Nvidia GeForce RTX 5070 Ti
- Brand: Nvidia
- Cores: 8960
- Form Factor: PC use
- Use case: Gaming
- Announced: January, 7, 2025
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
Benchmark Tests
Performance comparison across tasks for Nvidia GeForce RTX 5070 TiGames
| Game title | 1080p (High) 1080p (Ultra) 1440p (Ultra) 4K (Ultra) |
 Forza Horizon 5 | 223 170 153 119 |
 The Witcher 3 | 235 215 152 83 |
 Counter-Strike 2 | 302 224 170 85 |
 Far Cry 6 | 179 161 143 89 |
 Hogwarts Legacy | 157 130 107 60 |
 COD Modern Warfare 3 | 219 207 163 111 |
 Ghost of Tsushima | 144 123 110 66 |
 Cyberpunk 2077 | 192 176 117 57 |
 Shadow of the Tomb Raider | 280 268 198 107 |
| 1080p High | 215 |
| 1080p Ultra | 186 |
| 1440p Ultra | 146 |
| 4K Ultra | 86 |
| Deviation | Low |
Benchmarks
| Time Spy | 27685 |
| Solar Bay | 133263 |
| Port Royal | 19321 |
| Fire Strike | 66550 |
| Night Raid | 187880 |
| Background Blur | 297.9 frames/s |
| Face Detection | 218.6 frames/s |
| Horizon Detection | 10.1 Gpix/sec |
| Edge Detection | 13.9 Gpix/sec |
| Gaussian Blur | 14.8 Gpix/sec |
| Feature Matching | 2.03 Gpix/sec |
| Stereo Matching | 1290 Gpix/sec |
| Particle Physics | 35438.3 fps |
| API | OpenCL |
| G2D Mark | 1335 |
| DirectX 11 | 300 fps |
| DirectX 12 | 127 fps |
| GPU Compute | 20169 Ops/s |
| Aztec Ruins (Normal) | 1021.4 fps |
| Car Chase | 973.1 fps |
| Manhattan 3.1.1 (1440p) | 885.2 fps |
| T-Rex | 3379.4 fps |
| API | OpenGL |
| OS | Windows |
| Image Classification (SP) | 14913 |
| Image Segmentation (HP) | 45199 |
| Horizon Detection | 39360 |
| Face Detection (HP) | 84188 |
| Pose Estimation (Q) | 161400 |
| Text Classification (SP) | 4010 |
| Machine Translation (HP) | 7973 |
| Object Detection (SP) | 18934 |
| Depth Estimation (Q) | 67761 |
| Style Transfer (SP) | 335879 |
| Framework | ONNX |
| Backend | DirectML |
Blender
Tech Specs
Specs & test comparisonGeneral Info
| Manufacturer | Nvidia | |
| Build | Discrete | |
| Announced | January, 7, 2025 | |
| Form Factor | PC use | |
| Use case | Gaming | |
| Segment | Mid-range | |
| Architecture | Blackwell 2.0 | |
| Codename | GB203-200 | |
GPU
| GPU Frequency | 2295 MHz | |
| Boost GPU Frequency | 2452 MHz | |
| Shader Units | 8960 | |
| TMUs | 280 | |
| ROPs | 96 | |
| Execution Units | 70 | |
| Ray Tracing Cores | 70 | |
| Tensor Cores | 280 | |
| L1 cache | 128 KB/core | |
| L2 cache | 48000 KB/core | |
| Instructions per Clock | 2 IPC | |
Performance
| Pixel Fill Rate | 235 Gpix/sec | |
| Texture Fill Rate | 687 GTex/s | |
| FLOPS | 43.9 TFLOPS | |
Process
| Interface | PCIe 5.0 x16 | |
| TGP | 300 W | |
| Process | TSMC | |
| Process tech. | 4 nm | |
| Transistor Size | 378 mm² | |
| Transistor count | 45.6 billion | |
| Transistor density | 120.63 MTr/mm² | |
Memory
| Interface | GDDR7 | |
| Memory Frequency | 1750 MHz | |
| Bus | 256 bit | |
| ECC support | No | |
| Effective Memory Speed | 28000 MB/s | |
| Memory Size | 16 GB | |
Max Throughput
API
| DirectX | 12.2 | |
| Vulkan | 1.4 | |
| OpenGL | 4.6 | |
| CUDA | 12.0 | |
| DisplayPort | 2.1b | |
| Ray Tracing | Yes | |
| DLSS | DLSS 4 | |
User Rating
out Points ( Votes)
