To start off my testing I always prefer to take a look at the synthetic benchmarks. They don’t translate directly to an FPS that you can expect in game, but they are repeatable tests that make it a little easier to compare card to card. They also give us a good look at what to expect in the rest of our testing. My first tests were using 3DMark. I tested using all three Fire Strike benchmarks, those cover 1080p, 1440p, and 4k performance in DX11. Then the Time Spy benchmarks are DX12 focused and test 1080p and 1440p. In all three of the DX11 focused tests the RTX 2080 FE came in a little below the GTX 1080 Ti. Having already tested the RTX 2080 Ti I did know that DX11 performance isn’t as good as DX12 performance on the new cards, but I was still surprised that this card didn’t at least match the 1080 Ti, it came close on the 1080p performance test, but the other two the 1080 Ti has a slight gap. The Time Spy benchmarks, on the other hand, show a completely different picture. In those, the RTX 2080 is out to a 19% lead. That said the RTX 1080 Ti is still much farther ahead here with a 43% lead over the 1080 Ti and a 34% lead over the RTX 2080.
With Unigine Superposition being DX11 based, after seeing the 3DMark results I wasn't surprised that the 1080 Ti and the 2080 were neck and neck here. The only exception was with the 1080p Extreme test where the RTX 2080 FE was much faster than the GTX 1080 Ti.