[Quillan]

To get a firing solution, you normally have to input certain data: the bearing of the target, range to target, angle on bow, and target speed. The bearing and range tell where the target is. AOB tells what course the target is on, and speed tells how fast it is going. If you're using manual targeting, you have to input all this data yourself, while if you're using auto targeting it's given to the TDC whenever the ship is in the crosshairs of the scope.

The solution is nothing more than a mathematical calculation. You know the position and velocity of the target, you know the position and speed of your torpedo, so the TDC is solving for the heading it must take to collide with the target. Normally, the solution is only good for a few seconds, because the target is moving. Once the data is wrong the solution is wrong. In auto targeting, this isn't an issue because you can just keep the target locked and the data updates constantly.

The position keeper updates the data itself when it's turned on. It basically recalculates the solution constantly based on the original data input. You knew where the target was when you started, the direction it was heading, and the speed it was going. If none of those change, you can get a new firing solution by updating the target position any time in the future, which is what the PK does. Primarily it's good for firing without raising your periscope. You can get your data early, lower the scope, then wait for the optimal firing position without a chance of being observed before firing. It's largely useless if you're using auto targeting, and it overrides the input from the scope when it's on, so leave it off.