Consequences of violence -- have your say!

[Wilbefast]

Maybe this would add more to a choose-your-own-path game than "Do you kill this guy or that guy?". I think that branching possibilities would be much more believable if they were determined by how you play. Lugaru challenge levels already had a way of measuring how you play.

I agree with everything you're saying Noz, and this nicely brings us back to the original question: how do you make a game with some degree of branching, without have to design an exponentially large set of levels? I'm thinking the story would need to be written with interactivity firmly in mind.

Does anyone remember that little experiment where the start is always the same, with a woman telling you she misses you when you're gone, but depending on the path you take she ends up either being your wife or your daughter or your mistress?

[Djemps]

I agree that a branching story line is not the best idea for Overgrowth. Aside from all of the extra time and detail required to build the extra content, there is still the sad possibility that adding a choice tree might not actually produce any new 'choices'.

Here are two examples to think about...

The Mass Effect 3 debacle is a great example of how 'your choices matter' hardly matters at all. But it certainly isn't the first time that a game's promise of multiple endings has only resulted in tiny adjustments to the pre-scripted theatrical finale. How many old school SNES roleplaying games have multiple endings are just the same scene with an extra character standing somewhere in the corner? Something like that is not worth the time and effort for a replay.

Another problem might be called the 'Choose Your Own Adventure' syndrome. I remember going through those books as a kid and always ending up at the same two or three endings. Yet, when I flipped through the pages, there was at least an additional 30% of book content that I could never seem to 'choose' correctly. I would try over and over again to adjust my decision tree, but I hardly ever figured out how to get to those other pages. If built poorly, a video game's decision tree can end up having the same problem: there is all this possible content, but players hit certain bottlenecks and only pass through the same few results over and over again. At least as a kid I could see that there were other possibilities available to me, and thus I tried new things to get there. But a gamer has no way of knowing what else is possible in his 'Choose Your Own Adventure' game experience.

I have some ideas on how to create in-game results for varying actions, but sadly I have to jump off the computer for now. I will try to type up the second half of this tonight.

[Wilbefast]

I don't know why more game designers don't pull the fatalism card. Ie. no matter what choices you makes, your destiny is to play levels 1-10 then fight the boss

[+] Iji plot spoiler

In Iji you might want to save the Tansen but no matter what choices you make and how few you kill, they're all doomed.

Here's an idea: maybe you're destined to be in place P at time T but what you'll do there depends on past choices. For instance if P is a raider camp, you might be defending it from attackers or attacking it yourself, depending on past choices. Of course levels would need to be polyvalent, hence bland, to allow them to be played from multiple perspectives Scratch that one then...

[Noz]

The Mass Effect 3 debacle is a great example of how 'your choices matter' hardly matters at all. But it certainly isn't the first time that a game's promise of multiple endings has only resulted in tiny adjustments to the pre-scripted theatrical finale. How many old school SNES roleplaying games have multiple endings are just the same scene with an extra character standing somewhere in the corner? Something like that is not worth the time and effort for a replay.

Another problem might be called the 'Choose Your Own Adventure' syndrome. I remember going through those books as a kid and always ending up at the same two or three endings. Yet, when I flipped through the pages, there was at least an additional 30% of book content that I could never seem to 'choose' correctly. I would try over and over again to adjust my decision tree, but I hardly ever figured out how to get to those other pages. If built poorly, a video game's decision tree can end up having the same problem: there is all this possible content, but players hit certain bottlenecks and only pass through the same few results over and over again. At least as a kid I could see that there were other possibilities available to me, and thus I tried new things to get there. But a gamer has no way of knowing what else is possible in his 'Choose Your Own Adventure' game experience.

Therefore, I think it would be good to have the story mold to how the player plays the game. This way, the player will have the most satisfactory ending. I can think of one example of how this could work: As per my previous suggestion, if you are allied with a certain race/faction based on your gameplay style, your ending could be roughly the same, but tailored to how you played the rest of the game. The game could give you levels that would provide challenge, but not frustration. If you play the first few levels using stealth tactics, the game would take you down a path with a higher percentage of stealth based challenges. If you like big ol' brawls, the game will give you a suitable path and ending.

And, to make sure the player doesn't feel cheated out of game content, the ending could be roughly the same, but with a different approach as to how to accomplish it. Say you play stealthily, ally yourself with the rats, and in the end have to storm a city and assassinate a cat leader. The game might have you attack from one side and give you a set of Splinter-Cell-like challenges along the way. If you play like a maniac and ally yourself with the wolves, you attack from another side, bash the gate down, and paint the whole city like Jackson Pollock.

I realize that this might take away some of the freedom the player should have to make these choices himself, so to give the feeling of choice to the player, the game could simply suggest the path that you take and only seldom make you travel down it due to previous choices. Once the ball gets rolling on the choices you made in the past, the game could lock you in on one path until you have an absolute ending culminating from the choices that you made previously. Think Hammerfall meets Spore meets Mass Effect.

[Wilbefast]

The game could give you levels that would provide challenge, but not frustration. If you play the first few levels using stealth tactics, the game would take you down a path with a higher percentage of stealth based challenges. If you like big ol' brawls, the game will give you a suitable path and ending.

The trouble with this is it would require O(b^d) levels to be created, where 'b' is the number of choices per decision node ("branching factor") and 'd' is the number of choices ("depth"). In other words 3 decisions each with 2 options means something to the order of 8 levels, of which only d = 3 will be seen in a single play-through. 5 decisions with 4 options each would require 1024 levels. Exponential growth is scary.

the ending could be roughly the same, but with a different approach as to how to accomplish it. Say you play stealthily, ally yourself with the rats, and in the end have to storm a city and assassinate a cat leader. The game might have you attack from one side and give you a set of Splinter-Cell-like challenges along the way. If you play like a maniac and ally yourself with the wolves, you attack from another side, bash the gate down, and paint the whole city like Jackson Pollock.

The only problem with this is ...

levels would need to be polyvalent, hence bland, to allow them to be played from multiple perspectives

[Noz]

I was thinking that the game could measure your skills, weaknesses, and playing styles in the first few levels and put you on one track based on its calculations. This way, there wouldn't be a more than a few path changes during the game and the level quantity would remain fairly low.

[Wilbefast]

I was thinking that the game could measure your skills, weaknesses, and playing styles in the first few levels and put you on one track based on its calculations. This way, there wouldn't be a more than a few path changes during the game and the level quantity would remain fairly low.

So it would be more about catering to play-style then? Interesting. Bear in mind though that the sooner you do your branch, the more alternative content needs to be made: if the decision is at the very beginning you potentially need to make two versions of the entire narrative. Is it any wonder therefore that most games promise alternate endings, not alternate beginnings?

I'm going to dump a few previous posts here so I can refer to them:

• "The definitive notoriety thread"
• "An experiment in excessive freedom"
• "Story / challenge mode"

Also some stuff by "Extra Credits":

• "Narrative mechanics"
• "Choice and conflict"
• "Cutscenes"

I'll do the actually referring... um... sometime later

[Son of Sven]

I'm not quite sure how easy this would be to implement, but I think a good way to improve the challenge of the game would be to change enemies based on how you fight

For instance:
In the first few levels you go in all guns blazing and slaughter all your enemies, so the game adjustes the enemies to have better armour and weapons, making them harder to beat in direct combat, thus requiring you to take a different approach.

Or say you go stealthy for the first couple of levels, then the enemies would become more alert, but (for balance) less armoured.


I realise this would be an absolute bitch to implement, but it would mean that people would be better at the game overall. Once the enemies become tougher, people will have to learn other ways to kill them (or even not kill them).
This would make the game more of an overall challenge, e.g You've gone stealthy and you've gone all out, so the enemies are tougher and more alert. This mean the player will actually have to adapt and get better at fighting and stealth, and end up with most players being excellent all-rounders by the end of the game

Of course, it would have to be slow and subtle so that players have time to learn new tactics and adapt, but it would offer more challenging gameplay, which most gamers will enjoy

[Wilbefast]

[Endoperez]

I'm not quite sure how easy this would be to implement, but I think a good way to improve the challenge of the game would be to change enemies based on how you fight

For instance:
In the first few levels you go in all guns blazing and slaughter all your enemies, so the game adjustes the enemies to have better armour and weapons, making them harder to beat in direct combat, thus requiring you to take a different approach.

That comes close to punishing a player because he wants to play a certain way. Why couldn't the enemies get heavier armor and/or become more alerted any way, regardless of previous levels?

[Son of Sven]

I was saying that this would be a gradual thing, not to punish players, but to make them better at the game. You can still rush the enemy if they're armored, it would just be harder to do. And yes, enemies probably would get harder regardless of this, but with this idea enemies would be better and worse at specific things, which, if you change how you fight, could give you an edge over them later on

[Endoperez]

Lugaru punished player for constantly using a single move continuously during a single mission / level. That's understandable and worked. One reason it worked is that the player learns quickly what's happening. However, this happens gradually, and the player who didn't read the memo won't know something is wrong. It also doesn't force the player to change his style too much. You can't spam a single attack, but you can still beat the enemies if you vary your attacks.

The suggestions is meant to get the player to change his playstyle. I don't think that's a good choice. Why should there be LESS fighting and MORE sneaking for the people who want the opposite?

I would understand it if there were some levels where going rambo won't work, where enemies are grouped together and armored, and sneaking is the way to go. Some people probably wouldn't like it, but they can cope with it.

[Son of Sven]

I can see your point, but people would realise that the enemies are changing, even if it is gradual. And I never said that people had to completely change the way they play, it would just be harder, and might make some people consider trying different methods.

I suggested this because I thought it would make people more well-rounded in terms of skill when playing OG.

[Wilbefast]

I think the story should reflect your choices but not so much the levels themselves: all the levels should be playable, to a degree, in different ways. Would you rather a large number of levels organised into different paths, or a smaller number of better-made levels? This ain't cloud-cuckoo-land, there's only a couple of people working on this game and they can't do miracles.

So far the most practicable solution we've come by, in my not-particularly-humble opinion, is to have modules within levels which can be swapped in and out depending on the player's save-file (ie. previous choices, difficulty level, whether it's Christmas, etc). Diablo 3 apparently swaps out chunks of its otherwise linear (boo) levels at random. In the Doom level editor you tag enemies based on the difficulty levels for which they'll appear.



This takes the Iji et al. alternate-dialogue technique a little further: with the write framework you could have alternate enemies or weapons as well as dialogue, without building whole different sets of levels.

[mnelskerovergrowth]

Heck yeah! Morale!