I know. But, how would you retrieve the UI coordinates of the image bounds if it is different at each resolution? As far as I know, there is no way to retrieve in-game values of a dialog from a player's perspective.
I know. But, how would you retrieve the UI coordinates of the image bounds if it is different at each resolution? As far as I know, there is no way to retrieve in-game values of a dialog from a player's perspective.
Look at the map I posted (if you can). No matter what resolution you choose, if you clicked the targets accurately, the image of the cursor will always appear where you click. In essence, it estimates your resolution by sampling your clicks and adjusts accordingly. You can make it more accurate by rounding to the nearest valid resolution and also the number of checkpoints can be reduced to just 1.
As a side note, you mentioned mapping world coords to UI. This is actually possible to some extent. Early in beta I created a mouse tracker, which works by enabling mouse look to a small extent and moving a fake cursor on the screen based on how much the camera rotated. The fake cursor would then be projected from the UI down onto the world (it worked great back then, but was broken by a patch affecting retrieving yaw/pitch in late beta that was never fixed). The reverse should work without any flaws (although I never implemented camera roll to be taken into account).
@s3rius: Go
I know. But, how would you retrieve the UI coordinates of the image bounds if it is different at each resolution? As far as I know, there is no way to retrieve in-game values of a dialog from a player's perspective.
Ask the user to enter his resolution. Nasty, but works.
Look at the map I posted (if you can). No matter what resolution you choose, if you clicked the targets accurately, the image of the cursor will always appear where you click. In essence, it estimates your resolution by sampling your clicks and adjusts accordingly. You can make it more accurate by rounding to the nearest valid resolution and also the number of checkpoints can be reduced to just 1.
As a side note, you mentioned mapping world coords to UI. This is actually possible to some extent. Early in beta I created a mouse tracker, which works by enabling mouse look to a small extent and moving a fake cursor on the screen based on how much the camera rotated. The fake cursor would then be projected from the UI down onto the world (it worked great back then, but was broken by a patch affecting retrieving yaw/pitch in late beta that was never fixed). The reverse should work without any flaws (although I never implemented camera roll to be taken into account).
(removed)