GeForce RTX 3090 Ti vs GeForce RTX 4080
Dashed outline = score at launch
Why GeForce RTX 3090 Ti is better
- VRAM: 24 GB GDDR6X vs 16 GB GDDR6X(50% more)
- Memory Bus: 384-bit vs 256-bit(50% more)
- Bandwidth: 1,008 GB/s vs 717 GB/s(41% more)
- CUDA Cores: 10,752 vs 9728(11% more)
Why GeForce RTX 4080 is better
- Boost Clock: 2505 MHz vs 1.86 GHz(1346.8× more)
- TDP: 320 W vs 450 W(41% less)
Verdict
GeForce RTX 4080 leads outright in 6 of 6 dimensions today.
Spec by spec
| Spec | GeForce RTX 3090 Ti | GeForce RTX 4080 |
|---|---|---|
| Core Specs | ||
| CUDA Cores | 10,752● | 9728 |
| Boost Clock | 1.86 GHz | 2505 MHz● |
| Process Node | Samsung 8nm | — |
| Memory | ||
| VRAM | 24 GB GDDR6X● | 16 GB GDDR6X |
| Memory Bus | 384-bit● | 256-bit |
| Bandwidth | 1,008 GB/s● | 717 GB/s |
| Power & Thermal | ||
| TDP | 450 W | 320 W● |
| Recommended PSU | 1000 W | — |
| Connectivity | ||
| PCIe Interface | PCIe 4.0 x16 | — |
| Core | ||
| CUDA Cores | 10,752● | 9728 |
| Boost Clock | 1.86 GHz | 2505 MHz● |
| Architecture | — | Ada Lovelace |
| Power & Output | ||
| TDP | 450 W | 320 W● |
| Outputs | — | 3x DP, 1x HDMI |