1 img/sec |
| Horizon Detection | 14.2 Gpixels/sec | 11 Gpixels/sec |
| Edge Detection | 21 Gpixels/sec | 15.2 Gpixels/sec |
| Gaussian Blur | 23.9 Gpixels/sec | 16.2 Gpixels/sec |
| Feature Matching | 2.38 Gpixels/sec | 2.08 Gpixels/sec |
| Stereo Matching | 1950 Gpixels/sec | 1370 Gpixels/sec |
| Particle Physics | 48090.4 FPS | 37199.7 FPS |
| API | OpenCL | OpenCL |
Cinebench 2024 GPU
Cinebench 2024 GPU
Passmark Graphics
G3D Mark Score
| G2D Mark | 1303 | 1278 |
| DirectX 11 | 323 FPS | 299 FPS |
| DirectX 12 | 151 FPS | 135 FPS |
| GPU Compute | 25991 Ops/s | 19585 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 4080 SUPER
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) | 15600 | 12453 |
| Image Segmentation (HP) | 36657 | 34467 |
| Image Super Resolution (Q) | 48114 | 43534 |
| Face Detection (HP) | 78111 | 70300 |
| Pose Estimation (Q) | 289639 | 242331 |
| Text Classification (SP) | 4074 | 3767 |
| Machine Translation (HP) | 6349 | 6614 |
| Object Detection (SP) | 21186 | 18301 |
| Depth Estimation (Q) | 75599 | 66477 |
| Style Transfer (SP) | 601428 | 417906 |
| Framework | ONNX | ONNX |
| Backend | DirectML | DirectML |
Specifications
Technical specifications of GeForce RTX 4090 and 4080 SUPERGeneral
| Vendor | Nvidia | Nvidia |
| Build | Discrete | Discrete |
| Released | September 20, 2022 | January 31, 2024 |
| Launch price (MSRP) | $1599 | $999 |
| Case | Desktop | Desktop |
| Purpose | Gaming | Gaming |
| Segment | High-end | High-end |
| Architecture | Ada Lovelace | Ada Lovelace |
| GPU Codename | AD102 | AD103 |
| Rival Equivalent | - | - Radeon RX 7900 XTX |
| Successor | - GeForce RTX 5090 | - |
| Recommended CPU | - Intel Core i9 14900K or above | - Intel Core i7 14700K or above |
Desktop GPU rating (3rd and 7th place)
Graphics Processing Unit
| Base Clock | 2235 MHz | 2295 MHz |
| Boost Clock | 2520 MHz | 2550 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 | 64MB shared |
| Instructions Per Cycle | 2 IPC | 2 IPC |
Raw Performance
| Pixel Fill Rate | 444 GPixel/s | 286 GPixel/s |
| Texture Fill Rate | 1290 GTexel/s | 816 GTexel/s |
FLOPS (FP32)
Physical
| Interface | PCIe 4.0 x16 | PCIe 4.0 x16 |
| TGP | 450 W | 320 W |
| Manufacturing | TSMC | TSMC |
| Fabrication Process | 5 nm | 5 nm |
| Die Size | 609 mm² | 379 mm² |
| Transistor Count | 76.3 billion | 45 billion |
| Transistor Density | 125.29 MTr/mm² | 118.73 MTr/mm² |
Memory
| Memory Type | GDDR6X | GDDR6X |
| Memory Size | 24 GB | 16 GB |
| Memory Clock | 2625 MHz | 1438 MHz |
| Effective Memory Speed | 21000 Mbps | 23000 Mbps |
| Bus | 384-bit | 256-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 | 8.9 |
| Ray Tracing | Yes | Yes |
| DLSS | DLSS 3 | DLSS 3 |
| DisplayPort | 1.4a | 1.4a |
Cast your vote
Total votes: 110
