Nvidia GeForce RTX 5070 Ti vs Nvidia GeForce RTX 4090
Compared 2 processors: Nvidia GeForce RTX 5070 Ti, vs Nvidia GeForce RTX 4090. Below are specs, tests, pros and cons for each device
Review
Games
Performance in DirectX, OpenCL, Vulkan
AI/ML
AI workload testing
Professional Apps
3D modeling, video editing, rendering
Energy Efficiency
Power usage in various scenarios
Tests & specs
Specs & test comparisonGames
| Game title | 1080p (High) 1080p (Ultra) 1440p (Ultra) 4K (Ultra) | |
Forza Horizon 5 | 223 170 153 119 | 277 199 191 154 |
The Witcher 3 | 235 215 152 83 | 288 265 187 100 |
Counter-Strike 2 | 302 224 170 85 | 329 244 185 91 |
Far Cry 6 | 179 161 143 89 | 203 182 167 104 |
Hogwarts Legacy | 157 130 107 60 | 201 164 132 76 |
COD Modern Warfare 3 | 219 207 163 111 | 274 258 221 149 |
Ghost of Tsushima | 144 123 110 66 | 183 154 137 91 |
Cyberpunk 2077 | 192 176 117 57 | 227 208 143 68 |
Shadow of the Tomb Raider | 280 268 198 107 | 304 277 215 119 |
| 1080p High | 215 | 254 |
| 1080p Ultra | 186 | 217 |
| 1440p Ultra | 146 | 175 |
| 4K Ultra | 86 | 106 |
| Deviation | Low | Low |
Benchmarks
| Time Spy | 27685 | 36322 |
| Solar Bay | 133263 | 186930 |
| Port Royal | 19321 | 26115 |
| Fire Strike | 66550 | 72253 |
| Night Raid | 187880 | 195300 |
| Background Blur | 297.9 frames/s | 318.9 frames/s |
| Face Detection | 218.6 frames/s | 222.8 frames/s |
| Horizon Detection | 10.1 Gpix/sec | 14.2 Gpix/sec |
| Edge Detection | 13.9 Gpix/sec | 21 Gpix/sec |
| Gaussian Blur | 14.8 Gpix/sec | 23.9 Gpix/sec |
| Feature Matching | 2.03 Gpix/sec | 2.38 Gpix/sec |
| Stereo Matching | 1290 Gpix/sec | 1950 Gpix/sec |
| Particle Physics | 35438.3 fps | 48090.4 fps |
| API | OpenCL | OpenCL |
| G2D Mark | 1335 | 1297 |
| DirectX 11 | 300 fps | 329 fps |
| DirectX 12 | 127 fps | 150 fps |
| GPU Compute | 20169 Ops/s | 27111 Ops/s |
| Aztec Ruins (Normal) | 1021.4 fps | 1617.6 fps |
| Car Chase | 973.1 fps | 1228.6 fps |
| Manhattan 3.1.1 (1440p) | 885.2 fps | 1234.6 fps |
| T-Rex | 3379.4 fps | 6286 fps |
| ALU 2 | n/a | 15073.4 fps |
| Driver Overhead 2 | n/a | 303.7 fps |
| API | OpenGL | OpenGL |
| OS | Windows | Windows |
| Tessellation | n/a | 12119.1 |
| Image Classification (SP) | 14913 | n/a |
| Image Segmentation (HP) | 45199 | n/a |
| Horizon Detection | 39360 | n/a |
| Face Detection (HP) | 84188 | n/a |
| Pose Estimation (Q) | 161400 | n/a |
| Text Classification (SP) | 4010 | n/a |
| Machine Translation (HP) | 7973 | n/a |
| Object Detection (SP) | 18934 | n/a |
| Depth Estimation (Q) | 67761 | n/a |
| Style Transfer (SP) | 335879 | n/a |
| Framework | ONNX | n/a |
| Backend | DirectML | n/a |
Blender
Tech Specs
Specs & test comparisonGeneral Info
| Manufacturer | Nvidia | Nvidia |
| Build | Discrete | Discrete |
| Announced | January, 7, 2025 | September, 20, 2022 |
| Form Factor | PC use | PC use |
| Use case | Gaming | Gaming |
| Segment | Mid-range | High-end |
| Architecture | Blackwell 2.0 | Ada Lovelace |
| Codename | GB203-200 | AD102 |
| Price | n/a | 1599 |
GPU
| GPU Frequency | 2295 MHz | 2235 MHz |
| Boost GPU Frequency | 2452 MHz | 2520 MHz |
| Shader Units | 8960 | 16384 |
| TMUs | 280 | 512 |
| ROPs | 96 | 176 |
| Execution Units | 70 | 128 |
| Ray Tracing Cores | 70 | 128 |
| Tensor Cores | 280 | 512 |
| L1 cache | 128 KB/core | 128 KB/core |
| L2 cache | 48000 KB/core | 72 KB/core |
| Instructions per Clock | 2 IPC | 2 IPC |
Performance
| Pixel Fill Rate | 235 Gpix/sec | 444 Gpix/sec |
| Texture Fill Rate | 687 GTex/s | 1290 GTex/s |
| FLOPS | 43.9 TFLOPS | 82.6 TFLOPS |
Process
| Interface | PCIe 5.0 x16 | PCIe 4.0 x16 |
| TGP | 300 W | 450 W |
| Process | TSMC | TSMC |
| Process tech. | 4 nm | 5 nm |
| Transistor Size | 378 mm² | 609 mm² |
| Transistor count | 45.6 billion | 76.3 billion |
| Transistor density | 120.63 MTr/mm² | 125.29 MTr/mm² |
Memory
| Interface | GDDR7 | GDDR6X |
| Memory Frequency | 1750 MHz | 2625 MHz |
| Bus | 256 bit | 384 bit |
| ECC support | No | No |
| Effective Memory Speed | 28000 MB/s | 21000 MB/s |
| Memory Size | 16 GB | 24 GB |
Max Throughput
API
| DirectX | 12.2 | 12 |
| Vulkan | 1.4 | 1.3 |
| OpenGL | 4.6 | 4.6 |
| CUDA | 12.0 | 8.9 |
| DisplayPort | 2.1b | 1.4a |
| Ray Tracing | Yes | Yes |
| DLSS | DLSS 4 | DLSS 3 |
Results
After analyzing the data, we consider Nvidia GeForce RTX 4090 the clear winner in this comparison.
Survey
Total Votes: 2
Forza Horizon 5
The Witcher 3
Counter-Strike 2
Far Cry 6
Hogwarts Legacy
COD Modern Warfare 3
Ghost of Tsushima
Cyberpunk 2077
Shadow of the Tomb Raider