RyanSmithAT: For better or worse, you're going to get the same rate you always have.

RyanSmithAT: HDD read error stats are a bit like Whose Line Is It Anyway: they're (somewhat) made up.

This is a bit above average overall for the GTX 970 FTW, but it does conveniently highlight the fact that even if AMD officially pushed the clock speeds of R9 290XU a bit more, they’d still end up with GTX 970 right next to them. Meanwhile if we push on the gas pedal a bit harder with the GTX 970 FTW, we see the EVGA card pull ahead of the R9 290XU and stock GTX 970 by 8%. Which for NVIDIA is a great outcome, as it means they’re tying a card that's nearly 50% more expensive. The GTX 970 and R9 290XU trade blows from game-to-game, but in the end the two are a tie, just as we see here. This game ends up being a very good summary of what we’re going to see. Only at 1080p does the GTX 970 even begin to trail the R9 290XU. Even at the R9 290XU’s strongest hand – 4K – the GTX 970 is at parity and that remains for 1440p as well.

It seems fitting that we start with a game where the GTX 970 and R9 290XU start out tied. On the other hand it scales well with resolution and quality settings, so it’s still playable on lower end hardware. The original Metro: 2033 was a graphically punishing game for its time and Metro: Last Light is in its own right too.

As always, kicking off our look at performance is 4A Games’ latest entry in their Metro series of subterranean shooters, Metro: Last Light.

How to do metro last light benchmark