If (Key1ToHoldDown = %Key1ToHoldDownPrev%) and (Key2ToHoldDown = %Key2ToHoldDownPrev%) If the same keys are held down, Key2ToHoldDownPrev = %Key2ToHoldDown% Prev now holds the key that was down before (if any). Key1ToHoldDownPrev = %Key1ToHoldDown% Prev now holds the key that was down before (if any). GetKeyState, JoyU, JoyU Get position of U axis. GetKeyState, JoyV, JoyV Get position of V axis. GetKeyState, JoyY, JoyY Get position of Y axis. GetKeyState, JoyX, JoyX Get position of X axis. I suppose the 30/70 conditionals aren't necessary, then, but they don't hurt. Only one axis can be active at a time, and its value when active is always either 0 or 100. It seems to work, already, but I wanted to get critiques to see if there's a better way to go about the same results, or to make it faster, smaller, or more efficient.Įdit: I should mention, the pad on this device is digital. This should maximize smoothness in my games. like, when switching from pressing NE to N on the d-pad, the script should be holding Up and Right, then release only Right, keeping Up held down. I also wanted the script to respond sensibly when switching directions - that is, mimicking how I would press the keys. Then the V axis for NW-SW and the U axis for NE-SE. The usual axes, X and Y, for left-right and up-down. My fix involves telling the n52te to instead send joystick axis messages. The best it can do is send rapidly alternating keypresses, rather than actually sending two keypresses held down for a real diagonal (W and D, or Up and Left, for instance). The problem is that the software it comes with can't send anything which any game translates into a proper diagonal. As you may or may not know, the Belkin n52te keypad/gamepad, the successor of the excellent n52, was released with a completely nonfunctional d-pad.

Belkin n52te installation software