Tuesday, August 11, 2015

Good Bad Game Design

Skyrim is designed really badly. It's also designed really well.

For an RPG, Skyrim is hilariously bad. The characters are incredibly dull, the places are bland, the voice acting is generic to a hilarious extent, the graphics are uninteresting, the fighting is uninteresting.

Compared to non-open-world RPGs such as the Mass Effect series, Skyrim falls short on every measure.

Despite that, more people still play Skyrim than still play Mass Effect 3.

See, in that dry, generic, empty world, there is space for the player. Even in vanilla Skyrim, there is an endless variety of options. You can go in any direction, stumble across any number of little challenges, see any number of sights.

Yes, all the directions are boring. All the challenges are boring. All the sights are boring.

So why is it fun?

Something about a Role Playing Game that most people seem to forget is that the player gets out what they put in. Most RPGs have very generic protagonists. The game may offer you good and evil options, but even they are quite generic. Closed-world games like Mass Effect or Dragon Age supplement this with interesting characters: Shepard and, uh... Medieval Shepard... have no personality on their own. But you define a personality by surrounding them with characters that do. You build the personality of your hero by choosing their companions. Even if you just choose the companions you happen to like best, you are defining your Shepard as someone with a very similar personality to you.

If you offer enough secondary characters and constrain the number you can choose, you can allow a player some freedom to "put in" their personality and "get" role play out of it. Shepard has a character because you mentally justify why these party members are her favorite. You also get some characterization out of the choices you're presented with in the game, but I'd argue that the party member choice is the most compelling. At the very least, it's continuous and ongoing, rather than a one-off, so it wriggles in your brain and forces you to continually imagine how Shepard feels about how things are going.

In an open-world RPG, parties are rarer. Games like Fallout and Skyrim technically have a party mechanic, but it's very vague, and the characters you can add to your party have almost no impact, personality-wise. Is this a weakness?

No, not at all. In an open-world RPG, you "put in" actions rather than choices. Sure, there may still be choices. Maybe those choices are critical for planting the seed of personality.

But the continuous actions of play are where the character grows and blooms. Trying to sneak through a house or barracks. Deciding to shoot from afar. Deciding to rely on your dull party member to defend you. Looking for a secret inside the waterfall. Opening a creaking chest in the dark. Reacting to the sudden appearance of a pack of wolves.

Unlike a closed-world game, these events are all contiguous. Sometimes they move faster or slower, but they are almost never The Event You Should Be Having. Your avatar is living every second of this adventure, and it is developing in tandem with your actions. You are free to do anything and, in doing anything, you are free to be anyone.

Compound this with mods that change the world, and now you have even more options.

I think these ideas are important.

"Role play" requires the player to feel like the avatar exists. One way to do that is with pieces you painstakingly create for that purpose - a bitter choice, an amazing sight, an interesting companion. Another way to do that is to simply provide a world for the player to live in.

Anyway, I was going to go on and talk about the same "bad vs good" design in other kinds of games (comparing Space Engineers and Kerbal), but I think that's more than enough for today. Let me know what you think.


Random_Phobosis said...

Skyrim is a bad game and an astonishingly great toy. We are used to see conflicts of game and toy properties, but TES takes this conflict to new levels by using "toy parts" from one toy and "game parts" from another game entirely! Like playing hopscotch with... toy trains? I'm certain it would be a better game if the developers wouldn't blindly copy mechanics and dynamics from fundamentally different games.

The skill and equipment progression has constant growth, so after 20/50/100 hours the player finds himself at skill and equipment caps, which would be good for linear, carefully paced adventures like Dragon Age, but conflicts with eternal wandering TES-mode. Without strong resource sinks, resources of all kinds tend to lose value very quickly. The story has similar problems: constant growth forces to set some kind of story cap, and even then you end up being grandmaster of everything, disciple and avatar of all the gods and demons, and secretly the head of all the secret societies around.

If I could make a new TES game, I'd try to construct all the dynamics from scratch around my favorite TES thing: eternally wandering and discovering cool things. Here's just one of the solutions: maybe the player should be some kind of deity, which reincarnates in mortal all the time. The player would play an adventurer for some time, but the stats would not only grow from training, but also deteriorate from injuries and age. The player could decide to retire the adventurer and incarnate in new character. This would solve a lot of problems, both mechanically and theme-wise, and even make character death meaningful (which I consider impossible in many games).

Craig Perko said...

I largely agree, but I don't think I'd take your approach. Retiring an open-world character is something that already happens, but on the player's schedule. The complexities of how the character grows determine when characters are likely to "run dry".

I think the game should be built with natural "ending points" - situations where content and possibility run out. Rather than forcing you to retire your character or making your character worse, it's simply a natural place for you to stop playing that character and start the game over as a new one. Moreover, mods and expansions could revamp this content to give you more potential.

As an example, if you are a thief, there is a gentle tug towards joining the guild and taking on thieving missions, increasing your rank within the guild. As you rank up, the missions naturally start to run dry or feel repetitive due to the limited number of house configurations, loot, and mission parameters.

Rather than simply letting them run dry, we should embed a natural end state - a clearly final mission that ends with you dying or retiring. You don't have to do the final thief guild mission - you can go on looting as long as you want, or decide to become a mage - but it's a good "final destination" for that character.

Later on, maybe we introduce a mod that allows you to run the thief guild, and that becomes viable high-level play. The old 'final mission' is either canceled or simply segues into the new play, and a new final mission is created.

Paul Liverman said...

"Anyway, I was going to go on and talk about the same "bad vs good" design in other kinds of games (comparing Space Engineers and Kerbal), but I think that's more than enough for today. Let me know what you think."

Goddammit man, that's what I want to read! You were talking about it on Twitter, and that's what brought me here. xD

Craig Perko said...

OK, this week sometime!