Quick/Comparisons
A mid-range Nvidia GeForce RTX 3080 Ti graphics card with capable workstation performance (64/100), higher power consumption. Suited for 1080p gaming and everyday tasks.
| Gaming | 60 /100 |
| Workstation | 64 /100 |
| AI/ML | 46 /100 |
| Energy Efficiency | 35 |
Looking for more comparisons?
Compare GeForce RTX →| Quick Comparison Final Score | 59 /100 |
| Graphics Processor | GA102 |
| Cores | 10240 |
| TMUs / ROPs | 320 / 112 |
| Blender GPU | 5565.9 |
| Forza Horizon 5 | 128 |
| The Witcher 3 | 174 |
| Counter-Strike 2 | 205 |
| Far Cry 6 | 139 |
| Hogwarts Legacy | 108 |
| Call of Duty | 161 |
| Ghost of Tsushima | 96 |
| Cyberpunk 2077 | 128 |
| Tomb Raider | 219 |
| Average FPS | 151 |
| 1080p High | 180 |
| 1080p Ultra | 151 |
| 1440p Ultra | 117 |
| 4K Ultra | 69 |
| Margin of Error | Low |
| GB6 Compute Score | 196756 |
| Background Blur | 239.8 img/sec |
| Face Detection | 156.8 img/sec |
| Horizon Detection | 10.8 Gpixels/sec |
| Edge Detection | 14.8 Gpixels/sec |
| Gaussian Blur | 11.9 Gpixels/sec |
| Feature Matching | 1.53 Gpixels/sec |
| Stereo Matching | 889.6 Gpixels/sec |
| Particle Physics | 26054.6 FPS |
| API | OpenCL |
| G3D Mark Score | 26777 |
| G2D Mark | 1095 |
| DirectX 11 | 222 FPS |
| DirectX 12 | 109 FPS |
| GPU Compute | 15025 Ops/s |
| GB6 ML Single Precision | 25945 |
| GB6 ML Half Precision | 38430 |
| GB6 ML Quantized | 15331 |
| Image Classification (SP) | 9747 |
| Image Segmentation (HP) | 26654 |
| Image Super Resolution (Q) | 21475 |
| Face Detection (HP) | 49813 |
| Pose Estimation (Q) | 135370 |
| Text Classification (SP) | 2796 |
| Machine Translation (HP) | 4757 |
| Object Detection (SP) | 12924 |
| Depth Estimation (Q) | 36756 |
| Style Transfer (SP) | 274032 |
| Framework | ONNX |
| Backend | DirectML |
| Pixel Fill Rate | 186 GPixel/s |
| Texture Fill Rate | 533 GTexel/s |
| FLOPS (FP32) | 34.1 TFLOPS |
| Memory Type | GDDR6X |
| Memory Size | 12 GB |
| Memory Clock | 1188 MHz |
| Effective Memory Speed | 19000 Mbps |
| Bus | 384-bit |
| ECC | No |
| Memory Bandwidth | 912 GB/s |
| Interface | PCIe 4.0 x16 |
| TGP | 350 W |
| Manufacturing | Samsung |
| Fabrication Process | 8 nm |
| Die Size | 628 mm² |
| Transistor Count | 28 billion |
| Transistor Density | 44.59 MTr/mm² |
| DirectX | 12 |
| Vulkan | 1.3 |
| OpenGL | 4.6 |
| OpenCL | 3 |
| CUDA | 8.6 |
| Ray Tracing | Yes |
| DLSS | DLSS 2 |
| DisplayPort | 1.4a |
| Vendor | Nvidia |
| Build | Discrete |
| Released | January 3, 2021 |
| Case | Desktop |
| Purpose | Gaming |
| Segment | High-end |
| Architecture | Ampere |
| GPU Codename | GA102 |
| Recommended CPU | -Intel Core i9 12900Kor above |
| Steel Nomad Lite Score | 21408 |
| Time Spy | 19632 |
| Solar Bay | 96597 |
| Port Royal | 13265 |
| Fire Strike | 47867 |
| Wild Life Extreme | 42716 |
| Night Raid | 144955 |
| Base Clock | 1365 MHz |
| Boost Clock | 1665 MHz |
| Shading Units | 10240 |
| Texture Mapping Units (TMUs) | 320 |
| Render Output Units (ROPs) | 112 |
| Compute Units (Pipelines) | 80 |
| Tensor Cores | 320 |
| Ray-tracing Cores | 80 |
| L1 Cache | 128KB per cluster |
| L2 Cache | 6MB shared |
| Instructions Per Cycle | 2 IPC |
84/100