« Disagreement Is Near-Far Bias | Main | Data On Fictional Lies »

January 15, 2009

Comments

Was it Fallout3?

I can understand your point, but in general, games with skill trees that involve making decisions and so on about what to acquire, how to balance those skills, etc, often end up requiring (to do well) carefully preplanning the character. Ultimately, it's likely that a player may end up just looking up the info if it isn't already available.

IIRC, some of this is why, afaik, in Diablo 3 they're actually getting rid of the whole selectable skill tree thing, instead the upgrades (as I understand it) are automatic, and your choices involve collecting certain objects and arranging them in certain ways to create certain modifiers to the skills, or something like that.

I'd say though that in eutopia it may be nice to at least be able to see slightly ahead as to the possibilities, maybe having some notion of what actually is possible one or two steps (whatever a "step" may mean) ahead of where you are, just to give a taste and give a hint to you of what you might try to learn to do. ("learn to do" being intended to be sufficiently broad as to mean basically "solve/discover/invent the science/tech/etc that's needed")

I would definitely prefer the world where I know about my potential future as much as possible.
If I don't know something, then the chances are significantly higher that I will be screwed up about that something.
If I know -- I at least partially in control my life.
If I don't know -- somebody else is in control of my life, or even worse -- nobody is in control at all.

From evolution standpoint it's beneficial to have good idea about potential future. That's why evolution benefits individuals who want to know. That means that overall we should enjoy more the situation when we know what to expect.

I ask you to contemplate - not just which world you might prefer to live in - but how much you might want to live in the second world, rather than the first. I would even say that the second world seems more alive; when I imagine living there, my imagined will to live feels stronger. I've got to stay alive to find out what happens next, right?

I'll take the first. There's plenty of stuff you have to discover yourself anyway, without anyone hiding it, and other stuff you can only learn from experience and practice, it doing no good to merely be told. But having stuff withheld merely for the greater joy of receiving it later is ersatz fun. I want it all and I want it now, and there are enough genuine reasons that can't happen that there is no need to make an artificial scarcity.

Second Life has more in it to discover -- more Fun in it -- than any videogame. The videogame is artificial fun, substitute fun, a deliberately constructed time sink that sucks out people's life and pisses their hours on the floor, in return for seeing "You've Won!" at the end, which is just a suggestively named mental token. All you Won was a chance to scratch around for the easter eggs or do it all over again and again, the same but different.

My response to a demon offering a walkthrough of my life would be to aspire to have his abilities. Where is the walkthrough of his life? In real life, solving real problems always provides bigger real problems to solve. It is an inexhaustible source of challenge, novelty, and Fun. No matter much anyone tells you about it, you still have to stay alive to see what happens next.

Now you can expect the game designers that read OB to jump in and comment :) Skill trees and character progression lead to quite a lot of discussion in Game Design communities ...

I think one reason for giving the player full information is to prevent the player from agonizing too much about the early choices he makes - if he has more information about what those lead to, he might not agonize as much. It's a bit like "Hmm, giving people a menu with a lot of choices makes them unhappy. I know, we'll put a lot of information on each menu item so that the client can make the best decision!"

A better way of reducing the time the player spends agonizing over options is to give him a system where he can change his mind easily and at low cost. So, items rather than classes, spells rather than skills, etc.

But then, I don't know what game you're talking about, nor what were the exact reasons the designers had for making the system the way it is. They probably had pretty good ones too.

I think one reason for giving the player full information is to prevent the player from agonizing too much about the early choices he makes

That seems like an astoundingly naive "homo economicus" view of video game players. Unless the choices are overwhelmingly lopsided, more info = more agonizing.

Eliezer, so how do you account for chess being fun?

Also, I think you're failing to account for the fun of various forms of "metagaming". Among at least some players and in some games, a large amount of the enjoyment comes not from finding out about the skills, or acquiring them, or even using them--instead, it comes from planning and setting goals within the framework provided. When the enjoyment lies in the planning, I'm not convinced that the usual heuristic of "more choices = less fun" is applicable.

Note that this won't apply in cases where the resource is not limited (i.e., you can get every skill eventually) or when choices are not permanent (buying and selling equipment or items, instead of taking skills). Limited, irreplaceable resources combined with limited information is what will lead to the agonizing Emile mentioned, at least in my own experience.

As a matter of comparison, look at something like Magic: The Gathering, where you have three levels of abstraction in play:
1) The game -- drawing and playing cards, beating your opponent.
2) A solo meta-game -- planning your deck
3) A competitive meta-game -- figuring out what other players' decks look like
...and the first level is always the least important, and the third is the most important in tournament play.

Note that this combines very high surprise value in the most immediate level with no surprises and total, perfect information in the second.

Planning and optimising are definitely part of the fun that _some_ gamers get. Going into the system and finding "Power Word: Nuke" and then working out what choices to make to get there - and then seeing you getting closer to your destination - is a big pull.

So could it be said that whenever Eliezer says "video game" he really means "RPG", as opposed to strategy games which have different principles of fun?

If existential angst comes from having at least one deep problem in your life that you aren't thinking about explicitly, so that the pain which comes from it seems like a natural permanent feature - then the very first question I'd ask, to identify a possible source of that problem, would be, "Do you expect your life to improve in the near or mid-term future?"

Saved in quotes file.

I agree almost completely with Eliezer. It has been demonstrated time and time again that when the general public, the average person, is given a plethora of choices and options, they become overwhelmed and rendered useless. That is a practical reason why, I find, that those in public office are elected to such posts, so that they may act as a filter to the choices that the public will eventually recieve.

However, being that the subject is a game of colors and electrons, and not Einstein's physical reality of atoms and space, the argument for choice in the real world becomes moot. A person whole splurges on a +$50 game wants to do the best they can, if not at the beginning, then at some point during their virtual romp-about.

It has been observed that only games that have no plot whatsoever, are the ones that act "silly" and do not care about steady, regular progression. Those games that have a story focus the gamer into completing that story, and in the process get the whole story. That can be limiting if the individual does not know what is down each road in the fork, and if they could reach a higher level for a secret inside story if they had chosen a different skill path.

If every game did things the same way, the fun value of that method would decline over time. This is why we have genres, and then we have deliberate hybridization of genres.

Incidentally, "justified expectation of pleasant surprises" is exactly what I am assessing in the first few minutes of watching a movie. I am forming a judgement about the craft of the filmmakers, rather than anything particular with the plot, but whether I am in 'good hands' for the next couple hours.

The main fun and rewards structure in Diablo 2 (which I assume you're talking about - that or a clone) is the equipment, which you can't predict before you get, and which drops nearly constantly (with good equipment a bit rarer, but you're always assaulted by limited inventory and having to prioritise whether you need a piece for your own use, to sell, or whether you'll just leave it). The skill trees are there for the planner and optimiser - you can't really make bad decisions by choosing whatever sounds good, since your options at any given level are limited, but you can look ahead, sure - that prevents you from picking up "speedbump" powers that don't lead anywhere interesting.

Besides, just the text of the skill is hardly the reward for getting it. There are lots of properties of the skills you only see once you use the skill or even level it up some.

I think in general action and exploration games have a better "fun" structure though. Check out the talk here (begins about 10 minutes in) about the fun structure of learning used in most games and how to apply that to non-game applications.

When I watch a movie that I really like (and have already seen dozen times), I usually create some kind of imaginary persona that haven't seen the movie before and kind of pretend that everything surprises me. It works.

Let me switch my link to the gaming blog... Okay: no, in most contexts like that, the game designer is right and Eliezer is wrong. If the game has a limited number of irreversible training and specialization decisions, requiring players to make those decisions based on hope and faith is bad game design. Do not forget the negative surprise of discovering what you failed to gain because of uninformed decisions early on.

It would be like going to college without being allowed to see the graduation requirements. You can see what classes are available this semester and maybe next semester. You have four years to graduate, although you can delete your existing class credits and start over at any time. Good luck!

Calling this a naive homo economicus view works only if you assume that the player will play through once and never look back, or even look around for information about unchosen options. Otherwise, the player will find the information at some point, and the agonizing will take place. If you find out at level 10 that you needed to have put at least two points into the agility tree by level 6 if you ever want to fly, you start feeling those opportunity costs all at once. The player feels betrayed, goes online to look up what else he is missing because the developers decided to hide information, and might throw away the existing game to start over, grumbling back to level 10.

Remember, you can always play the game more than once. You may intend to, to see how the different options work. Your big opportunity costs come from the wasted game where you never got to explore any option fully because you locked yourself into a path you did not like.

Example of bad design: Antbuster (flash game you can find on a dozen sites). At any given point, you have up to three options for upgrading your cannon. They are part of a broad tree, and from the base you can see only the bottom branches. Can you guess which of the first three options would lead to the lightning gun, flamethrower, or insecticide? You can backtrack, but it costs money and resources are tight, so learning is losing.

See also Trial and Error Gameplay.

The designer should not give the player a massive info dump, but it should be readily available, especially since it will be online a few days after the game comes out. You are not protecting the player; you are just inconveniencing him. A common option is to have a toggle for "show options not currently available." Most MMOs have figured this out.

If I am deciding whether to spec my wizard in ice, fire, or lightning, I want to know the implications of that. Does one of them get more area effect attacks? Utility powers? Debuff effects? What is the ultimate power at the end of each spec tree? Is the end boss completely immune to one of the three?

There are many designs in which limited information works, but they require other supporting decisions. This is not a case where, all things being equal, vaguer information is better. In most cases, the best option is to present limited information with an option for complete information, plus an option to re-do those decisions somewhere along the way after you see how they work in practice.

Well, being a game developer, I can tell with some authority: you are wrong. It would (generally) mean error in game design if the abilities were hidden. Why? Because then the player couldn't plan anything.
If some or all abilities are hidden at the beginning, that forces the player to choose based on incomplete knowledge, and more often that not, leads to regrets: "I wish I purchased that ability which turned out to work in nice synergy with others, and not this one which turned out to be useless..". Especially if there's some finite pool of resources used to purchase these abilities.
And that is not fun, even if surpising.

A rule of thumb in game design is to never make players make uninformed choices, as that only leads to frustration. This beats any possible pleasant surpise that might be there.

EY: "In the other world, anyone older than you will refuse to talk about certain aspects of growing up; you'll just have to wait and find out."

Ewwwww. I want to kick everyone in that world. I'm not saying I want no surprises or that surprise does nothing for me. Just not that much. At least, compared to EY. I prefer small surprises. Surprises that fit into my goals and desires.


Vizikahn: "When I watch a movie that I really like (and have already seen dozen times), I usually create some kind of imaginary persona that haven't seen the movie before and kind of pretend that everything surprises me. It works."

When I really like a movie and enjoy watching it multiple times, it's because the primary appeal of the movie has very little to do with surprise or "finding out what happens next". In fact, even if I'm only allowed to watch a movie once, I'd rather it not depend much on surprise for its appeal. As artistic qualities go, surprise is rather low on my scale.

When I first heard the term "spoiler" I was utterly confused. Finding out the ending of a movie/story spoils it? Well, I guess it does if it's a lame story.

If some or all abilities are hidden at the beginning, that forces the player to choose based on incomplete knowledge, and more often that not, leads to regrets: "I wish I purchased that ability which turned out to work in nice synergy with others, and not this one which turned out to be useless..". Especially if there's some finite pool of resources used to purchase these abilities. And that is not fun, even if surpising.

This seems to miss the point -- you're talking about a surprise that isn't a pleasant surprise. Suppose the game was designed so that after achieving a goal, you get an unexpected bonus ability with awesome synergy with the character, no matter how the character had been developed up to that point? As a game designer, ignoring the difficulty of realizing such a design, how would you say the Fun-theoretic potential of this scenario stacks up?

A rule of thumb in game design is to never make players make uninformed choices, as that only leads to frustration. This beats any possible pleasant surpise that might be there.

This rule of thumb is overly broad as stated. It would rule out poker, "fog of war" in RTS games, etc.

I have no idea why you'd prefer not to know. Fun theory is mostly intuition, and my intuition says I really hate the feeling of loss of control that comes from not knowing important stuff, both in real life and while playing.

Another intuition is that irreversible decisions based on incomplete data feel horrible, in real life and while playing. Like deciding what temple to build in a city in Rome Total War without having any idea which one will be useful 30 turns from now when it starts to matter.

Cyan: I cannot think of any strategy game where fog of war was a good idea. The surprise works well enough in FPS where people are actively trying to use that to outsmart each other, but in strategy games it's just stupid, and reduces fun of gaming. AI in strategy games universally ignores fog of war, so there's not even fun of using that as a cover.

I don't agree with the premise that a surprise is more fun than the expected. I forget exactly where I read this but I've heard that women can have more fulfilling sexual experiences and/or stronger orgasms often when they know in advance that later in the day or week they will be having sex. So if the husband tells them "I'm coming home later and bringing the condoms" (or something, I'm not very imaginative at the moment) women can actually get more excited as the anticipation of having sex later actually makes receiving that sex all the more enjoyable. So, It may be that the anticipation of getting those skills further up the tree as you level is what makes that situation more fun than if you didn't have a clue what was coming and it just snuck up on you.

Perhaps the best world would be where one you wake up every morning and are told of the wonderful gift you'll be receiving that night before bed, -every- night before bed in fact. Maybe the AI can someday be the one to greet us before breakfast with a new amazing bedtime toy. Frankly I wouldn't mind such a world, though I would like to choose my "toy tree" and make sure I get fancy nanotech things with lots of blinking lights instead of apparel or jewelry.

Lots of people find planning their character design decisions, and exploring in detail the mechanical consequences of their designs, to be 'fun'.

Which is why there are so many sites that (for example) post in their entirety the skills for Diablo II and how each additional skillpoint affects the result - information that cannot be easily acquired from the game itself.

Although there are some basic principles behind 'fun', the specific things that make something 'fun' vary wildly from one person to another. If what the designers created wasn't to your taste, perhaps it's not that they failed, but that you're not a member of their target audience.

Slashdot had a story today about some academic research with a simple model of fun in games.

Yes, obviously you can't delete the advance information and keep the progressive irreversible skill tree which as the name "tree" implies is full of single possible paths and limited frangible resources.

The challenge would be to design a game that, oh, say, had more than one way to do something, like say real life, so that your choices were meaningful without having to hit an exact keyhole sequence in order to achieve the desired end. Which is the main point at which you start needing a "walkthrough" when you get "stuck". Then, perhaps, you could deliver pleasant surprises to the player without requiring them to plan out their whole future existence in advance.

I am not a game designer, but it still looks to me like what we have here is poor game design compensating for poor game design. There's only one way to do things and lots of irreversible choices, therefore, we have to give the player too much information about the future. This is easy on the game designer but hard on the player.

AI in strategy games universally ignores fog of war, so there's not even fun of using that as a cover.

Not quite true; Advance Wars: Days of Ruin has the AI dutifully obey the Fog of War limitations. However, the AI is pretty easy to beat anyway, so it doesn't matter.

It is good game design compensating for anything less than excellent game design.

Most people are not good game designers, even professional game designers. Even good designers make bad design decisions at times. Giving more information is a way of working around problems: it gives the player additional control and the ability to plan around problems. It is an error compensation system.

You exaggerate the point when you sprak of "single possible paths" and "one way to do something." That is prematurely halting at the obvious. We have a Diablo II citation above: the character trees have multiple paths, with many chances to pick up what you skipped before, and it still does better with more information. What the game tells you is your only way of finding out what those other ways of doing things might be. This is not real life, where you have lots of examples and feedback systems. You have what the game tells you and what you learn from trial and error in the game; if you have some meta-game knowledge, you can also guess what the designer might have done.

In an excellent game where things are designed very well, surprises can improve the experience without making you regret past decisions. They can even be adaptive to your past decisions to better avoid regrets. (At this point, you may approach calling for amputation of destiny, with the computer designing your ideal experience no matter what you choose or plan.) How many games things of that quality exist, where you trust the developer to have made all the right decisions?

If the point is that a really good game does not need to give you so much information up front, granted. It is bad design to give the player either too much or too little information. If the question resolves to that point, there is little interesting discussion to be had.

I haven't played computer games in a while, but I suspect the game designers know what they're doing better than Eliezer. When he creates a game that people want to play, I'll reconsider.

I would (or should I say "do") want to know if life is worth living, so I can cut my losses in advance.

I don't like surprises. That's part of why I like chain restaurants. That's an area where I am in sync with most people, as evidenced by their success and proliferation.

"I cannot think of any strategy game where fog of war was a good idea"

Fog of War can generate interesting gameplay in both the need for active scouting and in the need to make decisions based on limited information (like in poker).

Suppose the game was designed so that after achieving a goal, you get an unexpected bonus ability with awesome synergy with the character, no matter how the character had been developed up to that point? As a game designer, ignoring the difficulty of realizing such a design, how would you say the Fun-theoretic potential of this scenario stacks up?

Well, that is even worse, because essentially, you just took the choice away from player. No matter what he chooses, he'll always get some cool abilities.

This rule of thumb is overly broad as stated. It would rule out poker, "fog of war" in RTS games, etc.

That's a general rule of thumb, NOT an unbreakable pillar of game design. There may be, and usually are, other considerations. "Fog of war", for example, generally means unpleasant surprises for the player, but it is still a viable game mechanic in some cases.

If I may hazard a guess, it seems possible that the game designers' intent was to make interacting with the setting and the challenges fun, mysterious, and rewarding, not the power increase itself.

Let's say, for example, that this game has some string of abilities that, a few notches down the road, gives you the ability to leap much higher and farther. It's not immediate, but you can see that following this particular branch will give you that capability. You file this away for future reference; you're more interested in pumping your gun abilities for now.

Later on, you come to a location in the game where an enemy stronghold has a poorly-guarded back entrance... that's 10 feet off the ground. You realize, because you had a list of options to look through, that you have the ability to pour some skill points into these Athletics abilities and obtain the jumping ability needed to get through here. However, you've also been coming across a number of locked doors that a few points into Lockpicking would get you past... which do you choose? Each has its own potential for surprise and reward, and skill selection is just a tool for exploring that space.

I think the goal is to be up-front about skill choices and their repercussions because discovering those skills isn't supposed to be the point of the game. The point is to play the game, gain a feel for the challenges you encounter and your preferred methods of dealing with them, and then pick skills you KNOW will support those methods. The pleasant surprise is in encountering new challenges to test those skills against, not in randomly discovering that you can fly now because you picked the right 3 abilities.

Having to take action to avoid unpleasant surprises is usually pleasant, as long as your personal resources aren't stretched too much in the process.

If you eliminate the potential for unpleasant surprises, the game isn't much fun. (Imagine playing chess against an opponent that was so predictable as to never threaten to beat you. Why bother?)

Well, that is even worse, because essentially, you just took the choice away from player.

I can't help but feel that you didn't really bother to think this response through. Taken literally, you've just asserted that a surprising reward with character synergy is worse than a surprising rigid reward that makes the player feel regret. You assert that this is so because choice was taken away from the player even though neither situation involves player choice.

I get that yout design principle is to give the player choice and the ability to plan. So what is the right way to give "good news" to the player with the most hedonic impact?

Sword of the stars has a stochastic tech tree. For each technology there is a list of predecessor technologies that may lead to that technology, and each race has different chance of each of these actually enabling the following tech.

Taken literally, you've just asserted that a surprising reward with character synergy is worse than a surprising rigid reward that makes the player feel regret.

No. I asserted that guaranteed synergy is worse than random reward that might be synergistic or not, but the player does not know in advance. For example:

Let's say the player can get one of the abilities A or B at some point of the game (point 1), and C or D at some later point (point 2). And ability C is synergistic with A, that is, combination A+C is somehow superior to all others.

Eliezer's way would be to only show the player A and B at point 1, as opposed to showing also C and D. This would lead to frustration for players who had chosen B.

Your way would be to somehow devise abilities C' and D', at least one of which is synergistic with the ability that player chose at point 1, and present these abilities at point 2. This might be a good idea, if the trick is only used once or maybe several times; but soon the player will learn that whatever he chooses, he's guaranteed to receive a synergistic ability next, so there's no need to choose at all. At this point, the "hedonic impact" of this mechanic will almost disappear.

I get that yout design principle is to give the player choice and the ability to plan. So what is the right way to give "good news" to the player with the most hedonic impact?
I wish I knew. I can't name a single generally "right" way to do this. But from my experience (I certainly have no real theory to support this), the best "good news" are those directly caused by player's actions.
No. I asserted that...

Fair enough.

This might be a good idea... At this point, the "hedonic impact" of this mechanic will almost disappear.

I don't disagree with this. My scenario is premised on the reward being a surprise, so it implicitly assumes one-time use, or at least no overuse.

I'm surprised how many people justified seeing all the options ahead of time because they weren't allowed to cheaply change their mind later.

The way Diablo II's full information, no take-backs system rewards intentionally gimping your low level character is terrible. A game should be fun all the way through - you shouldn't avoid putting points in flare because fireball is coming up, and you shouldn't need to avoid lightning because more important endgame monsters are lightning immune. On another level, games shouldn't be presenting you with so many bad options masquerading as viable choices. Too many character builds in D2 will make it through normal difficulty, whereupon harder difficulties (and the remaining 2/3s of the levels) are impossible to complete. Games should not be punishing you like this.

The simple implementation of an inexpensive way to reorganize skills enables games to give the pleasure of discovery and the benefit of planning.

Brilliant.

I do however wonder Elizer, what is your current view of recreational drug use?


Very interesting. I guess there are two sides to this. All the best games I've played, going way way back, have contained constant surprise and multiple routes through. A single route and (or) a routemap that says 'this is what you'll learn/unlock here' has never done anything for me. But I think there is also a certain satisfaction in having the big book with the level 9 spells in to contemplate at your leisure, and to imagine what you might do with them if you had them, and how you might achieve them. Even with the level 9 spell book at your command, you know you're never going to have everything, and you're going to have to pick and choose what you hold in your head at any one time, trusting in your best skill to judge the path ahead. And having the book still leaves plenty of room for a few surprises that aren't in the book (the odd unique artefact, for example).

The comments to this entry are closed.

Less Wrong (sister site)

May 2009

Sun Mon Tue Wed Thu Fri Sat
          1 2
3 4 5 6 7 8 9
10 11 12 13 14 15 16
17 18 19 20 21 22 23
24 25 26 27 28 29 30
31