Key Differences
Advantages of GeForce RTX 5070 Ti
Advantages of GeForce RTX 3060
Specifications
Performance
| TGP | 300 W | 170 W |
Storage Details
| Bus | 256-bit | 192-bit |
Ports
| DisplayPort | 2.1b | 1.4a |
Other
| Forza Horizon 5 | 233165154123 | 114675740 |
| The Witcher 3 | 23820814979 | 1241007240 |
| Counter-Strike 2 | 29922716883 | 1851329549 |
| Far Cry 6 | 18016214588 | 94846435 |
| Hogwarts Legacy | 16313410860 | 68544220 |
| CoD: Modern Warfare III | 214206164115 | 80744932 |
| Ghost of Tsushima | 15012410868 | 64513519 |
| Cyberpunk 2077 | 19117012156 | 68623819 |
| Shadow of the Tomb Raider | 283268201107 | 1101017037 |
| 1080p High | 217 | 101 |
| 1080p Ultra | 185 | 81 |
| 1440p Ultra | 146 | 58 |
| 4K Ultra | 87 | 32 |
| Margin of Error | Low | Low |
| Time Spy | 27784 | 8751 |
| Solar Bay | 136303 | 41040 |
| Port Royal | 19439 | 5144 |
| Fire Strike | 67197 | 22307 |
| Wild Life Extreme | 57631 | 17727 |
| Night Raid | 190729 | 98329 |
| Background Blur | 297.9 img/sec | 149.6 img/sec |
| Face Detection | 218.6 img/sec | 91 img/sec |
| Horizon Detection | 10.1 Gpixels/sec | 4.07 Gpixels/sec |
| Edge Detection | 13.9 Gpixels/sec | 5.58 Gpixels/sec |
| Gaussian Blur | 14.8 Gpixels/sec | 3.92 Gpixels/sec |
| Feature Matching | 2.03 Gpixels/sec | 1.01 Gpixels/sec |
| Stereo Matching | 1290 Gpixels/sec | 372.9 Gpixels/sec |
| Particle Physics | 35438.3 FPS | 12226.3 FPS |
| API | OpenGL | OpenGL |
| G2D Mark | 1342 | 965 |
| DirectX 11 | 303 FPS | 128 FPS |
| DirectX 12 | 126 FPS | 69 FPS |
| GPU Compute | 18821 Ops/s | 8307 Ops/s |
| Aztec Ruins Normal Tier | 1450 FPS | 555.5 FPS |
| Car Chase | 1281.4 FPS | 492.3 FPS |
| Manhattan 3.1.1 (1440p) | 983.7 FPS | 407.9 FPS |
| T-Rex | 4055.8 FPS | 1940.6 FPS |
| Tessellation | 10135.4 FPS | 2965.4 FPS |
| ALU 2 | 9247 FPS | 2862.2 FPS |
| Driver Overhead 2 | 251.7 FPS | 295.2 FPS |
| OS | Windows | Windows |
| Image Classification (SP) | 14913 | 6645 |
| Image Segmentation (HP) | 45199 | 15338 |
| Image Super Resolution (Q) | 39360 | 14283 |
| Face Detection (HP) | 84188 | 28862 |
| Pose Estimation (Q) | 161400 | 61313 |
| Text Classification (SP) | 4010 | 2463 |
| Machine Translation (HP) | 7973 | 3793 |
| Object Detection (SP) | 18934 | 7712 |
| Depth Estimation (Q) | 67761 | 25461 |
| Style Transfer (SP) | 335879 | 123233 |
| Framework | ONNX | ONNX |
| Backend | DirectML | DirectML |
| Vendor | Nvidia | Nvidia |
| Build | Discrete | Discrete |
| Released | January 7, 2025 | January 12, 2021 |
| Launch price (MSRP) | $749 | $329 |
| Case | Desktop | Desktop |
| Purpose | Gaming | Gaming |
| Segment | Mid-range | Mid-range |
| Architecture | Blackwell 2.0 | Ampere |
| GPU Codename | GB203-200-A1 | GA106 |
| Rival Equivalent | Radeon RX 9070 XT | Radeon RX 6700 XT |
| Base Clock | 2295 MHz | 1320 MHz |
| Boost Clock | 2452 MHz | 1777 MHz |
| Shading Units | 8960 | 3584 |
| Texture Mapping Units (TMUs) | 280 | 112 |
| Render Output Units (ROPs) | 96 | 48 |
| Compute Units (Pipelines) | 70 | 28 |
| Tensor Cores | 280 | 112 |
| Ray-tracing Cores | 70 | 28 |
| L1 Cache | 128KB per cluster | 128KB per cluster |
| L2 Cache | 48MB shared | 3MB shared |
| Instructions Per Cycle | 2 IPC | 2 IPC |
| Pixel Fill Rate | 235 GPixel/s | 85 GPixel/s |
| Texture Fill Rate | 687 GTexel/s | 199 GTexel/s |
| Interface | PCIe 5.0 x16 | PCIe 4.0 x16 |
| Manufacturing | TSMC | Samsung |
| Fabrication Process | 4 nm | 8 nm |
| Die Size | 378 mm² | 276 mm² |
| Transistor Count | 45.6 billion | 12 billion |
| Transistor Density | 120.63 MTr/mm² | 43.48 MTr/mm² |
| Memory Type | GDDR7 | GDDR6 |
| Memory Size | 16 GB | 12 GB |
| Memory Clock | 1750 MHz | 1875 MHz |
| Effective Memory Speed | 28000 Mbps | 15000 Mbps |
| ECC | No | No |
| DirectX | 12.2 | 12 |
| Vulkan | 1.4 | 1.3 |
| OpenGL | 4.6 | 4.6 |
| OpenCL | 3.0 | 3.0 |
| CUDA | 12.0 | 8.6 |
| Ray Tracing | Yes | Yes |
| DLSS | DLSS 4 | DLSS 2 |
| Max. Temperature | - | - |