Monday, February 10, 2014

Decisions, Decisions: Information Disclosure Changes How We Play

(Warning: Mass Effect 1 Spoilers at the end of the post. Also, that game was released nearly 7 years ago. Statute of Spoiler Limitations is practically expired at this point.)

A game is a series of interesting choices.
- Sid Meier

Now, the cadence and magnitude of those choices can be quite variable, from the macro-choices of what class or civilization do I play, to the micro-choices of what to build in this city next, or what ability should I use in the next 1.5 seconds. But what differentiates a game from a movie is the fact that you’re making choices.

And nothing in a game is more frustrating than making choices in a vacuum, or being overloaded with information. Living beings collect input, think about it, and then make a decision. So when a game presents us with nothing or too much information, those choices feel arbitrary. We’re divorced from the consequences because as far as we’re concerned, we didn’t have the right information at hand to make a rational, reasonable, or emotional choice.

Take the game Final Fantasy XIII-2. As you gain experience points, you get the option of determining what role to level. The game lets you know at what levels of each role you get skills at, but the game doesn’t tell you is that depending on which role you level when, your character gets different stat allocations. For example, when activating a large node as opposed to a small node, you get bonus strength if you picked the Commando role, or bonus magic if you picked the Ravager role.

Leveling the Ravager role on a large node. Note the +2 Magic only shows up after you've spent the points to level it. Vexing.
The number of large nodes is finite, so at the end of the game, your character’s statistics can vary wildly depending on if you focused on one stat or another, or just leveled haphazardly. Because the decision is permanent, and the game doesn’t tell you in advance, this can lead to a frustrating experience if you haven’t figured it all out. Granted, some experimentation and then a reload would also solve this, but why should I have to do that?

Another example of this where you cannot be saved by reloading is Path of Exile. Their passive abilities grid, which is a lot like the Final Fantasy X Sphere Grid, is largely permanent. You get occasional points you can use to respec a single passive node, and I believe you get one full reset of the grid, but since you don’t know what abilities your character has available (as they’re gems which randomly drop), you don’t necessarily know what you want to spec into, and with the always online auto-saving nature of the game, there's no going back.

Because you cannot tell the future, and because decisions are permanent, it discourages experimentation. So you hedge your bets and take what you think is the safest path with what little information you have to work with, but you cannot know what path you've taken is actually the safest. Worse still, you potentially trap yourself into a substandard setup with no recourse to correct it besides throwing away potentially tens to hundreds of hours of effort and starting from scratch.

Path of Exile's Passive Skill grid. Decisions here are (mostly) permanent. And there are a LOT of decisions.
Contrast that to Diablo III or World of Warcraft, where Blizzard has the stance that information disclosure is key. And on top of that, any such decisions are not, generally, permanent. Talent points can be respecced, abilities can be swapped out, and any decision you make come with a plethora of information. As you level, you know what abilities you’ll get and at what levels, and those abilities are laid out in excruciating detail if you’re interested.

But there’s the trick, WoW these days has the concept of basic versus advanced tooltips. Advanced tooltips have enough information for the theorycrafters and the detail-oriented folks. For example, I came up with the near optimal rotation for my Frost Mage just by reading tooltips. Same thing with my Enhancement Shaman. Granted, oddities in the optimal rotation, such as hard-casting Lightning Bolt before you hit 5 stacks of Maelstrom Weapon is completely non-obvious, and needed simulation to suss out, but the rest of it is relatively straightforward if you do some basic math in your head. For those folks who don’t care about the minutiae or find that much information overwhelming, you can get close enough for most day-to-day things like questing, dailies, and the timeless isle by using the basic tooltips.

Compare that to a game like Final Fantasy XII, where you know that you have the spells Fire, Fira, and Firaga, and the descriptions are basically a weak, medium, or powerful fire attack respectively. You don’t need to know more detail than that because you can just choose the right one for the job with a little experimentation. Granted, the combat system in FFXII wasn’t as complex as figuring an optimal rotation in WoW, mind you.

But what about that thrill of discovery? Figuring out a lot of this stuff is part of the fun! Bravely Default scratches that itch in an interesting fashion. They don’t tell me ahead of time what abilities I’ll get as I level my jobs, but I know I’ll get something at every level, and of course it’s all themed to each job. The only decision I’m making here is which jobs to level first. The decisions aren't mutually exclusive, so given enough (a lot of) time and effort, I can level all of the jobs, but the real decisions come in how I mix and match abilities.

In Bravely Default, you don't know what you'll get at each level, but you know you'll get something.
As a game designer, knowing how much information to disclose and what to leave to the player to discover is a difficult question, and it also depends on your audience. Hardcore old school players may love everything to be obscure and require large amounts of digging to understand, whereas theorycrafters may want all the information in detailed numeric form. Casual or beginner players may want just enough guidance to be able to pick up the basics. Of course, there are infinite gradients of in-between.

As a player, I can say what I prefer: if the decision is something that is mutually exclusive and permanent with respect to my character advancement, I’d like to know what the consequences and effects of that decision are going to be. If it’s with respect to plot, story, or character interactions, there should be enough knowledge in-game such that I can reasonably guess where that decision might take me.

Feel free to throw plot twists at me, but if I decide that I’m going to let Space Racist Ashley Williams die to save Kaiden Alenko (not biased, at all), then after the fact, I’d expect Williams would be dead and I’d have the opportunity to save Alenko. I wouldn’t expect that Williams would suddenly turn into an Angel and sit on my shoulder whispering in my ear, telling me to burn them all. Doesn’t fit the narrative or the world that had been built up to that point.

I want my decisions to have an effect on my gameplay, and to be interesting and have that meaningful effect, I need to understand what it is I am deciding. Otherwise it will feel like I'm just playing a slot machine with pretty graphics.


  1. Very interesting thoughts.

    I have not played any of the Mass Effect titles yet btw, so I think making spoilers avoidable is still a good idea.

  2. You also have games like Spec Ops: The Line that give you several trivial choices early on. They make it seem like your choices will matter, but, at the most crucial point in the narrative, they take any choice away. It made the whole game after that point someone else's story. Which is fine, but don't make me think I'm going to be making narrative decisions when I'm not.

    1. That bugs me too. I understand that branching narratives and story are difficult from a developer perspective. You end up with pieces of story that maybe only a quarter of your players will see, so financially it may not make sense. But at the same time, don't insult your players' intelligence.

      As you say, a linear story isn't bad. It can be quite good if well crafted, but putting in false choices ensures that your players will be frustrated when they realize that their choices didn't matter.

  3. "And on top of that, any such decisions are not, generally, permanent."

    This is a big one. You aren't penalized for guessing wrong in a situation where you simply don't know enough. No matter how badly you mess up a WoW character, you can still fix it with some effort.

    And as much as I love Neverwinter Nights, this is one of my gripes with it -- normally you can't fix anything about a character. With console commands you can delevel and relevel yourself, but even then you can't fix the choices you had to make at level 1.

    Hence why the online world I'm building in the game will require characters to start with certain stats and abilities because those are the one thing I can't fix for players.

    P.S. If you ever want to play ME3 multiplayer let me know. Only good thing from that game (slight exaggeration but not that much).

    1. Effectively pre-made templates for level 1 characters so people can't hose themselves? That's a pretty good workaround.

      Sadly, I don't play multiplayer ME3. I did a bit to get the ending, but I'm not really big on shooters in general. ME3 I played for the story, not the combat.

    2. "Effectively pre-made templates for level 1 characters so people can't hose themselves? That's a pretty good workaround."

      Basically, yeah. I know people will whine about how I'm limiting their choice but, unfortunately, it's the only way to prevent people from shooting themselves in the foot (even if it's better at the time I might change something down the line that then screws them over which I don't want).

      "ME3 I played for the story, not the combat."

      You poor thing. The reverse is the only "redeeming" feature of the game.

      ME1 = Great overarching story, decent side stories, terrible combat
      ME2 = Good overarching story, good side stories, good combat
      ME3 = Terrible overarching story, decent side stories, good combat (better than ME2 in some ways, worse in others)

    3. Perhaps I should've said I played ME3 to bring closure to the rest of the series. But I did actually enjoy the story for the most part in ME3. The first 99% of it, anyhow. *shrug*

    4. To be fair, I enjoyed myself in general (even though there were obvious plot/story issues) the first time through the game -- some of the side missions particularly were really good.

      But trying to play through a second time? I made it like halfway through Palaven before saying "Yeah...this is just stupid, even discarding the ending."