Sections
Mouse and Joypad in Video Games
Comparison of the two input methods and possible explanations for the success of specific game genres in consolles and personal computers
This article as been adapted and expanded from a forum post I wrote in the MacityNet.it website.
The standard input controller that ships with consolles like Nintendo or X-Box is, nowadays, a joypad with a simple four-directions controller, some buttons and, since few years ago, also a small cloche. My idea is that the choice of this kind of joypad has completely excluded some kind of games from the consolle market, more specifically all the games requiring a fast and accurate movement of a pointer.
We can find at least two type of video games: games where the player has to rapidly aim an object on the screen and those where a continuous control is preferred. The first type of games is represented by strategy games, where the player has to select and move different units shown on the screen, and the "shoot'em all" games, where the player has to repeatedly aim and shoot targets moving on the screen. The second kind of games are simulation games, for example car simulators and flight simulators.
This idea came out for the first time when I played at "Halo" on a X-Box, since the game seemed to me quite "slow" and aiming seemed complicated: instead of moving the mouse as fast as I wanted to aim a pecific point on the screen, I had to carefully move the cloche to tell the game where to move the aim and I had to continuosly check on the monitor the correctness of my movements. Playing more with it helped a lot, but it was still more complicated than using the mouse at the PC.
I thought about it and I came up with the following explanation: using the mouse forces the brain to correlate the hand linear movement with the pointer (or aim) movement, the relation is a simple linear function. No additional thinking is required, once you know the factor, you can automatically know where the pointer will be after the hand movement. This process has been highly optimised during millennia of evolution.
Obviously, I refer here only to games, where the function is really linear, but in modern operating systems (and in Mac OS since a long time ago) the movements of the pointer are not linear but they depend on the hand speed.
The joypad, on the opposite, relates the movement of the hand to the "speed" of the movement of the pointer on the screen. This different process forces the brain to either:
- calculate the expected pointer position using the movement of the hand and the time elapsed during this movement (in other words, to calculate the integral in the time of the hand movement multiplied by a constant);
- continuosly check the pointer position in the screen and consequently correct the hand position, until the pointer reaches the final position (this is a feedback control).
Both solutions are slower and less accurate than using a direct relation, as when the mouse is involved:
- calculating the integral is possible (not consciously, of course) and is probably what experienced players do, but it is clearly slow. I think a direct match between a joypad user and a mouse user would lead to a quick victory of the latter;
- the feedback control is extremly slow, since it depends on the eye-hand response time: the brain takes time to recognise the movement of the hand by seeing the pointer on the screen (first delay), then it must calculate the corrections (second delay) and then send the "commands" to the hand (third delay). The sum of them often becames too big for a realtime shooting or action game.
What about the joypad, then? I think it is a clear winner every time the user needs a continuous control of an object during the time and when the control must automatically be reset to the initial position when not used. Simulators are the ideal applications: the steering wheel needs to be reset, the airplane cloche need to be stabilised. A mouse is not able to do that automatically, but I found the following solutions in some games:
- the mouse position becomes the same of the joypad position, for example, the center of the screen is the reset position. The mouse becomes just like a joypad with too much dust, blocked where you leave it. This solution was chosen by Microsoft Flight Simulator (at least version 4). This solution is acceptable if you can see the pointer, but becomes useless if you cannot check visually the position of the pointer (you'll never know if it is really centered if not after some minutes). I landed on a completely different city...
- the joypad position is simulated with the speed of the mouse movement: mouse standing still means reset position, mouse moving slowly on the left mean joypad just a bit on the left, mouse moving fast on a direction means joypad pushed a lot in that direction. I really found a game using this method of input (as secondary input), but I don-t remember the title. I think it was a car simulator. I remember that I had to move the mouse, then lift it and put it back in the mouse pad, then move it, ...
In the PC world, where there are few constraints, the joypad has always been there, since the four direction keys are simply a joypad without radial sensitivity (the first controller with it was the Playstation controller, I think). With the keyboard, pressing the keys with different speeds partly solves this problem.
Of course, action games can be played with joypads as well, but it seems there are less games of that genre for consolles than for PCs. The difference is maximum with strategy games, since selecting and moving units around the monitor with good precision and speed is quite difficult with joypads.
Author: Olaf Marzocchi
First version: October 2006.