DLSS frame generation takes two frames from a game plus the mathematical data about what went into making them and feeds them into an AI and asks it to guess what the frame in between would have looked like if there was one.
That’s more efficient than actually rendering the extra frame and uses a different bit of the graphics card, but using it for too many frames causes noticeable artifacts because it starts making wrong guesses if the difference between the two frames it is given are too big.
Latest Answers