Anyone else befuddled by how similar, and similarly expensive, all the top-end cards are? The numbers start to all smeg together after a while of staring. I just bodged this together in 10 minutes, as a quick reference to keep it straight in my head, and thought it might be useful to others. Interesting (but not surprising) to see value still slopes off horribly in the top tier. 'Twas ever thus, right? Those last 3 prices were hard-to-find outliers, too, average prices were about £100 higher than that. I used to do big, sprawling charts that averaged across several games and all that shite. This one is just taken from the Bit-Tech Battlefield 1 3840x2160 numbers, averaging min. and avg. FPS. It's Price over FPS *10, so a higher value coefficient is better. This was actually a hugely beneficial exercise for me, because my first thought while browsing eBuyer was "ooh! A 1080 is only £490 now, that's cheap!" Turns out it's not cheap for the performance - it's just cheap relative to its original - horrible - price. [common sense] Bear in mind that there will be minimum acceptable FPS for your particular games and resolution, so figure out that minimum, draw a line above the cheapest card that meets it and don't bother with the ones above the line. A great value card that produces 20fps in your game of choice is still only producing 20fps, no matter how cheap each of those 20 frames were it's still a crap result. [/common sense] If you hate my lazy-a$$ method or want to plug your own values in, I've left the Google Sheet open here, you can download an XLS copy and go nuts: https://docs.google.com/spreadsheets/d/1v3Idal81Qgr_Ho7REuPWm9gEkSaaHVir1gbpBzmVjvI/edit?usp=sharing edit- oh and the 1070 Ti isn't on there because really, who gives a ****. edit - okay I admit it, the real reason isn't prejudice, it's just that there was no test result for it.