Nvidia GeForce RTX 4070 Ti SUPER vs Nvidia GeForce RTX 3080
72
out 100
VS
59
out 100
Compared 2 processors: Nvidia GeForce RTX 4070 Ti SUPER, vs Nvidia GeForce RTX 3080. 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 | 207 146 140 104 | 168 118 106 78 |
 The Witcher 3 | 225 190 142 72 | 195 162 116 61 |
 Counter-Strike 2 | 292 216 165 85 | 275 206 150 83 |
 Far Cry 6 | 164 149 136 85 | 150 136 114 71 |
 Hogwarts Legacy | 143 120 96 55 | 119 96 76 41 |
 COD Modern Warfare 3 | 191 187 144 100 | 153 147 108 72 |
 Ghost of Tsushima | 130 110 96 58 | 109 90 72 43 |
 Cyberpunk 2077 | 175 151 107 48 | 133 121 80 36 |
 Shadow of the Tomb Raider | 263 247 196 104 | 207 196 151 81 |
| 1080p High | 199 | 168 |
| 1080p Ultra | 168 | 141 |
| 1440p Ultra | 136 | 108 |
| 4K Ultra | 79 | 63 |
| Deviation | Low | Low |
Benchmarks
| Time Spy | 24225 | 17662 |
| Solar Bay | 116030 | 85932 |
| Port Royal | 15768 | 11595 |
| Fire Strike | 56271 | 42429 |
| Night Raid | 174807 | 132288 |
| Background Blur | 331.7 frames/s | 217.3 frames/s |
| Face Detection | 211.6 frames/s | 141.9 frames/s |
| Horizon Detection | 9.63 Gpix/sec | 8.19 Gpix/sec |
| Edge Detection | 13.7 Gpix/sec | 11 Gpix/sec |
| Gaussian Blur | 13.9 Gpix/sec | 8.43 Gpix/sec |
| Feature Matching | 2.15 Gpix/sec | 1.29 Gpix/sec |
| Stereo Matching | 1210 Gpix/sec | 780.7 Gpix/sec |
| Particle Physics | 33566 fps | 22538.5 fps |
| API | OpenCL | OpenCL |
| G2D Mark | 1218 | 1052 |
| DirectX 11 | 280 fps | 207 fps |
| DirectX 12 | 117 fps | 100 fps |
| GPU Compute | 18522 Ops/s | 14559 Ops/s |
| Aztec Ruins (Normal) | 1300.6 fps | 950.5 fps |
| Car Chase | 1119.6 fps | 809.5 fps |
| Manhattan 3.1.1 (1440p) | 943.4 fps | 705.4 fps |
| T-Rex | 4724.8 fps | 3525.4 fps |
| ALU 2 | 8528.7 fps | 6005.6 fps |
| Driver Overhead 2 | 289.1 fps | 283.8 fps |
| API | OpenGL | OpenGL |
| OS | Windows | Windows |
| Tessellation | 8597.7 | 6034.6 |
Tech Specs
Specs & test comparisonGeneral Info
| Manufacturer | Nvidia | Nvidia |
| Build | Discrete | Discrete |
| Announced | January, 24, 2024 | September, 1, 2020 |
| Form Factor | PC use | PC use |
| Use case | Gaming | Gaming |
| Segment | High-end | High-end |
| Architecture | Ada Lovelace | Ampere |
| Codename | AD103 | GA102 |
| Price | 799 | 699 |
GPU
| GPU Frequency | 2340 MHz | 1440 MHz |
| Boost GPU Frequency | 2610 MHz | 1710 MHz |
| Shader Units | 8448 | 8704 |
| TMUs | 264 | 272 |
| ROPs | 96 | 96 |
| Execution Units | 66 | 68 |
| Ray Tracing Cores | 66 | 68 |
| Tensor Cores | 264 | 272 |
| L1 cache | 128 KB/core | 128 KB/core |
| L2 cache | 48000 KB/core | 5000 KB/core |
| Instructions per Clock | 2 IPC | 2 IPC |
Performance
| Pixel Fill Rate | 251 Gpix/sec | 164 Gpix/sec |
| Texture Fill Rate | 689 GTex/s | 465 GTex/s |
| FLOPS | 44.1 TFLOPS | 29.8 TFLOPS |
Process
| Interface | PCIe 4.0 x16 | PCIe 4.0 x16 |
| TGP | 285 W | 320 W |
| Process | TSMC | Samsung |
| Process tech. | 5 nm | 8 nm |
| Transistor Size | 379 mm² | 628 mm² |
| Transistor count | 45 billion | 28 billion |
| Transistor density | 118.73 MTr/mm² | 44.59 MTr/mm² |
Memory
| Interface | GDDR6X | GDDR6X |
| Memory Frequency | 1313 MHz | 1188 MHz |
| Bus | 256 bit | 320 bit |
| ECC support | No | No |
| Effective Memory Speed | 21000 MB/s | 19000 MB/s |
| Memory Size | 16 GB | 10 GB |
Max Throughput
API
| DirectX | 12 | 12 |
| Vulkan | 1.3 | 1.3 |
| OpenGL | 4.6 | 4.6 |
| CUDA | 8.9 | 8.6 |
| DisplayPort | 1.4a | 1.4a |
| Ray Tracing | Yes | Yes |
| DLSS | DLSS 3 | DLSS 2 |
Results
After analyzing the data, we consider Nvidia GeForce RTX 4070 Ti SUPER the clear winner in this comparison.
Survey
Total Votes: 0
