Annomander wrote...
Human reactions data is skewed, as humans in test situtaions are invariably reacting to something which they know is going to occur. Reacting in real time to things which you do not have prior knowledge of yields much less impressive reaction times.
They ran some tests on a bunch of Q3/UT2k4/CS players back in 2006 or something at a LAN event and found that most were able to twitch react at an average of around 250ms.
Human reaction data is skewed. In context of the crusader, problem, though, the 250ms reaction times are wrong because they are, as I said, "randomized" visual stimulus followed by reaction. In context of the crusader, the important part is our ability to predict the firing time (press button, get response from game) during the twitch reaction (a subreaction, if you will) which we can do very precisely - once you start the twitch reaction, you body/mind knows the time it'll take you to arrive at that location so that you can start a firing action at the first possible moment, to a precision in the range of noticing a 100ms delay, if you are expecting closer to 20.