Failure: Who's Fault is it Anyway?

This may seem like a pretty strange title for an admittedly short article, but it's definitely an aspect of game design that all designers should consider when both designing a game, and evaluating an implementation.

Most games revolve around the ideas of success and failure. The exact definition of these two things differ from game to game - in some, failure means death and success is progression to the next area, in others, failure means conceding a point, and success is scoring a point.

However, whatever the game, there is a very important point that has to be remembered.

Failure should be the player's responsibility.

By this I mean that failure must be in the player's own hands. If the character the player is controlling fails due to a mistake by the player, that is fine. If the character fails due to an area of design or implementation mistake, then this can destroy an otherwise fine game.

In most of the best games, you really feel that there is a direct link between your ability and your chance of success.

However, in some games, failure just feels like some random act of god. You know what you are SUPPOSED to do, but you have to achieve it in spite of the game, or ones where you feel you have failed because the computer has been unfair in some way. These games tend to annoy rather than frustrate, and is something to be avoided if you can help it. Especially since annoyed players tend to jump on to the net and slag your game off in a fit of rage.

Game mechanics and design can cause the player to fail without fault of his own in a number of ways, and I've included a few examples ranging from mild to incredibly serious. I'll add to these as I think of them :-)

Character Controls

Good character control is precise and responsive, and players can easily manipulate their character through the obstacles. If character control is poor, players can fall off ledges when they were trying to jump, fail to avoid obstacles etc etc. Whilst these things can obviously happen even with the best controls, there is a big difference in the player's mind between "ooh, just failed to make that jump" and "£&%*£% controls!"

Collision Detection

Collision detection should really err on the side of the player, giving them the benefit of the doubt effectively. The problems caused by poor collision detection has changed over the years. In The Old Days (tm), sloppy collision detection meant that the player could be killed by bullets that missed him for example, but more recently the problems are more likely to do with characters getting caught on bits of scenery, This also needs care from the level builders as well as the implementers of the collision detection system, as the problem can occur in either area.

Artificial Intelligence

In many games the AI can directly or indirectly result in the player failing through no fault of their own, whether due to enemy characters having unfair advantages (knowing information they should not etc), or co-operative computer controlled characters performing in such a way as to make the player fail. An example of the first may be a computer's knowledge of the entire terrain before a game starts, or it's ability to know exactly where the player is immediately, as opposed to using the traits of the enemy characters. An example of the second may be the pathfinding of your units in an RTS game, which makes your units wander around the map getting picked off by the enemy instead of actually attacking as a group.

Computer Predicting Player Desires

This could be known as the Word syndrome. It's where the computer seems to try and predict what you are trying to do, and then "helps" you do it, regardless of what you are actually trying to do. My favourite examples of this are many team sport games, where the computer decides which character you should be in control of. Of course, the computer, being helpful, is constantly switching focus of the character you are controlling, which leads to much running in the wrong direction as you desperately try and work out which character you are controlling. I've been playing a lot of ISS Pro Evolution Soccer recently, and whilst being a brilliant football game, I get really frustrated with some of the computer's "helpful" decisions. Another example is auto- weapon switching. Although more games allow the player to override this these days, some games still automatically switch weapons when a new one is picked up/ an old one runs out of ammunition. Helpfully switching the player to a high explosive weapon when in confined quarters is something that most players don't find particularly desirable.

Computer Not Predicting Anything

This is basically the opposite of the problem listed above. Anyone who has played a number of RTS games will have come across games where the computer is quite merrily attacking one of your units/ buildings whilst your defensive units sit nearby, not doing anything. It's certainly a fine line between the computer predicting too much, and not predicting enough, but in some game types, this can be very important.

Unavoidable Consequences

This occurs in a number of games, and is one of my pet hates. By unavoidable consequences, I mean situations where the player has no clue that the result of his actions are going to be detrimental to his character. Some games feature traps or obstacles that are impossible to overcome by skill at the first attempt, simply because the player has no idea that a dangerous situation exists. Some consequences can be fairly minor - the loss of a little health for example, whilst others can be very severe, and result in the end of the game, or necessitate a game reload. This problem often leads to a try and reload scenario, where the player approaches a new area for the first time with the express purpose of finding out HOW to do the section, so they can reload and do it properly the second time around. If, as a designer, you wish to put down traps or obstacles, put some thought into how the player will spot the problem, and how they will work out to avoid it. If you need to put in an instadeath situation, try and make sure the player realises that this will be a consequence.

Automated Camera Control

For some reason, when originally creating this article, I overlooked one of the mainstays of game related failure. Sitting at my computer right now, I still can't believe that this wasn't one of the first points I noted down.

Bad cameras only affect a portion of the gaming market, most notably third person games based in confined environments, but they have a severe effect on those games they do affect. In slower paced games, bad cameras tend to be more a source of frustration than of complete failure. In general, the problems increase as the environments grow more cramped.

Poor camera control can cause player failure by various means, and to various degrees of severity, but most of them are a result of a single problem. The camera isn't showing the information that the player requires. Sometimes this is simply an issue of the camera pointing the wrong way whilst at other times it can be due to objects obstructing the player's view.

There are other problems caused by bad camera systems - the player can become disoriented if the camera is moving rapidly, especially if the control system is based around screen direction as opposed to character direction (i.e. pushing forward moves the character into the screen as opposed to moving the character in the direction it is facing).

Some games allow a certain degree of player control, which can help alleviate some of the issues, but most do not, and it is fairly common to find a player swearing at the screen wondering why they can't see the obstacle they have to avoid or the creature they have to fight, and instead are looking at a blank wall, or their character's head.

I suspect most designers are reading this with amazement, wondering when this article is actually going to provide any scarcely known information instead of stating the obvious. Well, it doesn't really. But like a lot of areas in computer game design and implementation, just because something is obvious or known doesn't mean that everyone always considers it. The visual results of not adhering to the idea that failure should be the player's fault may be small, but the impact can be enormous, even in games where everything else is faultless.