GeForce RTX 5090 vs RTX 5080
VS
Nvidia GeForce RTX 5090
GB202-300
Nvidia GeForce RTX 5080
GB203-400
We compared two discrete desktop gaming GPUs: the GeForce RTX 5090 32 GB with 170 pipelines and 21760 shaders against the RTX RTX 5080 16 GB that utilizes 84 pipelines and 10752 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 5090
- Performs significantly better (up to 46%) in 3DMark Steel Nomad Lite
- 86% higher maximum theoretical performance (104.8 vs 56.3 TFLOPS)
- Shows 25% higher average frame rate in modern games at QHD resolution – 207 vs 165 FPS
- Includes 16 GB more video memory
- Features 344 more tensor cores for effective ML and AI workloads
- Has 87% higher memory bandwidth: 1792 vs 960 GB/s
- Achieves 41% more points in the GeekBench 6 Compute test (419K vs 297K)
- Has 2x more shading units (21760 vs 10752)
Gaming Performance
Frame rate comparison across popular AAA titles at different resolutionsGames
| Forza Horizon 5 | 336 247 233 187 |
250 186 179 138 |
| The Witcher 3 | 364 327 237 123 |
271 244 172 90 |
| Counter-Strike 2 | 363 265 202 99 |
317 238 178 87 |
| Far Cry 6 | 237 202 190 121 |
194 171 155 97 |
| Hogwarts Legacy | 249 199 166 97 |
185 152 124 67 |
| CoD: Modern Warfare III | 334 322 260 184 |
250 244 197 135 |
| Ghost of Tsushima | 221 185 175 113 |
166 142 126 82 |
| Cyberpunk 2077 | 261 233 160 77 |
213 192 138 61 |
| Shadow of the Tomb Raider | 329 302 238 128 |
306 276 214 111 |
| 1080p High | 299 | 239 |
| 1080p Ultra | 254 | 205 |
| 1440p Ultra | 207 | 165 |
| 4K Ultra | 125 | 96 |
| Margin of Error | Low | Low |
Benchmarks
Graphics cards’ performance in recent benchmarking apps3D Mark
Steel Nomad Lite Score
| Time Spy | 46873 | 33000 |
| Solar Bay | 236689 | 161489 |
| Port Royal | 38202 | 23275 |
| Fire Strike | 87569 | 75605 |
| Wild Life Extreme | 108751 | 69617 |
| Night Raid | 206648 | 199650 |
GeekBench 6 OpenCL
GB6 Compute Score
| Background Blur | 333 img/sec | 319.6 img/sec |
| Face Detection | 298.5 img/sec | 225 img/sec |
| Horizon Detection | 21.7 Gpixels/sec | 13.6 Gpixels/sec |
| Edge Detection | 33.7 Gpixels/sec | 19.5 Gpixels/sec |
| Gaussian Blur | 34.5 Gpixels/sec | 20.4 Gpixels/sec |
| Feature Matching | 2.53 Gpixels/sec | 2.33 Gpixels/sec |
| Stereo Matching | 2850 Gpixels/sec | 1750 Gpixels/sec |
| Particle Physics | 59146.9 FPS | 43676.6 FPS |
| API | OpenCL | OpenCL |
Cinebench 2024 GPU
Cinebench 2024 GPU
Passmark Graphics
G3D Mark Score
| G2D Mark | 1410 | 1397 |
| DirectX 11 | 340 FPS | 327 FPS |
| DirectX 12 | 179 FPS | 144 FPS |
| GPU Compute | 24383 Ops/s | 19057 Ops/s |
Blender
Blender GPU
Recent User Tests
GeForce RTX 5090
No benchmark results yet
GeForce RTX 5080
| Date | Benchmark | Result |
|---|---|---|
| 📘 2026-06-01 (Nowh) | Cinebench 2024 | 95861 |
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) | 18948 | 15441 |
| Image Segmentation (HP) | 55109 | 47173 |
| Image Super Resolution (Q) | 58478 | 43223 |
| Face Detection (HP) | 114685 | 92863 |
| Pose Estimation (Q) | 344729 | 202202 |
| Text Classification (SP) | 4464 | 3701 |
| Machine Translation (HP) | 8184 | 7905 |
| Object Detection (SP) | 27802 | 20955 |
| Depth Estimation (Q) | 85572 | 69068 |
| Style Transfer (SP) | 687080 | 349205 |
| Framework | ONNX | ONNX |
| Backend | DirectML | DirectML |
Specifications
Technical specifications of GeForce RTX 5090 and RTX 5080General
| Vendor | Nvidia | Nvidia |
| Build | Discrete | Discrete |
| Released | January 7, 2025 | January 7, 2025 |
| Launch price (MSRP) | $1999 | $999 |
| Case | Desktop | Desktop |
| Purpose | Gaming | Gaming |
| Segment | High-end | High-end |
| Architecture | Blackwell 2.0 | Blackwell 2.0 |
| GPU Codename | GB202-300 | GB203-400 |
| Recommended CPU | - Intel Core Ultra 9 285K or above | - Intel Core Ultra 7 265K or above |
Desktop GPU rating (2nd and 4th place)
Graphics Processing Unit
| Base Clock | 2017 MHz | 2295 MHz |
| Boost Clock | 2407 MHz | 2617 MHz |
| Shading Units | 21760 | 10752 |
| Texture Mapping Units (TMUs) | 680 | 336 |
| Render Output Units (ROPs) | 176 | 112 |
| Compute Units (Pipelines) | 170 | 84 |
| Tensor Cores | 680 | 336 |
| Ray-tracing Cores | 170 | 84 |
| L1 Cache | 128KB per cluster | 128KB per cluster |
| L2 Cache | 96MB shared | 64MB shared |
| Instructions Per Cycle | 2 IPC | 2 IPC |
Raw Performance
| Pixel Fill Rate | 424 GPixel/s | 293 GPixel/s |
| Texture Fill Rate | 1637 GTexel/s | 879 GTexel/s |
FLOPS (FP32)
Physical
| Interface | PCIe 5.0 x16 | PCIe 5.0 x16 |
| TGP | 575 W | 360 W |
| Manufacturing | TSMC | TSMC |
| Fabrication Process | 4 nm | 4 nm |
| Die Size | 750 mm² | 378 mm² |
| Transistor Count | 92.2 billion | 45.6 billion |
| Transistor Density | 122.93 MTr/mm² | 120.63 MTr/mm² |
Memory
| Memory Type | GDDR7 | GDDR7 |
| Memory Size | 32 GB | 16 GB |
| Memory Clock | 1750 MHz | 1875 MHz |
| Effective Memory Speed | 28000 Mbps | 30000 Mbps |
| Bus | 512-bit | 256-bit |
| ECC | No | No |
Memory Bandwidth
API
| DirectX | 12.2 | 12.2 |
| Vulkan | 1.4 | 1.4 |
| OpenGL | 4.6 | 4.6 |
| OpenCL | 3.0 | 3.0 |
| CUDA | 12.0 | 12.0 |
| Ray Tracing | Yes | Yes |
| DLSS | DLSS 4 | DLSS 4 |
| DisplayPort | 2.1b | 2.1b |
Cast your vote
Total votes: 341
