GeForce RTX 3080

GPU: Ampere (GA102) | GPU Cores: 8704 | Boost Clock: 1,710 MHz | Video RAM: 10GB GDDR6X 19 Gbps | TDP: 320 watts

Nvidia’s GeForce RTX 3080 sports the new and improved Ampere architecture. It’s over 30% faster than the previous gen 2080 Ti, for $500 less. The new RTX 3080 Ti didn’t manage to supplant the incumbent, thanks to its significantly higher pricing. If you’re serious about maxing out all the graphics settings and you want to play at 4K or 1440p, this is the card to get — it’s mostly overkill for 1080p gaming, though enabling all ray tracing effects in games that support the feature makes 1080p still reasonable.

If you skipped the first round of RTX GPUs, the RTX 30-series might finally get you you on board the ray tracing train. With potentially double the ray tracing performance of Turing, and games like Cyberpunk 2077 using even more ray tracing effects, the RTX 3080 is your best bet at playing games in all their ray traced glory without nuking the piggy bank.

Ampere also brings improved tensor cores for DLSS, a technology we’re bound to see more of in future games now that it doesn’t require per-game training by a supercomputer. We’re seeing a lot more games with DLSS 2.0 these days, helped by the fact that it’s basically a toggle and UI update to get it working in Unreal Engine. Nvidia’s RT and DLSS performance are also quite a bit faster than what you get from AMD’s new RX 6000 cards, which is a good thing as Nvidia sometimes falls behind in traditional rasterization performance (which is what our raw numbers are based on).

The biggest problem with RTX 3080 by far is going to be finding one in stock, at prices that aren’t straight up terrible. Rumors swirling around the upcoming RTX 3080 Ti suggest it doubles the VRAM to 20GB, but other rumors indicate the 20GB cards probably won’t come out for a while due to component shortages