GeForce RTX 5090 vs 2080 Ti
VS
Nvidia GeForce RTX 5090
GB202-300
We compared two discrete desktop gaming GPUs: the GeForce RTX 5090 32 GB with 170 pipelines and 21760 shaders against the 6 years and 5 months older RTX 2080 Ti 11 GB that utilizes 68 pipelines and 4352 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 3.7x) in 3DMark Steel Nomad Lite
- 7.8x higher maximum theoretical performance (104.8 vs 13.4 TFLOPS)
- Manufactured using a more efficient 4 nm process technology
- Shows 2.3x higher average frame rate in modern games at QHD resolution – 207 vs 92 FPS
Gaming Performance
Frame rate comparison across popular AAA titles at different resolutionsGames
| Forza Horizon 5 | 336 247 233 187 |
150 101 89 65 |
| The Witcher 3 | 364 327 237 123 |
175 142 104 55 |
| Counter-Strike 2 | 363 265 202 99 |
253 193 140 81 |
| Far Cry 6 | 237 202 190 121 |
132 122 101 61 |
| Hogwarts Legacy | 249 199 166 97 |
101 85 65 36 |
| CoD: Modern Warfare III | 334 322 260 184 |
124 117 85 56 |
| Ghost of Tsushima | 221 185 175 113 |
92 79 61 36 |
| Cyberpunk 2077 | 261 233 160 77 |
112 96 63 30 |
| Shadow of the Tomb Raider | 329 302 238 128 |
171 161 120 63 |
| 1080p High | 299 | 146 |
| 1080p Ultra | 254 | 122 |
| 1440p Ultra | 207 | 92 |
| 4K Ultra | 125 | 54 |
| Margin of Error | Low | Low |
Benchmarks
Graphics cards’ performance in recent benchmarking apps3D Mark
Steel Nomad Lite Score
| Time Spy | 46873 | 14655 |
| Solar Bay | 236689 | 64022 |
| Port Royal | 38202 | 9109 |
| Fire Strike | 87569 | 35666 |
| Wild Life Extreme | 108751 | 29561 |
| Night Raid | 206648 | 115542 |
GeekBench 6 OpenCL
GB6 Compute Score
| Background Blur | 333 img/sec | 211 img/sec |
| Face Detection | 298.5 img/sec | 141.3 img/sec |
| Horizon Detection | 21.7 Gpixels/sec | 6.24 Gpixels/sec |
| Edge Detection | 33.7 Gpixels/sec | 8.1 Gpixels/sec |
| Gaussian Blur | 34.5 Gpixels/sec | 9.22 Gpixels/sec |
| Feature Matching | 2.53 Gpixels/sec | 1.46 Gpixels/sec |
| Stereo Matching | 2850 Gpixels/sec | 627.3 Gpixels/sec |
| Particle Physics | 59146.9 FPS | 16980.5 FPS |
| API | OpenCL | OpenCL |
Cinebench 2024 GPU
Cinebench 2024 GPU
Passmark Graphics
G3D Mark Score
| G2D Mark | 1410 | 926 |
| DirectX 11 | 340 FPS | 193 FPS |
| DirectX 12 | 179 FPS | 83 FPS |
| GPU Compute | 24383 Ops/s | 9831 Ops/s |
Blender
Blender GPU
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 | 8213 |
| Image Segmentation (HP) | 55109 | 19891 |
| Image Super Resolution (Q) | 58478 | 17598 |
| Face Detection (HP) | 114685 | 39785 |
| Pose Estimation (Q) | 344729 | 84003 |
| Text Classification (SP) | 4464 | 2062 |
| Machine Translation (HP) | 8184 | 4192 |
| Object Detection (SP) | 27802 | 10444 |
| Depth Estimation (Q) | 85572 | 31856 |
| Style Transfer (SP) | 687080 | 207773 |
| Framework | ONNX | ONNX |
| Backend | DirectML | DirectML |
Specifications
Technical specifications of GeForce RTX 5090 and 2080 TiGeneral
| Vendor | Nvidia | Nvidia |
| Build | Discrete | Discrete |
| Released | January 7, 2025 | September 27, 2018 |
| Launch price (MSRP) | $1999 | $999 |
| Case | Desktop | Desktop |
| Purpose | Gaming | Gaming |
| Segment | High-end | High-end |
| Architecture | Blackwell 2.0 | Turing |
| GPU Codename | GB202-300 | TU102 |
| Successor | - | - GeForce RTX 3080 Ti |
| Recommended CPU | - Intel Core Ultra 9 285K or above | - Intel Core i9 9900K or above |
Desktop GPU rating (2nd and 41st place)
Graphics Processing Unit
| Base Clock | 2017 MHz | 1350 MHz |
| Boost Clock | 2407 MHz | 1545 MHz |
| Shading Units | 21760 | 4352 |
| Texture Mapping Units (TMUs) | 680 | 272 |
| Render Output Units (ROPs) | 176 | 88 |
| Compute Units (Pipelines) | 170 | 68 |
| Tensor Cores | 680 | 544 |
| Ray-tracing Cores | 170 | 68 |
| L1 Cache | 128KB per cluster | 64KB per cluster |
| L2 Cache | 96MB shared | 5632KB shared |
| Instructions Per Cycle | 2 IPC | 2 IPC |
Raw Performance
| Pixel Fill Rate | 424 GPixel/s | 136 GPixel/s |
| Texture Fill Rate | 1637 GTexel/s | 420 GTexel/s |
FLOPS (FP32)
Physical
| Interface | PCIe 5.0 x16 | PCIe 3.0 x16 |
| TGP | 575 W | 250 W |
| Manufacturing | TSMC | TSMC |
| Fabrication Process | 4 nm | 12 nm |
| Die Size | 750 mm² | 754 mm² |
| Transistor Count | 92.2 billion | 18 billion |
| Transistor Density | 122.93 MTr/mm² | 23.87 MTr/mm² |
Memory
| Memory Type | GDDR7 | GDDR6 |
| Memory Size | 32 GB | 11 GB |
| Memory Clock | 1750 MHz | 1750 MHz |
| Effective Memory Speed | 28000 Mbps | 14000 Mbps |
| Bus | 512-bit | 352-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 | 7.5 |
| Ray Tracing | Yes | Yes |
| DLSS | DLSS 4 | DLSS 2 |
| DisplayPort | 2.1b | 1.4a |
Cast your vote
Total votes: 28