How to make a game fun and interesting

IN A TALK called Fault Tolerance: From Intentionality to Improvisation, Clint Hocking from Ubisoft went over how to guide players, or not, in a game. The example was one he worked on, Far Cry 2.

Intentionality is a loaded word, and is best summed up as allowing the player to figure out the goals and how to accomplish them given a few tools and implicit rules. Doing this is a tough thing, player improvisation can often lead to odd results, sometimes even to failure. Letting people do what they want a little is good, a lot can be problematic.

An example of this was shown with a clip from one of Clint's former works, one of the Splinter Cell games. A player used a series of cameras to guide an NPC (non-player character) through a building, then convinced him to kick down a door while exclaiming that he wasn't a fool. The door was booby-trapped, and the flaming NPC was blown backwards into an open elevator shaft, and filmed as the flaming corpse hit the bottom of the shaft.

Intentionality is the whole idea that nowhere in the tutorials did this scenario come up, nor did the designers intend for this to happen. Ever. Really. But it is cool, and the player used the tools at hand to accomplish the goals, kill the NPC, that the designers did want. Instead of A -> B -> C, they got A -> R -> L -> C.

Far Cry 2 had nine pillars of gameplay during the design, and they lead to a lot of ideas for levels and intentionality. During development, there was something called HMR, Health, Morale, Repair, standing for the obvious things. They were based on infamy, basically people were afraid of your reputation. Enemy factions were affected by this and other things. The factors looked like this.

Far Cry 2 design pillars

The level in development was planned out so that your character would get a mission to blow up a rail car containing petrol that the bad guys were selling to finance their evil deeds, and using the money to buy medicine for the troops. Blowing it up meant the bad guys lacked money, so their weapons wouldn't get fixed as much, and they couldn't buy anti-malaria drugs. When you did this, the succeeding enemies would have worse weaponry and lowered health, and you would gain infamy.

It would also draw a head enemy from hiding to see what was going on, and you could assassinate him. Of course, enemy morale would then drop in line with your infamy gain, and they wouldn't fight as hard. From there, there was a big battle. The plan was to let the player scout out the battlefield, plan what weapons they would need, and assault the enemy stronghold. Simple enough.

Another gameplay pillar was fire: you could set one in the African grasslands, and use it to flush people out, occasionally kill them, and provide you with cover. It also hurt enemy morale. The design plan was to have the player set fire to the grass near the bad guys, and go in under the cover of smoke picking off enemies. Since they were demoralised and weak, many would run, those that didn't would be easy pickings because of the malaria, and would have unreliable guns. Easy pickings for someone as infamous as you.

During development, some of the features got cut, infamy being a big one. HMR was problematic to implement implicitly, and it basically became H, and the game ended up looking like a lot of other FPSes instead of a deeper game. Not a bad thing in and of itself, and the nine pillars ended up looking like this.

Meet the new 4 pillars

Not a problem, FPSes like this are still fun if done well, right? Yes, unless you have a mission design that was based around the old gameplay that used intentionality heavily. Now the mission was to blow up a rail car for no apparent reason and shoot the guy who came out to look at what happened. Both of these didn't do anything other than take up your time.

Since HMR became H, there was a little health loss for subsequent bad guys, but no morale loss, and no reliability loss for the guns. If you were like 99.999% of the gamers out there playing FPSes, you got to the final fight and went charging in. Planning is for wusses in games like this, right?

If you set a fire, the enemies, not lacking morale, didn't break and run, had great guns, could take a little scorching before they fell. They shot you to bits, and ran you back into your own fire. In level design circles, they call this a 'whoopsie', even thought the level is done right, it is done right for the wrong system, and intentionality falls flat.

This is where the improvisation part comes in. When life gives you a battalion of angry bad guys charging you, improvising is mighty handy, RUN! It was solvable, but not in a way that the game designers intended, or even designed for.

Other than "Don't cut necessary game mechanics without modifying levels", the real lesson here is that the more open a game, the more intentionality breaks down. Intentionality is a very tenuous thing in sandbox games like GTA4, you are expected to do what you want in weird ways, that is the 'game'.

That brings us to the other half of the talk, fault tolerance. In the end, things like malaria, reliability and wounds were put back in instead of more generic concepts like infamy and implicit reactions. People could be demoralised, but it wasn't a reaction to how 'bad' you were, they just basically were. Your gun jammed at semi-regular intervals, and things fell back in line toward the original ideas.

There are two concepts here to be aware of, composition and execution. To dumb it down quite a bit, composition is planning and thinking about how you are going to solve a problem or attack a situation. Things like run up to the rock, duck, shoot the guard, and sneak in below the watchtower while using weapons 4, 8 and 9.

Execution, on the other hand, is doing what you planned, or what you didn't plan. If there is an enemy fortress, an execution heavy game would basically give you a single gun, and have you charge up killing everything that popped into view.

Most games have a mix of both, but many skew things to one side or the other. Games like Thief are composition heavy, Wolfenstein 3D is execution heavy. The more execution heavy things are, the more they seem to be on rails, not necessarily a bad thing, but some find it tedious.

Games that let you plan, then do, are fine, but if you fail, many games have you die and restart. You could call this being taken out of execution and back to composition. If it is not handled gracefully, it can kill a game. Nothing is worse than having to redo 15 minutes of a level because you pressed the wrong button by accident.

A fault-tolerant game will take you out gracefully and let you retry without a crushing failure. This in and out makes a game good. Instead of dying and reloading, then trudging across the map for several minutes, you get shot and a medic pulls you away from the fight. Back at base, you get reloaded, healed and try again without much of a break.

Far Cry 2 did just that, if you got shot up too badly, your buddy grabs you and patches you back up. Into the breach once more my, blam, ow, thud. Thanks again bud, time to go back and.... rinse and repeat.

One more definition that was thrown in was initiative. In the framework of the talk, it is defined as kicking the player out of execution into composition in a way that makes them want to get back in. This entails little repetition, and makes a customer happy. Happy customers are repeat customers, and games that balance things right rarely become coasters.

In the end, the concepts in the talk are simple. Let players make their own path, and they will do things they like while having fun. Too much rope lets them hang themselves, and leads to something that is a sandbox, not really a game.

You need to plan things out for them to a limited extent, and give players the tools they need. If you do this, and allow them to use those tools gracefully, players will like your game. Balance things wrong, and a new career in something other than game design awaits. Balance carefully, grasshopper. µ