I see. I thought the reason why they say to go up slowly was due to it diminishing life span if gone too far. Now I realize it was because every card is different and top to bottom is more difficult than working from bottom to top. On another side note, if I was to theoretically overclock it to 10ghz, which is the stock memory clock for the 1080, then wouldn't it technically be gddr5x memory at that point? From what I understood, gddr5x is simply gddr5 clocked at a higher speed. I know I probably will never get that type of overclock, however, I am just curious.
The reason for increasing the OC slowly (step by step) is to verify the system actually works at the increased OC.
If we assume a base of 1.0, and a theoretical target of 2.0...
If you go straight to 2.0, and it crashes, you don't where the actual max was.
So you increase step by step.
1.0, 1.1, 1.2, 1.3, 1.4, 1.5, 1.6, 1.7, 1.8 ooohhhh...that's where it crashed. Back off to 1.7 and accept that.