?

Log in

No account? Create an account
I had forgotten just how much I hate game systems that adaptively… - Silicon Rose [entries|archive|friends|userinfo]
Silicon Rose

[ userinfo | livejournal userinfo ]
[ archive | journal archive ]

[Nov. 30th, 2008|09:33 am]
Silicon Rose
[Current Mood |contemplativecontemplative]
[Current Music |"we don't have the power / but we never say never"]

I had forgotten just how much I hate game systems that adaptively level your opponents. Adaptive level systems plus non-level-based character advancement(*) just makes it one step worse. Both FFVIII and Last Remnant have the same problem: you're wandering along, having a great old time, and abruptly the enemies get a boost over a particular threshold and they're hitting you with insta-kill abilities that you don't have any counter for. With non-level-based character advancement, you don't even have a clear idea of how the game assesses your fitness for greater challenges, and your characters' advancement could be lopsided; both of these together mean that you could cross the magical HP boundary and all of a sudden you're getting slaughtered.

There are plenty of other ways the game can assess your fitness, such as number of advancements (you've raised your HP twice, your AP once, and your Strength twice - that's five advancements, so you're now "level 2"), or on the basis of the skills or number of skills you possess, on a stat threshold basis (your HP is above 400 and your Strength is above 20 -- you'd have to do this on a per-class basis with appropriate attributes). However, I think it's a hard problem to assess the actual fitness of a party for a higher level of combat on the basis of pretty much any system I can think of. I understand the desire on the part of game designers that the challenge scale regardless of how the player progresses through the game, but what you frequently end up with is the player feeling like everything was going along hunky-dory and then they abruptly had a high-velocity collision with a wall.

Writing this actually made me think of D&D 4th Edition, because it could end up having the same problems. I think that it's balanced well overall, but when the DM actually puts his hand to creating monsters, he could really mess it up... like, how much does the "shifty" ability contribute to a kobold's level? Nothing? Well, when properly used, it could seriously reduce the damage output of the players. One level? What if the terrain doesn't give them much of a chance to use it? Do you take that level away? Is it a half-level, now?

Now, in reality, 4th Ed. isn't too much of a problem because, as always, the DM has fiat power to make the game system work however he damn well pleases. If the players are getting slaughtered and he doesn't want that to happen, he can insert an out for them, or just tone down the combat to the level where the players are challenged, but not overwhelmed. The difference here is the intelligence behind the game. In Last Remnant, if they mixed up the power balance, the player is unequivocally treated to either a cake-walk or a murderous challenge. In D&D, the DM would have to be uninspired or unempowered for that to happen. It's not out of the question -- I'm certainly not good enough at adaptation to scale encounters well yet, witness CORM1-1 with the 1 damage minions where the enemies got slaughtered and I didn't do much about it -- but there's the possibility of a saving hand.

So, there is one thing that I've danced around, which is the opportunity for the game designers to scale based on perceived challenge. If the game detects that I'm having a miserable time, it can shift the combat down a level in an attempt to make it easier. It doesn't fix that "magic threshold" problem, but it can make the game playable again when something just went wrong. However, this solution also has a drawback. I can't remember a perceived challenge system where the game estimated your challenge level on the basis of the current map. They always assess your ability based on the maps that have occurred beforehand. This makes sense, because otherwise, you've ripped the challenge out of the game; every time you fail, the game makes the map easier, until the player can pass it. It becomes ensured progression. Then it becomes: what's the point of the challenge at all? I see this issue come up frequently in the gaming community, where there's this natural tension between the people who want a very casual challenge, a walk to the finish line if they're willing to just keep plodding forward, and the people who want beating the game to mean something. Look at the reaction people had to Ninja Gaiden on the Xbox. There was a loud minority who thought it was a cardinal sin for Team Ninja to make a game that blisteringly difficult.

Now I'm going to furiously backpedal to a distant point of view to make a sweeping conclusion about human nature. Most people want a challenge only insofar as they can meet it. They want challenges where they can pretend that they're a real hurdle to be overcome, that doing so is an unabashed success, but they want that hurdle to be set up for them so it's possible with just a bit of effort. Deep down inside, I know it's that way for me. It's hard to face the idea that something is just too difficult for you to accomplish. Anyways, humans need a feeling of success to keep them going, and "you're one step closer to the goal, only five hundred miles minus one step to go" doesn't cut it. We have to feel a sense of progress. And when it comes to games, you've got the additional pressure that it's supposed to be a leisure time activity. It's tolerable for me, though depressing, to know that publication is my goal, five hundred miles down the road, and I'll keep shuffling towards it in my own good time. But how many hours do I want to put into a game that's too difficult for me? There's accomplishment there, certainly, and ideas and concepts that could be put to use in real life, but video games, at least, aren't an inspiring bastion of success. Being great at Last Remnant, with its particular battle system and constraints, isn't so much of and interesting goal to me. I want a challenge that doesn't make me throw down my controller in frustration, that gives me the immediate feelings of success without too much effort.

That said, I'm a stubborn bastard, and I'm going to go beat the living snot out of those stupid birds.

(*) Non-level-based character advancement is when character advancement happens piecemeal over skills, abilities, and stats instead of coming at a level-based boundary.
LinkReply

Comments:
[User Picture]From: ketsugami
2008-11-30 07:08 pm (UTC)
A long time ago, back in my Rifts days, I was talking with Luke about GMing, and we came to the conclusion that the role of the GM is to "let the players win, but trick the players into thinking that winning is difficult." In other words, the ideal (most fun) state of an RPG is a bunch of players who are absolutely convinced that they triumphed by the skin of their teeth and that only their awesome playing skill and a bit of luck pulled them through.

So, if you kill all the players, they won't be happy. But if you cheat in their favor too obviously, or make things too easy, they ALSO won't be happy. The role of the GM is to keep things right at the fever-pitch as much as possible.

I think game designers (for video games) have a similar problem -- obviously you *want* people to be able to finish you game. (Ever since we moved away from Nintendo-style procedurally generated content, designers have realized that its stupid to include content that only .01% of the population gets to see.) At the same time, they have to make the game hard enough to give you that "Yeah, I'm awesome!" feeling when you do make progress. They have the additional handicap of not seeing their audience face to face, and having an audience that includes widely different types of gamers. Any sort of adapting challenge, difficulty levels, etc, are attempts to cope with this problem.

RPGs have the additional problem that they can fudge a bit by saying, "Well, if its too hard, go back and grind more!" =\

(Reply) (Thread)
[User Picture]From: siliconrose
2008-11-30 07:15 pm (UTC)
That's the worst part of adaptive leveling systems -- you can't grind.
(Reply) (Parent) (Thread)
From: cloakedwraith
2008-11-30 11:43 pm (UTC)
Sorry for the long reply and quote spam...here's hoping that "quote" is a valid tag...

There are plenty of other ways the game can assess your fitness, such as number of advancements (you've raised your HP twice, your AP once, and your Strength twice - that's five advancements, so you're now "level 2"), or on the basis of the skills or number of skills you possess, on a stat threshold basis (your HP is above 400 and your Strength is above 20 -- you'd have to do this on a per-class basis with appropriate attributes).

That still sounds like skill-based levelling to me. It's exactly what Bethesda did with Oblivion. While I'm not sure whether it was a fundamentally unsound idea or just a poor implementation of a generally-sound idea, since the second looks like the first in practice, I suspect the former. If Last Remnant has significant sandbox-game elements, I imagine that the adaptive levelling was introduced to encourage free-roaming. After all, statically-levelled enemies would effectively railroad the player anyway or condemn him to a grindfest.

So, there is one thing that I've danced around, which is the opportunity for the game designers to scale based on perceived challenge. If the game detects that I'm having a miserable time, it can shift the combat down a level in an attempt to make it easier. It doesn't fix that "magic threshold" problem, but it can make the game playable again when something just went wrong. However, this solution also has a drawback. I can't remember a perceived challenge system where the game estimated your challenge level on the basis of the current map. They always assess your ability based on the maps that have occurred beforehand. This makes sense, because otherwise, you've ripped the challenge out of the game; every time you fail, the game makes the map easier, until the player can pass it. It becomes ensured progression.

I'm not following that last bit. They assess the player's ability based on the maps that have occurred beforehand, as opposed to the current map, and that makes sense. So far, so good. That makes sense because you've actually done the previous maps, so it has an idea of how to scale the current map. If you consistently fail at the current map, it can reduce the difficulty, but they can still set a floor on the difficulty level, and it might be that the player just isn't ready. Also, as a means of discouraging players from suiciding themselves to get the easiest variant of a given boss battle (let's say), there can be a progressively lessened reward for victory.

As an alternative to dynamically-scaled enemies, a simpler approach might be to adjust the encounter rate. This could take the form of more encounters, with a higher percentage of them being easier, while the character is on the low end of the expected level for the map. As the party levels up, the game can start to throw in more of the rougher encounters, and when the party is about where it should be to fight the boss, the encounter rate can taper off. As a significant side benefit for players, it would discourage grinding.

Lost Odyssey took a slightly different approach to achieve the same goal with its de-facto level ceilings on each map. I thought it worked well and kept the challenge pretty solid until the Temple of Munchkins..err, Enlightenment. (I imagine that under the hood, you're still getting the same amount of XP per battle, but they just made the jump from level X to level X + 1 enormous compared to what came before.)
(Reply) (Thread)
[User Picture]From: siliconrose
2008-12-02 01:19 am (UTC)
You're right, skill and attribute leveling are much the same. However, in many RPGs they're treated differently, so I guess I just naturally did that myself.

You might be able to force adaptive leveling to work if you did a full statistical rollout for every monster and player in the game. For example, say Curse has a 50% chance of killing a given unit each turn, it lasts five turns, and there's a 5% chance of them resisting the original attack. Then, over the next five turns, the HP the Curse does is .95 * hp * (the chance that it'll kill him over the five turns, which I'm too lazy to calculate right now). Now, what's the chance of it casting Curse? Of course, chess is peanuts next to that kind of calculation.

...and it doesn't stop the fact that Curse is a cheap, cheap ability. Grrrrrr.

Okay, I agree with you on the maps. The problem is that players who are performing badly get punished in game (such as not getting the best weapons), and they hardly need any help at getting themselves killed. As a side note, you have one of the SRWa games where you only got certain mechs and pilots if you were performing BADLY enough. Now that sucked.

Yeah, I liked how Lost Odyssey's level system tapered off in each dungeon. That's one of the good things about a dynamic experience system. I seem to remember that the way they represented it to players was that you needed 500 experience each level, and you dropped down to 1xp per battle if you were at the plateau point.

I'm not sure about the encounter rate dropping. Something is niggling in the back of my head, but I can't figure it out.
(Reply) (Parent) (Thread)