Archive | November, 2009

First Person Shooters have tunnel vision and no spine.

24 Nov

With the release of Call of Duty: Modern Warfare 2 millions of gamer are back online playing mock soldier.

These games are also a constant reminder of the shortcomings of the genre when used as a model for real life experience.  One of the most glaring is the lack of peripheral vision. Stare straight ahead and then put your hand up somewhere to the immediate right of your head. You’ll notice how wide your eyes true visual range is in being able to see things that are nearly parallel to your body. The second aspect of the human eye is that it is extremely sensitive to movement.  In a hostile environment the ability to react to movement in our periphery has inevitably saved countless lives from prehistoric times to wars in the modern age.

In FPS games like COD:MW2 however players have a cone of vision that is designed to be much narrower and what you’ll immediately realize is that players are extremely vulnerable from the sides. Walk into a room in Call of Duty and you’ll notice that u cant see much to your side. In games like this you are much more likely to be shot in the side of the head simply because of the limited way in which vision is modeled.  To design peripheral vision into games would require programmers to remodel the way the environment is perceived by the player. I believe that in doing so however would tremendously improve the player experience in first person games.

One other aspect of FPS games pertains more specifically to the way targeting works in consoles. One of the major differences in the way the Xbox/PS3 controller differs from the mouse is that by pressing down on the joystick you tell the game to turn in a direction at a certain rate. To turn 90 degrees to the left then requires a player to press down on the joystick , wait for the visual confirmation that we have achieved our goal then release the joystick. Compare this to the way in which the mouse operates. If you want to turn 90 degrees to the left you simply move the mouse the distance that would require to make that change.  Which is closer to the way in which our body really operates? Imagine pointing a gun forward and needing to aim at something 90 degrees to your left. We certainly don’t tell ourselves to spin counterclockwise at a rate of 3 rotations a minute and then stop once we’re pointing the right way. Instead like using the mouse we tell our body where we would like it to point and how fast we would like it to do so. How can this be solved on the console?

With the advent of the two joystick controller players can now control their characters along two simultaneous motions. Move forward,back, left and right. Pivot the field of view up down left and right. What is interesting though is that while looking up and down does not change the way your character is oriented (it acts like our head would in real life) turning left and right requires the entire character to pivot. It is designed this way because joysticks have a limited range of motion. It would be like if a persons spine were fused and we could not turn from the hip up. Similarly it is impossible to run one direction while facing another. (Strafing exists but is dramatically different due to the speed at which players move)  What does this have to do with the targeting problem above?

Developers should create a secondary movement/targeting system that resembles the way our body turns and pivots from the waist up. It could simply be done by utilizing a button (i.e. left trigger) to act as a SHIFT key. Holding this button down would change the way the right joystick operates. Instead of pivoting the whole body it would simply pivot the game version of our upper torso. Since the range of motion of the upper body is limited we can then map the range of activity to particular amplitudes in the joystick and have it act exactly as we would a mouse. Centering the joystick would place our field of vision and target straight ahead, while pushing the joystick to the upper left would immediately have our character look to the upper left most range. Similarly everything in between would be mapped accordingly and players could train their muscle memory the same way they would if they were holding a real gun.

Many games already do this second target system, but it is used most often in a game that is trying to simulate some kind of turret system. The Mechwarrior genre of video games was one which used this “upper torso” action as their primary method of targeting. By improving the current Console movement system with the upper torso action we can create a better simulation of how the human body works and improve the game mechanics as well.


Fourth and 2

16 Nov

Last night at the end of the Patriots-Colts game, leading by 6 with a little more than 2 minutes to play Bill Belichick decided to go for a first down at their own 34 yard line. Conversion fails, Colts Score, Sports commentating mayhem ensues.

Let us at least look at the basic problem being proposed in this case. The simplest way to look at this is the following equation:  You go for it if  x + (1-x)*y >  z

x: probability of Patriots converting the 4th down.

y: probability of Colts failing to score a TD from 34 yards.

z: probability of Colds failing to score a TD from the 70-80 yards.

If the Patriot’s chances of converting combined with their chances of stopping the colts when they fail to is greater than the chances of the colts scoring from further then they should go for it. For various values of x,y,z that are very realistic going for it proves to be a good decision.

We can complicate it by asking what x,y,z really represent. What’s we’re implicitly weighing is the efficacy of the (Patriots Offense vs Colts Defense)  against the (Colts Offense vs Patriots Defense)  The better both teams are performing offensively the better the result for going for it versus punting. Another factor that arises in this question of going for it is the the value/penalty of having to go an extra 40-50 yards.  y is essentially the Touchdown function at 34 yards. Whereas z is the Touchdown function at 70 yards. (y= f(34) , y=f(70) ) Yards to end-zone have a diminishing return. Going from the 5 yard line to the 10 yard line is a much greater penalty than going from the 45-50. For better offensive teams the penalty for each incremental yard is less than it would be for poor offensive teams. (In the extreme scenario the Colts + Raiders both have a HIGH likelihood of scoring from 1 yard out, but the Colts have a significantly greater likelihood from scoring 50 yards out than the Raiders would)

I don’t propose to have an answer as to which is actually the correct decision but simply ask that people analyze and consider the different factors and solutions in this scenario. It drives me crazy that  most of the sporting world refuses to do any analysis but instead exclaim “Play the percentages and punt.” Exactly what percentages are people referring to here. Surely there must be a study that the coaches, sports writers and radio talk show hosts have all studied. My guess is that they simply relied on conventional practices and based on history believe them to be best.

Boundaries of Imagination

16 Nov

The blog’s name comes from the name of one of my favorite trance albums by Armin Van Buuren. The name aptly describes what this blog will explore, ideas especially new ideas. The whole of human history has been the expression in one form of new thoughts and previously unimagined possibilities.