• NAME:

• There are 5 questions, on 8 pages (including this cover page and two blank pages at the end of the exam).
  Please check to ensure your exam is complete.

• Each question is worth 9 marks, attempt all questions.

• Answer directly on the exam paper. If more space is required, use the backs of the pages and/or the two blank pages at the end of the exam.

• The questions are quite open-ended, so be sure not to spend excessive amounts of time on one question at the expense of the rest of the exam.

• The exam is open book, open notes. No calculators or other electronic devices are permitted.

• You have eighty (80) minutes to complete the exam.

• Good luck!

Question 1: Game challenge and difficulty

"A player must always feel like the failure of a challenge is entirely his own responsibility, and not a fault of a poorly designed product."

"The player must understand how and why he failed, so that he can learn from his mistake and increase the feeling of failure being his responsibility."

Provide reasoned arguments either in support of or in opposition to those two ideas.

Sample solution: While it may sound like a harsh way of phrasing it, the logic is that when the player wins or loses they should feel like it is either their accomplishment or their loss. Wins should come with a sense of accomplishment, not feeling 'cheap' or easy, and conversely losses shouldn't come with a feeling that they were unfair or that the situation was unwinnable. Many different factors can contribute to the player's feeling that it wasn't their fault they lost, or that it wasn't really their skill that led them to a win. Prevalent factors include the scenario being poorly balanced, poorly explained, placing too high a reliance on luck or random chance, bugs (whether damaging to the player or exploitable by the player), and faulty design of levels or scenarios leading to dead ends or unwinnable situations. When the player does lose, there should be a clear lesson in the loss - giving the player a better idea of how to win in the future. Sometimes the lesson is one of timing (e.g. your character needs to be more advanced before tackling this opponent), sometimes it may be a lesson of how to attack a particular situations (e.g. stealth versus brute force), sometimes it may be a lesson in balancing short term needs versus long term planning, but it should always be clear that there is a better way of tackling the current situation. This does drive home that the player chose poorly somehow, but gives them the knowledge base to choose more wisely in the future.

Question 2: Risk and reward

Many games increase tension by encouraging players to pursue paths that could become mutually self destructive.

Perhaps the simplest example is "chicken", where two players race towards one another risking a head on collision: if one player turns away the other gains a small victory, but if neither turns away they both suffer a major loss.

Discuss how similar ideas can be used on a much larger scale to produce tension in mainstream games.

Sample solution: In the 'chicken' example there are three basic outcomes: one player turns away, for a small loss, giving the other a small victory both players turn away, each getting a small loss neither player turns away, both suffer catastrophic losses. The player is left with the choice of taking the safe route, knowing it means a small loss, or accepting a significant risk to make a small gain. In many games the player fully expects to put their character's life and limb on the line over and over for a series of small gains, expecting them to build into a significant advantage over time and accepting the possibility of a big loss - surviving the risks is a major part of the appeal of most games. The big questions are how big to make the risks and how big the associated rewards should be. Another example arises when there are multiple mutually-hostile opponents to deal with - do players A and B attack one another in a fight to the death over some resource, knowing that the winner may be too weak to fight off other opponents afterwards? Yet another related example is the 'prisoner's dilemma', in which two players are independently (and simultaneously) given the chance to betray one another.   A betrays B A doesn't betray B B betrays A Each gets 5 year sentence A gets 10 years, B gets none B doesn't betray A B gets 10 years, A gets none Each gets 1 year sentence Betraying the other player means you get no penalty or 5 years, not betraying them means you get 1 year or 10 years. If your opponent's behaviour was completely random then choosing betrayal would seem like the sensible approach, but if you both think that way theyn you're both in trouble, while if you both think cooperatively then you both get off lightly. ... which should you try?

Question 3: Interfaces

In lectures, we discussed several different approaches to visual interfaces for games:

1. minimalist interfaces, in which no actual controls are shown on-screen
2. total control interfaces, in which controls for all player actions are always on screen
3. accessible option interfaces, in which access to the seperate control modes or panels are always on screen

For each of the three modes, describe a style of game and a target audience the mode would be most appropriate for, and a a style of game and a target audience the mode would be inappropriate for.

Sample solution
Interface style	Appropriate game/audience	Inappropriate game/audience
Minimalist	Point-and-click games, games with just a handful of control actions that are fairly intuitive	Any game with a complex set of controls that would need to be memorized
Total control	Games with complex control sets but fast required response times (e.g. detailed flight/racing/combat simulators)	Games with too many controls, options, and abilities to realistically squeeze onto a single display simultaneously
Accessible options	Complex turn-based strategy games, with substantial short and long term planning but without the pressures of real-time responses	Games requiring fast reaction times across a wide range of actions (i.e. where searching through the menus is too slow to be viable)

Question 4: Game balance

Dynamic balancing is the process of automatically updating the difficulty of a game to keep the game interesting - i.e. keep it from getting too easy (boring) or too difficult (frustrating).

This can involve giving the player or the AI a wide range of handicaps or advantages, increasing or decreasing the hazards a player must face, etc. However, dynamically altering the balance needs to be done carefully, otherwise the player's actual skill/involvement seems less important ("If I play really poorly the game will just dumb down the AI for me, or if I play really well the game will give the AI a huge boost that wipes out the advantage I gained through all my hard work.")

In the context of a the game you are developing for your course project, discuss ways in which the game difficulty can be dynamically balanced to keep interest levels up, but without making the dynamic balancing obvious to the player.

Sample solution: This one is tough to generalize on since you're asked to answer in terms of your project, but some considerations include: To dynamically scale the difficulty, you need to be tracking some set of player stats that indicate how 'strong' they are at the moment, and a strength range you expect as typical/reasonable for the players at this point in the game. When players fall outside that range, you apply your scaling ideas to make the game more/less difficult. Make sure that any alterations aren't directly observable by the player. For example, if the player can see something like 'hit points' for their opponent, and Orcs suddenly start showing up with 12 hit points instead of 20, they'll know something fishy is going on. Even if they just see a health bar for the opponent but recognize that the health bars are running down quicker than before then there could be a problem.a If you do include an observable change in the opponents' statistics, then make sure a believable rationale exists for it (rumours that the enemy is running out of troops, and is relying on less trained or less fit individuals, etc). In terms of scaling the strength of the opposition, possibilities include varying the frequency, composition, armament, or sheer numbers of random opponents. You could also scale the frequency with with the player encounters aid, inns, rest stops, power ups, etc. Another possibility is adjusting the difficulty of the terrain or the weather conditions the player moves through.

Question 5: Testing and playtesting

The IGDA Quality Assurance group includes the following in its "Ten commandments of QA":

"Understand the difference between playing a game and testing a game. Spend most of your time doing the latter."

Explain what this means, and its importance in the development process for commercial games.

Sample solution: As we discussed in lectures, one of the biggest problems in testing is ensuring you get thorough coverage - getting all areas and aspects of the game tested, not just the ones that are most common/most fun/most used. If your testers are simply playing the game as they would ordinarily play, and reporting anything that seems broken or wrong, then you have no guarantee that anyone has actually checked every available region of your game world, every character type, every weapon/item/skill type, every playing style, every reasonable approach to solving specific problems, etc etc, so you have no confidence that there isn't a substantial problem lurking there untested. This becomes even worse when you start considering combinations - have we tried every weapon with every character type on every level, have we tried playing through every puzzle, level, scenario, etc with every character type, etc. Thorough testing is achieved through carefully mapping out everything that needs to get tested and in which combinations, ensuring someone is responsible for each test set, and ensuring that every test set is executed, recorded, and analyzed.