A series of interesting events.

[Paul Debrion]

A bit of a followup to my earlier thread:
http://www.shrinemaiden.org/forum/index.php/topic,7459.0.html

Say for example your walking along searching for anything and a random NPC of Tribe A comes to meet you. Sure there is a bit more to it, most likely pathfinding and the state of the current terrain and the task the NPC is trying to accomplish, but from your perspective it's still pretty random. Now say that the NPC is not actively hostile, but you see him carrying an object that you want and that Tribe A greatly values and you kill him to get it. You bring this object back to your home village to share with all your buddies. Now since you have taken this object from Tribe A, they are now actively hostile towards your people.
Tribe A generally has a couple forms of attack, they either attack you directly at ground level, or they find a high ground and launch explosives at you to wear you down before attacking directly. There are possible variations of these two of course, but these are the types of attack that you easily recognize. You an your buddies decided to build in your village in a place surrounded by high mesas because you thought it was pretty. Given this high ground, the Tribe A employs the latter form of attack due to your decision.
Now that you are under siege by Tribe A. You decide cease any attempts to attack any other tribes because quite frankly being at war with one tribe is enough. In fact, you give gifts to improve relations with Tribe C to get them to supply you with stuff to help fight of the hostility coming from Tribe A.
Eventually you attack Tribe A's village and destroy it. This stops Tribe A's attacks on your village.
Unfortunately, the bombing from Tribe A has resulted in much of the land being blown apart around your village, resulting in terrain that is deeply fractured and difficult to navigate. This results in the task of building routes in and out of the village and perhaps eventually filling in the holes to allow for easy navigation.
Unfortunately, Tribe A is the only tribe that can produce forcefields that you can only aquire through theft or trade, and it is temporarily unavailable by either means because Tribe A was destroyed and it won't be available until they rebuild. That something would have helped defend against an attack by Tribe B, and that along with the fractured land around your settlement making it difficult to defend results in your village being destroyed by Tribe B. You and your friends have to relocate and start a new settlement afterwards.

Because many systems were allowed to affect each other, events were able to lead to other events. Examples include where your robbing of the NPC caused Tribe A to become hostile AI relations leading to an attack your village that led to damaging the ground around your settlement due to destructable terrain, and the lack of an item produced by Tribe A due to the destruction of Tribe A's village and the damage to the ground leading to your village being easily defeated Tribe B.

Of course, there was human input as well. Examples include when you decided to rob the NPC, when you decided to build your village in that location with surrounding high ground, and your decision to ally with Tribe C for assistance. When these human decisions affect the series events is where there is interactivity.
Things could have turned out differently. Perhaps you could have built in a flatter location which would have resulted in a ground level attack that would have done less damage to the ground. Perhaps you could have given the object back to Tribe A and ended the conflict early, or perhaps you could have decided not to steal it in the first place which would have meant forcefieds would still be available (through trade or theft of course) to help you defend against Tribe B.

The random number generator was involved as well. It was the random number generator that led the NPC to encounter you in the first place and it was the random number generator that made Tribe B attack afterwards. The later attack was of course made fatal through a combination of events stemming from both the NPC encounter and interaction with the human players. The world generator was also partly responsible for creating the area you decided to build in, perhaps making it a bit more attractive than it would have been otherwise.

In fact this example actually comes from something I actually played through in the game Love. The object stolen being the Source of Life, villages being settlements, Tribe A being the Clondo Tribe, and so on.


This is what in my opinion is one of the most beautiful things you can see in a game, human input and random chance providing a start and the game's systems allowing that start to grow into something more.

This isn't the only game that does this. Games like AI War: Fleet Command, Dwarf Fortress, and even as old as X-Com: UFO Defense and Elite have done similar to some degree, even if it wasn't the main point of the game.

Where have you seen an memorable series events emerge from a system?

[shadowbringer]

just my opinion, but here it goes: competitive fighting games and shmups often require (among many other virtues) a large amount of quick decisions (and reassessment of the current situation and the ability to see what to do next and further), and give the feeling of being in charge of the situation (and with more experience, having the situation under relative control and trying to avoid having such control be taken from you).

About the random factor, it could be said (imho) that, let's say, when you're playing a shmup and is trying to avoid getting cornered, either you may react instintively and corner yourself a bit more than needed, or you may have foreseen that it was possible to get near the middle where, let's say, you would have more escape options (I mean, in a situation where there weren't as many potentially dangerous bullets) but this decision wasn't obvious enough because it appeared to be more riskier to react to if you had less time to react. So, in both shmups and fighting games, players try to decide between sets of best options (to increase their chances of winning and avoid having the opponent/AI reduce these) and (with more experience) enjoy employing their knowledge on the game's possibilities for doing so.

[hyorinryu]

I'm curious. How do you feel about the director system in Left 4 Dead 2(Maybe 1 to, but haven't played it)?

[Paul Debrion]

just my opinion, but here it goes: competitive fighting games and shmups often require (among many other virtues) a large amount of quick decisions (and reassessment of the current situation and the ability to see what to do next and further), and give the feeling of being in charge of the situation (and with more experience, having the situation under relative control and trying to avoid having such control be taken from you).

About the random factor, it could be said (imho) that, let's say, when you're playing a shmup and is trying to avoid getting cornered, either you may react instintively and corner yourself a bit more than needed, or you may have foreseen that it was possible to get near the middle where, let's say, you would have more escape options (I mean, in a situation where there weren't as many potentially dangerous bullets) but this decision wasn't obvious enough because it appeared to be more riskier to react to if you had less time to react. So, in both shmups and fighting games, players try to decide between sets of best options (to increase their chances of winning and avoid having the opponent/AI reduce these) and (with more experience) enjoy employing their knowledge on the game's possibilities for doing so.

It's certainly a step in the right direction.

I was discussing ways the game itself would facilitate this, but on the other hand this is about interaction between the player and the game. What you described is more about the player side of things, but that's very important as well.

I'm curious. How do you feel about the director system in Left 4 Dead 2(Maybe 1 to, but haven't played it)?

The director can move around enemies and items, but the elements that are really supposed to change how the players approach a given situation (the special infected and various items) don't seem potent enough. If every special infected had as great an effect as the tank and the witch and didn't put as many homogeneous normal infected it would help create a more interesting game.

Of course, that would make the game more interesting for me. I think others might find such a chaotic take on things to be rather annoying and frustrating, so I guess taking that into consideration it's probably fine the way it is.



On an interesting side note. There is a monotony that comes with randomizing things that way.

While this isn't too much of an issue in Left 4 Dead given the game's length, it does become a problem in other games that try to randomize things, like Diablo. While random number generators can be beneficial, many developers put them too close to the player, which results in rather repetitive results. In my opinion, a better way to use a random generator is to put more systems between it and the player and that is how you fight the monotony that often comes from using one.

I mentioned allowing systems to affect one another earlier. This is the key to fighting to repetitiveness caused by using a random generator. By doing this you allow whatever your random number generator came up with to form larger patterns of cause and effect.

Say for example you have a cup of five dice and you roll the whole cup a hundred times and you get a bunch of numbers between five and 25. Not very interesting is it?

Now say you allow each roll to affect the next one, say we establish values for which we change the number of dice in the following roll. Say if we get a number above 15 we add a die, and if we get a number below 15 we subtract a die. Now since each roll of the cup of dice influences the next one, you start to see patterns emerge. A given roll that was high will add a die which will cause the next roll to be more likely to roll higher, you start seeing cause and effect. Now the chaotic mess starts to make some sense and have some meaning to it.

Now say you run into a trader that has an excess of goods he's selling unusually cheaply. This was because trader couldn't sell his goods in down as he expected because the town is too poor. The town is too poor because they were raided by bandits a few days ago. The bandits raided the town because they needed to fill their pockets after devoting resources to rescuing a few of their own who got captured by the authorities a few days prior. The authorities happened to be within the area because an unrelated random murder that occured at a town near the bandit's hidout and they just so happened to run into each other.

Ok, so that last example might be rather extreme overkill and you could probably do away with couple of those steps to make it easier to program, but you probably see what I'm getting at. In this case there is a random number generator at the end, but since you allow it to have either a direct or indirect impact on multiple things, you start to get patterns that makes sense and chains of cause and effect begin to emerge. There are multiple systems involved, bandits that try to rescue their buddies and raid towns more often if they're running low on cash, towns that can only afford to buy things when they have money or things to trade with, traders who travel to where they can sell for the best price, and so on. What started as random and boring starts to become deeper and more interesting.