Welcome to the Wolfire Blog! This is where we keep everyone up to date on our progress on Overgrowth and other new stuff. Be sure to subscribe to get the latest news! Also, be sure to check out the forums for even more up to date news - or get on IRC for up to the second updates.
I am trying a new kind of blog post: a casual chat with my friend Jack Monahan. He is a professional level designer, concept artist, and writer of the blog Design Reboot. More info on Jack can be found in this previous dev chat, or in this post showing some of his awesome art.
Aubrey: Was I talking to you about the definition of games? How there are "game" games-- i.e. chess and sports and then single player video games are really something new?
Jack: No, I don't think I remember this.
Aubrey: Basically, I think one reason it's so hard to explain why games are neat to people is because there is no vocabulary for this distinction.
Jack: Right, we're still trying to figure out that one.
Aubrey: Real single player games have a lot more in common with puzzles than other games, but I think the core difference is this: Do you have to play to learn the rules, or do you have to learn the rules to play? This is why there is so much confusion when non-gamers sit down to play-- i.e. the "what am I supposed to do?" problem
Jack: Ahh yes. That's a really important distinction, and yeah it does get to our own frustration when we talk about a kind of cultural switchover that has happened--hinging on exactly this. Gamers now expect it all to be dropped into their lap. They in turn will complete some short-term task, in exchange for another story pellet.
Aubrey: To a real gamer, it is self evident what you are supposed to do in a video game. You do what you can do.
Jack: It's weird.. thinking back, I had a lot more fun on the playground with many games--basketball, football, or just playground games--rather than when I was on the school sports team. Which reflects in what I play now if i think about it. Competitive play in multiplayer is about high levels of performance that can often be exclusive in nature. Losing a round of Counter-Strike sucks a lot harder than having the least kills in a round of L4D.
Kids know good balancing. Only an asshole beats someone so thoroughly they no longer want to play, yet in really competitive sports you play all those kids all the time and it sucks if you're not one of the high performers. Half your team sits on the bench. It's about performance. Which can have its own thrilling highs--I played basketball in school, and I played Counter-Strike along with everyone else--but it's a lot less interesting now to me than it used to be because of the kind of fun it encourages.
Aubrey: So I have been thinking about these distinctions in terms of Platonic universals. What would be the perfect single player style video game and why?
Jack: Oh interesting--let's hear it.
Aubrey: Basically what I was saying earlier about the ultimate extreme of single player games as being a reality sim. That is the paradigm for a game about learning rules from doing. So if you look at old games, you have simulations and then you have arcade versions of sims, but the direction games are coming from now is from a totally invalid paradigm of games being more like movies. All these cinematic games adhere to simple game logic that everyone understands, but it's because they want the player to act in a predictable way-- not because they are trying to make a good game-game. The game expects you to play a well defined role.
Jack: Thinking about my experience as a level designer, so much was about closing off possible avenues or solutions to the player, which I hated, but you had to do it because otherwise the player wouldn't play the story "right". I hate that idea. So much of what players thought to do was really clever, but because it wasn't in line with the scripted progression of the level or the game, it had to get cut. Invisible walls get put up.
Aubrey: So the thing is that a sim as we traditionally think of it is not really right either. Like those crazy ones that require a ton of preparation and stuff are not what we think of as the holy grail of single player games.
Jack: Yeah that's not the answer either. Not for most games anyway.
Aubrey: But, mechanical simulations are a tiny subset of reality. I think we need to prune away rules until the rules form a story by learning them.
That is why game design is so hard. We only know a few handles in this new medium.
Jack: The medium is the message [Marshall McLuhan -ed]--or for a game, your mechanics encapsulate what you're trying to say.
Aubrey: We can make people pissed off or feel hopeless. We can compel people to play with stat grinding and collection. We can make people feel powerful. I think we can design a set of rules that never changes, but allow people to go from feeling weak to powerful. Which is what games like X-com pull off.
Jack: Yes. There are some pretty clear seeming constants. To me now it seems about gutting your own inner author/director that I think is important. Or at least keeping him in check because to tell stories is in our nature. Stories are great, but that storytelling impulse can strangle a game, edge out the player.
Aubrey: Yes. Games tell a story in the way reality does-- with a lot of trail and error and no editing. We don't even have a word for the nascent form of a story-- the things that happen that can be turned into a story. Maybe "proto-story", which is all video games can really tell. I think the current challenge is to make simulations of things people care about-- love, friends, enemies; when was the last time you felt hated in a game by the enemies?
Jack: Rare that a game lets you. The thing I get really obsessed with is that it's really hard to program games to give a moral reading, or really finding meaning in anything the player does. For the player it's frustrating to be misread when your intentions or what you did was completely different than what the game thinks. But cause and effect games will always do great--so why not leave more room for interpretation/moral readings of actions solely to the player? Isn't that why running over people in GTA is fun/creepy? The game can't tell an accidental veering off onto the sidewalk from an intentional murder spree, but we can.
Like in that Rev Rant, Burch tells the Dwarf Fortress story about how a whole colony is wiped out except a baby, who despairs and drowns himself in a pool of its parents blood. The game can't understand how messed up this is, and that's fine. All it has to do is let things play out--spit out results. And on the other side then, the players are screaming because obviously a baby drowning itself on human terms is hugely significant. This kind of "drama" is inherit to gaming, it can happen all the time--but only if we let the game play these things out, instead of railroading everything for the sake of a canned story.
Aubrey: yes-- I think that the DF guy has a better grasp of game design than almost anyone. He isn't a gamer.
Jack: In cinematic games It's all forced in your face. Anything "real" happens in a cutscene. Which of course isn't real at all to players.
Aubrey: Right. Red Dead Redemption is terrible that way--you know the win condition is to get to the last cutscene. Most games totally fail on basic problems, like depicting people as being anything like people.
Jack: Yep. I'm playing my war game, yet how is there any drama in a warzone with no civilians and all the soldiers are robotically proficient and perfectly unafraid to die as you are? It doesnt take much to elicit pity, or fear. Yet most games the biggest reaction you just get is annoyance.
Aubrey: Have you listened to the "fear" Irrational podcast? One thing they talk about is how you can't empower the player and expect them to be afraid.
I have been thinking a lot about timed experiences like X-Com. Imagine Half-Life as a simulation- you always start in the testing chamber at the base of the complex and need to escape Black Mesa but the Xen shit keeps getting worse at a constant rate.
HL made me feel pity for the scientists, and anger over how the bad guys treated them but only because the scientists were part of the sim
Jack: Even if you're not eliciting emotional reactions for NPCs, you can still make games interesting and tense. Mount and Blade pissed me off (in a good way) when I had a great 3-4 battle run with this nice warhorse and then some bandit crippled it. It whipped me into a murderous frenzy, and I still won the battle--even though to the game that battle was statistically similar to any of the others. Overkill, revenge, when you can execute on them can be great.
Aubrey: Yes. My wife was playing RDR and these procedural assholes accidentally killed our favorite horse. So she hunted them down and executed them, even though they were "good" guys. It was great.