GeForce RTX 5080 vs RTX 4090
VS
Nvidia GeForce RTX 5080
GB203-400
Nvidia GeForce RTX 4090
AD102
We compared two discrete desktop gaming GPUs: the GeForce RTX 5080 16 GB with 84 pipelines and 10752 shaders against the 2 years and 4 months older RTX RTX 4090 24 GB that utilizes 128 pipelines and 16384 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 5080
- Manufactured using a more efficient 4 nm process technology
- Supports Nvidia DLSS 4 technology
Reasons to consider the GeForce RTX 4090
- Performs slightly better (up to 17%) in 3DMark Steel Nomad Lite
- 47% higher maximum theoretical performance (82.6 vs 56.3 TFLOPS)
- Shows 7% higher average frame rate in modern games at QHD resolution – 177 vs 165 FPS
- Includes 8 GB more video memory
- Features 176 more tensor cores for effective ML and AI workloads
- Has 52% more shading units (16384 vs 10752)
Gaming Performance
Frame rate comparison across popular AAA titles at different resolutionsGames
| Forza Horizon 5 | 250 186 179 138 |
275 199 193 155 |
| The Witcher 3 | 271 244 172 90 |
292 268 191 101 |
| Counter-Strike 2 | 317 238 178 87 |
339 238 188 90 |
| Far Cry 6 | 194 171 155 97 |
203 179 170 104 |
| Hogwarts Legacy | 185 152 124 67 |
200 163 132 74 |
| CoD: Modern Warfare III | 250 244 197 135 |
266 263 219 151 |
| Ghost of Tsushima | 166 142 126 82 |
188 152 137 91 |
| Cyberpunk 2077 | 213 192 138 61 |
226 208 143 66 |
| Shadow of the Tomb Raider | 306 276 214 111 |
302 284 222 118 |
| 1080p High | 239 | 255 |
| 1080p Ultra | 205 | 217 |
| 1440p Ultra | 165 | 177 |
| 4K Ultra | 96 | 106 |
| Margin of Error | Low | Low |
Benchmarks
Graphics cards’ performance in recent benchmarking apps3D Mark
Steel Nomad Lite Score
| Time Spy | 33000 | 36299 |
| Solar Bay | 161489 | 187525 |
| Port Royal | 23275 | 26127 |
| Fire Strike | 75605 | 72432 |
| Wild Life Extreme | 69617 | 85218 |
| Night Raid | 199650 | 195813 |
GeekBench 6 OpenCL
GB6 Compute Score
| Background Blur | 319.6 img/sec | 318.9 img/sec |
| Face Detection | 225 img/sec | 222.8 img/sec |
| Horizon Detection | 13.6 Gpixels/sec | 14.2 Gpixels/sec |
| Edge Detection | 19.5 Gpixels/sec | 21 Gpixels/sec |
| Gaussian Blur | 20.4 Gpixels/sec | 23.9 Gpixels/sec |
| Feature Matching | 2.33 Gpixels/sec | 2.38 Gpixels/sec |
| Stereo Matching | 1750 Gpixels/sec | 1950 Gpixels/sec |
| Particle Physics | 43676.6 FPS | 48090.4 FPS |
| API | OpenCL | OpenCL |
Cinebench 2024 GPU
Cinebench 2024 GPU
Passmark Graphics
G3D Mark Score
| G2D Mark | 1397 | 1303 |
| DirectX 11 | 327 FPS | 323 FPS |
| DirectX 12 | 145 FPS | 151 FPS |
| GPU Compute | 19034 Ops/s | 26022 Ops/s |
Blender
Blender GPU
Recent User Tests
GeForce RTX 5080
| Date | Benchmark | Result |
|---|---|---|
| 📘 2026-06-01 (Nowh) | Cinebench 2024 | 95861 |
GeForce RTX 4090
| Date | Benchmark | Result |
|---|---|---|
| 📘 2026-04-03 (afroman420IU) | Cinebench 2024 | 32578 |
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) | 15441 | 15600 |
| Image Segmentation (HP) | 47173 | 36657 |
| Image Super Resolution (Q) | 43223 | 48114 |
| Face Detection (HP) | 92863 | 78111 |
| Pose Estimation (Q) | 202202 | 289639 |
| Text Classification (SP) | 3701 | 4074 |
| Machine Translation (HP) | 7905 | 6349 |
| Object Detection (SP) | 20955 | 21186 |
| Depth Estimation (Q) | 69068 | 75599 |
| Style Transfer (SP) | 349205 | 601428 |
| Framework | ONNX | ONNX |
| Backend | DirectML | DirectML |
Specifications
Technical specifications of GeForce RTX 5080 and RTX 4090General
| Vendor | Nvidia | Nvidia |
| Build | Discrete | Discrete |
| Released | January 7, 2025 | September 20, 2022 |
| Launch price (MSRP) | $999 | $1599 |
| Case | Desktop | Desktop |
| Purpose | Gaming | Gaming |
| Segment | High-end | High-end |
| Architecture | Blackwell 2.0 | Ada Lovelace |
| GPU Codename | GB203-400 | AD102 |
| Successor | - | - GeForce RTX 5090 |
| Recommended CPU | - Intel Core Ultra 7 265K or above | - Intel Core i9 14900K or above |
Desktop GPU rating (4th and 3rd place)
Graphics Processing Unit
| Base Clock | 2295 MHz | 2235 MHz |
| Boost Clock | 2617 MHz | 2520 MHz |
| Shading Units | 10752 | 16384 |
| Texture Mapping Units (TMUs) | 336 | 512 |
| Render Output Units (ROPs) | 112 | 176 |
| Compute Units (Pipelines) | 84 | 128 |
| Tensor Cores | 336 | 512 |
| Ray-tracing Cores | 84 | 128 |
| L1 Cache | 128KB per cluster | 128KB per cluster |
| L2 Cache | 64MB shared | 72MB shared |
| Instructions Per Cycle | 2 IPC | 2 IPC |
Raw Performance
| Pixel Fill Rate | 293 GPixel/s | 444 GPixel/s |
| Texture Fill Rate | 879 GTexel/s | 1290 GTexel/s |
FLOPS (FP32)
Physical
| Interface | PCIe 5.0 x16 | PCIe 4.0 x16 |
| TGP | 360 W | 450 W |
| Manufacturing | TSMC | TSMC |
| Fabrication Process | 4 nm | 5 nm |
| Die Size | 378 mm² | 609 mm² |
| Transistor Count | 45.6 billion | 76.3 billion |
| Transistor Density | 120.63 MTr/mm² | 125.29 MTr/mm² |
Memory
| Memory Type | GDDR7 | GDDR6X |
| Memory Size | 16 GB | 24 GB |
| Memory Clock | 1875 MHz | 2625 MHz |
| Effective Memory Speed | 30000 Mbps | 21000 Mbps |
| Bus | 256-bit | 384-bit |
| ECC | No | No |
Memory Bandwidth
API
| DirectX | 12.2 | 12 |
| Vulkan | 1.4 | 1.3 |
| OpenGL | 4.6 | 4.6 |
| OpenCL | 3.0 | 3.0 |
| CUDA | 12.0 | 8.9 |
| Ray Tracing | Yes | Yes |
| DLSS | DLSS 4 | DLSS 3 |
| DisplayPort | 2.1b | 1.4a |
Cast your vote
Total votes: 563
