NVIDIA GeForce RTX 3080 Ti vs NVIDIA GeForce RTX 3090
What is the difference between NVIDIA GeForce RTX 3080 Ti and NVIDIA GeForce RTX 3090. Find out which graphics card has better performance.
Benchmarks
| 26758 | PassMark G3D | 25855 |
| 1042 | PassMark G2D | 989 |
| 17532 | 3DMark Time Spy | 18067 |
| 8999 | 3DMark Time Spy Extreme | 9397 |
| 33084 | 3DMark Fire Strike | 32959 |
| 21580 | 3DMark Fire Strike Extreme | 21878 |
| 12319 | 3DMark Fire Strike Ultra | 12550 |
| 100577 | 3DMark Wild Life | 98491 |
| 42414 | 3DMark Wild Life Extreme | 43705 |
| 61444 | 3DMark Night Raid | 60794 |
Graphics Processor (GPU)
| GA102 | GPU Name | GA102 |
| GA102-225-A1 | Model number | GA102-300-A1 |
| Ampere | Architecture | Ampere |
| Samsung | Foundry | Samsung |
| 8 nm | Process Size | 8 nm |
| 28,300 million | Transistors | 28,300 million |
| 628 mm² | Die Size | 628 mm² |
Graphics Card
| May 31st, 2021 | Release Date | Sep 1st, 2020 |
| GeForce 30 | Family | GeForce 30 |
| Active | Production | Active |
| 1,199 USD | Launch Price | 1,499 USD |
| Jun 3rd, 2021 | Availability | Sep 24th, 2020 |
| GeForce 20 | Predecessor | GeForce 20 |
| PCIe 4.0 x16 | Bus Interface | PCIe 4.0 x16 |
Memory
| 12 GB | Memory Size | 24 GB |
| GDDR6X | Memory Type | GDDR6X |
| 384 bit | Memory Bus | 384 bit |
| 912.4 GB/s | Bandwidth | 936.2 GB/s |
Performance
| 186.5 GPixel/s | Pixel fillrate | 189.8 GPixel/s |
| 532.8 GTexel/s | Texture fillrate | 556.0 GTexel/s |
| 34.10 TFLOPS (1:1) | FP16 (half) performance | 35.58 TFLOPS (1:1) |
| 34.10 TFLOPS | FP32 (float) performance | 35.58 TFLOPS |
| 532.8 GFLOPS (1:64) | FP64 (double) performance | 556.0 GFLOPS (1:64) |
Clock Speeds
| 1365 MHz | Base Clock | 1395 MHz |
| 1665 MHz | Boost Clock | 1695 MHz |
| 1188 MHz 19 Gbps effective | Memory Clock | 1219 MHz 19.5 Gbps effective |
Render Config
| 10240 | Shading Units | 10496 |
| 320 | TMUs | 328 |
| 112 | ROPs | 112 |
| 80 | RT Cores | 82 |
| 128 KB (per SM) | L1 Cache | 128 KB (per SM) |
| 6 MB | L2 Cache | 6 MB |
| 80 | SM Count | 82 |
| 320 | Tensor Cores | 328 |
Board Design
| Dual-slot | Slot Width | Triple-slot |
| 285 mm 11.2 inches | Length | 336 mm 13.2 inches |
| 112 mm 4.4 inches | Width | 140 mm 5.5 inches |
| 40 mm 1.6 inches | Height | 61 mm 2.4 inches |
| 350 W | Thermal design power (TDP) | 350 W |
| 750 W | Suggested PSU | 750 W |
| 1x HDMI 3x DisplayPort | Display Connectors | 1x HDMI 3x DisplayPort |
| 1x 12-pin | Power Connectors | 1x 12-pin |
| PG132 SKU 18 | Board Number | PG132 SKU 10 |
API support
| 12 Ultimate (12_2) | DirectX | 12 Ultimate (12_2) |
| 4.6 | OpenGL | 4.6 |
| 3.0 | OpenCL | 3.0 |
| 1.2 | Vulkan | 1.2 |
| 6.6 | Shader Model | 6.6 |
| 8.6 | CUDA | 8.6 |