The Galactic Line is a game about crews aboard space ships. It has one particularly glaring challenge: the camera.
Looking at the crew of a space ship is typically done via a birds-eye view with the roof stripped away, just like how you see people in The Sims, or FTL. This approach is particularly useful in games where you also build things or manage complex facilities, because this kind of "map" view is a natural fit with arranging pieces and laying floorplans.
Originally, I had planned to have The Galactic Line feature several different cameras, including a first-person or close-third-person for realtime events. However, I think I've changed my mind. Let's talk about the drawbacks of the birds-eye view, my workarounds, and why I'm leaning in that direction again.
Floors
One of the biggest limitations of the birds-eye camera is when things are stacked. You can see this even in The Sims: when you're looking at the ground floor, you can't see people in the basement or upstairs. This keeps you from having a good overview, despite having a literal over view.
The Galactic Line uses modular construction, rather than tiled construction. You click prefab rooms and sections together. This means we can arrange our modules so that they naturally click together in a manner that doesn't have overlapping rooms.
1) Multi-layer modules (stairs, atriums, etc) give us a good control point for how ships can be arranged vertically. If these modules have doors on each floor offset from the floor beneath, then the attached modules will also be offset, allowing us to naturally stagger our layouts.
2) Use of "gray" space such as storage and life support means that a habitation module can have lots of areas we can simply leave unrendered, allowing for rooms above or below to naturally fill that space. Moreover, hallways themselves are low-priority and could be rendered softly, if at all, allowing us to stack halls over rooms without blocking the rooms.
3) Larger modules for larger ships may have a dozen rooms in one package. If we make each of those modules largely vertical, we can arrange the rooms so they don't block each other vertically. Additional ship modules will generally snap on horizontally, meaning our large modules rarely vertically overlap with other large modules.
4) Sparse modules. Using a variety of layout tricks such as arbitrary angles, slopes, and gentle curves, we can make it difficult to tightly pack our modules. These work well if the modules have a number of windows or other external-facing elements to give an excuse as to why you don't want to pack them tightly. This would be, visually, fairly unique: most settings have tightly-packed hab areas.
5) Subsystem filters rather than floor filters. Instead of looking at "deck 6", we can look at "off-duty rooms" or "engineering rooms" or "command rooms". By arranging these rooms using the aforementioned techniques, we can show a spiderweb of rooms on several floors, none overlapping. The focus on functionality means we can also see the operations of the sort we care about, rather than the arbitrary "nth floor" filter.
People
In The Sims, you can't see everyone all the time - they might be on a different floor, or just far enough to the side that they're not on screen. So there's a bunch of portraits on the screen which show their faces and needs, letting you keep track of them no matter what you're looking at.
Our situation is a bit more complex. The Sims is about a few people, and the focus is on tending moods. In The Galactic Line, you could theoretically have a crew of thousands, and the focus is on the specific event that's unfolding rather than on moment-to-moment moods.
The good news is that the event focus means we are also able to focus on only a few of our potentially thousands of crew members: the ones involved in the event.
This means we can have a manageable number of portraits, but it also means we can have a manageable map presence. We can arrange our highlighted crewmembers to be in rooms with no overlap, so we'll always have a clear view of them. We can even build an adaptive interior view which highlights only the rooms they're in, making any potentially overlapping rooms invisible unless you manually walk into them for some reason.
Functionally, this means we don't need portraits. We can use portraits for a variety of things, especially when arranging or building the crew, but for the event scenes we don't need them. You can see everyone involved. You can see what they're involved with because they're standing at a specific place in a specific room. There's no need for floating heads to pump numbers into your eyes.
Faces
The reason I wanted to go with a more personal camera is easy: this is a game about people, so I wanted closeups of people. I planned to do some camera manipulation when you get near/start talking. Get a reasonable closeup of the face.
I need this because A) faces are distinct, and it's one of the important ways crew members are unique. B) The expressions they make are a critical part of empathizing with them and their situation.
However, we don't need to rely on the player camera. There's no reason we can't have a separate dialog system, like thousands of games of all sorts. A popup of some kind that shows their face up close.
There's loads of different approaches. For example, the full-screen dialog tree vs the chatroom-style messages with faces alongside. There are a fair number of distinct options - perhaps a sidebar element that folds out to show the ship's chat room would feel right... although the faces might be too small.
Any way I slice it, there's no big problem with having a separate dialog engine. It's common.
Advantages to the Bird
Birds-eye camera has a few distinct advantages. I've already mentioned that it fits well with the construction engine. It also shows a fair number of people and the state of the ship all at a glance, which is nice.
Another advantage is the low resolution required. Because the camera never gets too close, I can use relatively low-res assets for furniture, for bodies and rooms. I don't need to have high-res assets for books and posters.
The main problem I have with the camera angle is simply that it's not very personal. Moving a pawn around on a giant board game is not very immersive. Walking around inside the ship is a hugely immersive experience that I would love to focus on, especially because the experience of being stuck aboard the ship is the focus of the game.
Well, let me know what you think.
Wednesday, March 15, 2017
Monday, March 06, 2017
Glowing Open Worlds
I was following some Twitter thoughts by Jake Monahan, and it got me thinking. He was talking about how open world games are inherently built to be strip-mined, discarded once you've finished pillaging them. He points out that most of the things you do that change the world (such as resolving a gang war) actually make the world less fun to play in.
This is very interesting to me, so I thought I'd analyze it in way too much detail!
First, the constraint: we're specifically talking about designed open worlds, not randomized ones. Fallout and GTA, not Minecraft.
The most obvious reason why open world games are built like this is because that's the play we have gotten used to. Killing and collecting are our focus, and both of those remove your targets from the game.
A less obvious reason is that most games are about progressing through the game. If the world is steady-state and nothing you do changes anything, there's no reason to move to the next zone. And if you do decide to move on, there's no reason to actually grapple with the next zone, you might as well just skate through as fast as possible.
So the "strip mining" is not simply a side effect of the preferred play. It's a tactic that pushes the player to move through the game at a steady pace. Different players will engage at different depths - some might spend a lot of time hunting for every scrap in a given area, others might just hit the high notes. Either way, they are steadily lured into finishing this zone and moving into the next one.
It's possible to design a constructive open world, where your gameplay makes more gameplay. However, you would need a different mechanic to lure the player forwards, or they'll never get very far into your world. They'll just mine out all the play options in the first few areas and then get bored of your mechanics.
In short, if the player isn't lured into moving through your open world, it's not an open world.
Whether it's a linear game or an open world game, as we move through the game we discard levels behind us. In a linear game, those levels are simply gone. In an open world game we can physically return to those places, shrines dedicated to our past adventures, but they are no longer part of our path forward.
...
It's worth considering replay value. In the first play-through, players usually behave fairly predictably. This is definitely worthy of its own conversation, because 'predictably' doesn't mean 'exactly the same path through the game', but it does mean that you can typically use standard tricks and layouts to pull them into a few paths you consider to be ideal for new players.
For example, in Fallout 4 you aren't physically forced to get Preston and Sanctuary, but nearly all first-time players will. After that it appears to open up and tell you to go do anything you want, but the way the map and sidequests are laid out pushes you to find a few small farms to recruit and a few bandit hives to clear. In theory you can spend a lot of time farting around this open world, but in practice they limit your available play options so that you'll get a little bored of doing that and decide to start for the city. They city features a combination of physical funnels, golden paths, and meat walls to pressure you into finding Diamond City.
The design of this is "first run" open world design, and it's not too different from a linear game. But when we replay the game, we are more familiar with the world and take more advantage of the variations.
This isn't a proper "new game+" situation, because many people restart the game long before they've beaten it. For example, I've played the first five hours of Skyrim approximately 300 times, and I still haven't actually beaten the game. Trying new paths, or perfecting your abuse of paths your familiar with, or challenging yourself with a new constraint... these are powerful play elements!
What I'm leading up to is this:
The open world we strip mined? It doesn't stay stripped.
We return to the fresh, untrammeled world again and again, whenever we want.
To me, that is the truest power of an open world, and that is what I would like to focus on when building open world maps.
... maybe I'll talk about that for fifty hours some day soon.
This is very interesting to me, so I thought I'd analyze it in way too much detail!
First, the constraint: we're specifically talking about designed open worlds, not randomized ones. Fallout and GTA, not Minecraft.
The most obvious reason why open world games are built like this is because that's the play we have gotten used to. Killing and collecting are our focus, and both of those remove your targets from the game.
A less obvious reason is that most games are about progressing through the game. If the world is steady-state and nothing you do changes anything, there's no reason to move to the next zone. And if you do decide to move on, there's no reason to actually grapple with the next zone, you might as well just skate through as fast as possible.
So the "strip mining" is not simply a side effect of the preferred play. It's a tactic that pushes the player to move through the game at a steady pace. Different players will engage at different depths - some might spend a lot of time hunting for every scrap in a given area, others might just hit the high notes. Either way, they are steadily lured into finishing this zone and moving into the next one.
It's possible to design a constructive open world, where your gameplay makes more gameplay. However, you would need a different mechanic to lure the player forwards, or they'll never get very far into your world. They'll just mine out all the play options in the first few areas and then get bored of your mechanics.
In short, if the player isn't lured into moving through your open world, it's not an open world.
Whether it's a linear game or an open world game, as we move through the game we discard levels behind us. In a linear game, those levels are simply gone. In an open world game we can physically return to those places, shrines dedicated to our past adventures, but they are no longer part of our path forward.
...
It's worth considering replay value. In the first play-through, players usually behave fairly predictably. This is definitely worthy of its own conversation, because 'predictably' doesn't mean 'exactly the same path through the game', but it does mean that you can typically use standard tricks and layouts to pull them into a few paths you consider to be ideal for new players.
For example, in Fallout 4 you aren't physically forced to get Preston and Sanctuary, but nearly all first-time players will. After that it appears to open up and tell you to go do anything you want, but the way the map and sidequests are laid out pushes you to find a few small farms to recruit and a few bandit hives to clear. In theory you can spend a lot of time farting around this open world, but in practice they limit your available play options so that you'll get a little bored of doing that and decide to start for the city. They city features a combination of physical funnels, golden paths, and meat walls to pressure you into finding Diamond City.
The design of this is "first run" open world design, and it's not too different from a linear game. But when we replay the game, we are more familiar with the world and take more advantage of the variations.
This isn't a proper "new game+" situation, because many people restart the game long before they've beaten it. For example, I've played the first five hours of Skyrim approximately 300 times, and I still haven't actually beaten the game. Trying new paths, or perfecting your abuse of paths your familiar with, or challenging yourself with a new constraint... these are powerful play elements!
What I'm leading up to is this:
The open world we strip mined? It doesn't stay stripped.
We return to the fresh, untrammeled world again and again, whenever we want.
To me, that is the truest power of an open world, and that is what I would like to focus on when building open world maps.
... maybe I'll talk about that for fifty hours some day soon.
Friday, February 24, 2017
Foreshadowing and Layered Storytelling
Spoilers for Life is Strange and Night in the Woods. Night in the Woods spoilers are hidden behind a button (in the web version) and you can skip them.
Recently I've seen a lot of games stumble over the concepts of foreshadowing and layered storytelling. Even really good games. So let's talk a little bit about it.
This might be a bit rough, it's my first take.
Games have twists. At the end, in the middle, twists everywhere. All these twists have to feel like they make sense.... but they usually don't. They often feel like they "came out of nowhere".
There's usually foreshadowing: nobody simply forgets to set up the ending. They just... foreshadow badly.
The two most common kinds of bad foreshadowing are the accidental reveal and NG+shadowing.
The accidental reveal is pretty obvious: the foreshadowing is so blatant that it simply gives away the twist.
NG+shadowing is the opposite. The foreshadowing is so minuscule or unrelated that you can only see it as foreshadowing when you're playing the game a second time.
You might think we're on the hunt for goldilocks moments: foreshadowing that isn't too blatant, but isn't too subtle.
Not me. I say both those kinds of foreshadowing are misunderstanding the role of foreshadowing.
Let's get... hoity-toity.
The twists that stay with me are the ones that are thematically relevant.
Let's say we're writing a game about a main character struggling through a rough time in their life. The backdrop to this is a series of mysterious murders: our MC lives their life, notices murders happening in the background. At the end of the game, the two plot lines come together and there's... some kind of cool twist.
If you're a writer, you need to think about how these two plot lines support each other. They're not just two stories happening in the same place: both need to be stronger because the other exists.
The first instinct is to make them related physically. Our struggling MC witnesses a murder. Or is suspected of being a murderer. Or one of the friends is a suspect. Simply moosh the two plot lines together and then stitch things up. It works. Ish.
A good writer will think about more than that. We need the two to support each other thematically. The serial killing is a twisted reflection of our MC's internal struggle. The twist at the end is how the two separate or merge, and it reflects not the meanderings of the plot, but the slow syncing of the two themes.
...
Let's discuss and compare Life is Strange and Night in the Woods. They both have the same main character and a similarly laid-back approach to gameplay, so they're a good contrast. Since NitW is new, I put the NitW spoilers behind a spoiler block. You can skip them.
...
In Life is Strange, our main character has psychic powers. These help her to mask her social anxieties by allowing her to simply undo social interactions. Initially, her personal arc is about coping with or overcoming her social anxiety - perhaps with a highlight on the awkward fact that she's attracted to her childhood friend.
This arc is probably strong enough to sustain us through the whole game. Our B-plots each episode could easily integrate in. We can talk about Kate, who faces social execution due to a video of something that was not her fault. We could talk about Chloe, who has no 'social' because people keep leaving her and she's burning on a combination of deeply loyal raging and deeply bitter self-isolation.
These segments integrate well with our personal arc. Our worries about our own inadequacies and mistakes are shown as funhouse mirrors, blown out of proportion. Some segments are minor, like Victoria's cronies briefly admitting their own social worries. Some segments are major, lasting almost the whole game and providing a constant thematic companion - Chloe, in this case.
However, our personal arc doesn't require any superpowers. That story could be told with no superpowers at all.
The role of the superpowers is not to advance Max's arc. Max is responsible for advancing Max's arc. Instead, the superpowers are a mirror of Max's arc, showing in more concrete terms how Max is advancing, and giving us a simple tool to help tie other arcs to Max's own arcs. Think about it: if Max's powers reflect Max's place in her own arc, then naturally they'll be good at interacting with arcs that also reflect Max's place in her arc. Max's powers can be a needle and thread to stitch arcs together. They give us, as authors, a tool to make Max's arc feel palpable and clear.
Let's make it plain:
We start Max with the power to briefly rewind time, which we initially use mostly to let the player re-do awkward social moments. This gives us plenty of opportunities to cement both who Max is and who she wants to be. This basically represents Max's ability to consider other people, rather than just wallowing in herself: via retrying, she can consider how other people will react and plan her conversation accordingly.
That said, the limits are more important than the capabilities.
We can meddle with other people's lives, such as stepping in for Kate and Chloe. But we quickly learn that no matter what we say, there are things that can't be fixed with a few careful lines of dialog.
The arc of the powers reflects her own personal journey: she's just starting off, so she is pretty limited.
We slowly ramp up her powers to interacting with physical things. First with miraculous timing, then by being able to hold items while rewinding. In the real game, these are all mixed together... but we're talking about our own theoretical version, so we can separate things out.
This growing power reflects Max's interest in being better at peopling. The things she can't do with a snippet of dialog, she can do with physical objects.
It's very possible to use these powers to tie arcs to our current situation.
For example, the "Victoria on the stairs" puzzle, where she blocks our way until we pour paint on her. We can't convince Victoria with words, so we convince her with actions. In the game, this is one of the worst puzzles, since it makes so little sense, takes so long, and is immediately rendered pointless with another nonsense puzzle. But thematically, it's a good idea. A similar puzzle could have worked well.
For example, the Warren Gets Beat Up scene could have been replaced with a Making Warren An Accidental Badass scene. That would have felt awesome, and would have allowed us to take Warren down a self-destructive path for later use in a different arc.
If we really wanted to have it be a Victoria scene, why not have it be something that makes sense? Victoria's goons found Max's satchel, and are saying that nobody can be SURE whose satchel it is, and that they should probably go turn it in to the lost in found... in Pensacola.
Both of these puzzles are about situations dialog can't resolve. Both of them are about people who get things done beyond the reach of simple dialog, and about people who are also suffering from concerns dialog can't reach. In fact, we could get away with outright stating that as a throwaway line. Max sneaks up on Victoria and friends, hears them talk: "It's all just talk! If talking could solve everything, my mom wouldn't be in the hospital dying." Etc, etc.
This section of Max's arc could culminate in the second episode, when Kate tries to commit suicide. In canon, this situation is beyond Max's powers of time control, and represents a chance for Max to step up and personally resolve the situation without relying on "cheats" - proving that she no longer needs them. It represents Max closing the circle: dialog spoken from the heart is powerful enough to do what her carefully calculated superpower dialog could not do. Max has outgrown her powers.
In canon, Max's arc changes from social anxiety to heroism. More specifically, the theme becomes "how far will you go to stay with someone you love", which is also a very potent arc. I have no problem with this arc change, because it builds off of Max's earlier arc in an organic way and is completely believable.
Unfortunately, the arc doesn't follow through: the author clearly didn't think it out well enough, and the story devolves into a macabre serial killer dream. While the individual elements are written well, they don't thematically tie in with either of Max's arcs, and the whole thing falls apart. The finale does drag it back around, but only after hours of serial killer nonsense actively detracted from it.
That weird serial killer sequence? It was a twist that came out of nowhere.
It was foreshadowed, but it still felt awkward and forced, because the themes didn't line up.
Creating gameplay based on your theming is a powerful tool for a lot of reasons. First, it gives you a great lead: no matter what you're trying to do in the game world, you know how it should feel and what it should be about, so it's much easier to develop. Second, it draws the player in and keeps them in.
We can craft moments of triumph and pain that work within our arc - either our social anxiety or our heroism. Finding that Rachel died and watching Chloe crumple will gut us either way, but subtle highlights on Max's reaction will tell us whether Max is struggling to support Chloe or rescue Chloe. Those are very different feelings and players will pick up on it, especially as Max's subsequent actions struggle to do one of those things specifically. IE, are we trying to go back in time to rescue Rachel? How about to prevent Chloe from meeting Rachel? How about just supporting Chloe with hugs, pot, and ice cream? How about taking advantage of Chloe to bind her to you emotionally? All of these are valid paths which support different arcs in different ways, and develop Max in different, interesting directions.
Which direction you choose is guided by what path you want to take. And, at the end of the game, the ending sequence will be about resolving that mess.
The canon ending is: sacrifice the town for Chloe, or visa-versa?
However, there are other, equally powerful possible endings. For example: wipe yourself out of existence. Chloe's dad wasn't there cooking lunch for you, never died. Choose which of two or three Chloes is "real", and face down your nightmares of the other Chloes - it's even worse because they forgive you. Rewrite history so many times that there is no "real" Chloe, only your Frankenstein's Chloe made up of a hundred different pasts stamped together with superglue.
Those are all "heroism" endings, but there are just as many "social anxiety" endings. Solve Chloe's problems, or let her work through them? Leave Chloe straight, or just tweak reality slightly so she's your love interest? Etc, etc.
Knowing our goal allows us to build up a path towards it, choosing gameplay, puzzles, B-arcs, and supporting cast that highlight our journey. Usually this is about building up Max's vector until it is out of control, then simply pointing out that it's out of control.
I say "vector" deliberately. This isn't about letting Max use her power to stop a serial killer or whatever. That's not part of either of these personal arcs.
Technically, sure her powers are being used. But not in a way that reflects her personal journey. It's just out of desperation - and live-or-die desperation isn't part of her personal arc, not unless it's on someone else's behalf.
The serial killer arc could have worked if Chloe was the captured victim - Chloe ran off to check the grave without you, and she was gone when you arrived. It would have been interesting to let Max go off the rails trying to rescue Chloe from ever-earlier points in time, watching reality disintegrate.
But I still hold that a serial killer wasn't even needed. Max's powers are the serial killer. Jefferson's role could easily have been to narrate Max's fall from innocence in a much less murdery way, by simply being a smooth-talking pervert. This would have been useful in the part of Max's social anxiety arc where she learns that she can't really know someone by simply collecting their responses.
In the end, Max choosing whether to sacrifice Chloe or the town only makes sense if Max's arc is "Chloe and I will be together at all costs". And that can be established best by creating this universal, stitched-together flow.
That's the best kind of foreshadowing and layered storytelling to me. The two episodes spent on serial killer stuff weren't stitched in properly, and that was the game's clearest failing... despite the fact that it was technically both foreshadowed and layered.
What I'm saying is all of the B-arcs, puzzles, and twists should come out of their thematic connections to the core arc.
...
Now, how about Night in the Woods?
Night in the Woods is really good and the same length as Life is Strange, go play it instead of reading spoilers!
Just foreshadowing doesn't make something feel like it belongs.
Recently I've seen a lot of games stumble over the concepts of foreshadowing and layered storytelling. Even really good games. So let's talk a little bit about it.
This might be a bit rough, it's my first take.
Games have twists. At the end, in the middle, twists everywhere. All these twists have to feel like they make sense.... but they usually don't. They often feel like they "came out of nowhere".
There's usually foreshadowing: nobody simply forgets to set up the ending. They just... foreshadow badly.
The two most common kinds of bad foreshadowing are the accidental reveal and NG+shadowing.
The accidental reveal is pretty obvious: the foreshadowing is so blatant that it simply gives away the twist.
NG+shadowing is the opposite. The foreshadowing is so minuscule or unrelated that you can only see it as foreshadowing when you're playing the game a second time.
You might think we're on the hunt for goldilocks moments: foreshadowing that isn't too blatant, but isn't too subtle.
Not me. I say both those kinds of foreshadowing are misunderstanding the role of foreshadowing.
Let's get... hoity-toity.
The twists that stay with me are the ones that are thematically relevant.
Let's say we're writing a game about a main character struggling through a rough time in their life. The backdrop to this is a series of mysterious murders: our MC lives their life, notices murders happening in the background. At the end of the game, the two plot lines come together and there's... some kind of cool twist.
If you're a writer, you need to think about how these two plot lines support each other. They're not just two stories happening in the same place: both need to be stronger because the other exists.
The first instinct is to make them related physically. Our struggling MC witnesses a murder. Or is suspected of being a murderer. Or one of the friends is a suspect. Simply moosh the two plot lines together and then stitch things up. It works. Ish.
A good writer will think about more than that. We need the two to support each other thematically. The serial killing is a twisted reflection of our MC's internal struggle. The twist at the end is how the two separate or merge, and it reflects not the meanderings of the plot, but the slow syncing of the two themes.
...
Let's discuss and compare Life is Strange and Night in the Woods. They both have the same main character and a similarly laid-back approach to gameplay, so they're a good contrast. Since NitW is new, I put the NitW spoilers behind a spoiler block. You can skip them.
...
In Life is Strange, our main character has psychic powers. These help her to mask her social anxieties by allowing her to simply undo social interactions. Initially, her personal arc is about coping with or overcoming her social anxiety - perhaps with a highlight on the awkward fact that she's attracted to her childhood friend.
This arc is probably strong enough to sustain us through the whole game. Our B-plots each episode could easily integrate in. We can talk about Kate, who faces social execution due to a video of something that was not her fault. We could talk about Chloe, who has no 'social' because people keep leaving her and she's burning on a combination of deeply loyal raging and deeply bitter self-isolation.
These segments integrate well with our personal arc. Our worries about our own inadequacies and mistakes are shown as funhouse mirrors, blown out of proportion. Some segments are minor, like Victoria's cronies briefly admitting their own social worries. Some segments are major, lasting almost the whole game and providing a constant thematic companion - Chloe, in this case.
However, our personal arc doesn't require any superpowers. That story could be told with no superpowers at all.
The role of the superpowers is not to advance Max's arc. Max is responsible for advancing Max's arc. Instead, the superpowers are a mirror of Max's arc, showing in more concrete terms how Max is advancing, and giving us a simple tool to help tie other arcs to Max's own arcs. Think about it: if Max's powers reflect Max's place in her own arc, then naturally they'll be good at interacting with arcs that also reflect Max's place in her arc. Max's powers can be a needle and thread to stitch arcs together. They give us, as authors, a tool to make Max's arc feel palpable and clear.
Let's make it plain:
We start Max with the power to briefly rewind time, which we initially use mostly to let the player re-do awkward social moments. This gives us plenty of opportunities to cement both who Max is and who she wants to be. This basically represents Max's ability to consider other people, rather than just wallowing in herself: via retrying, she can consider how other people will react and plan her conversation accordingly.
That said, the limits are more important than the capabilities.
We can meddle with other people's lives, such as stepping in for Kate and Chloe. But we quickly learn that no matter what we say, there are things that can't be fixed with a few careful lines of dialog.
The arc of the powers reflects her own personal journey: she's just starting off, so she is pretty limited.
We slowly ramp up her powers to interacting with physical things. First with miraculous timing, then by being able to hold items while rewinding. In the real game, these are all mixed together... but we're talking about our own theoretical version, so we can separate things out.
This growing power reflects Max's interest in being better at peopling. The things she can't do with a snippet of dialog, she can do with physical objects.
It's very possible to use these powers to tie arcs to our current situation.
For example, the "Victoria on the stairs" puzzle, where she blocks our way until we pour paint on her. We can't convince Victoria with words, so we convince her with actions. In the game, this is one of the worst puzzles, since it makes so little sense, takes so long, and is immediately rendered pointless with another nonsense puzzle. But thematically, it's a good idea. A similar puzzle could have worked well.
For example, the Warren Gets Beat Up scene could have been replaced with a Making Warren An Accidental Badass scene. That would have felt awesome, and would have allowed us to take Warren down a self-destructive path for later use in a different arc.
If we really wanted to have it be a Victoria scene, why not have it be something that makes sense? Victoria's goons found Max's satchel, and are saying that nobody can be SURE whose satchel it is, and that they should probably go turn it in to the lost in found... in Pensacola.
Both of these puzzles are about situations dialog can't resolve. Both of them are about people who get things done beyond the reach of simple dialog, and about people who are also suffering from concerns dialog can't reach. In fact, we could get away with outright stating that as a throwaway line. Max sneaks up on Victoria and friends, hears them talk: "It's all just talk! If talking could solve everything, my mom wouldn't be in the hospital dying." Etc, etc.
This section of Max's arc could culminate in the second episode, when Kate tries to commit suicide. In canon, this situation is beyond Max's powers of time control, and represents a chance for Max to step up and personally resolve the situation without relying on "cheats" - proving that she no longer needs them. It represents Max closing the circle: dialog spoken from the heart is powerful enough to do what her carefully calculated superpower dialog could not do. Max has outgrown her powers.
In canon, Max's arc changes from social anxiety to heroism. More specifically, the theme becomes "how far will you go to stay with someone you love", which is also a very potent arc. I have no problem with this arc change, because it builds off of Max's earlier arc in an organic way and is completely believable.
Unfortunately, the arc doesn't follow through: the author clearly didn't think it out well enough, and the story devolves into a macabre serial killer dream. While the individual elements are written well, they don't thematically tie in with either of Max's arcs, and the whole thing falls apart. The finale does drag it back around, but only after hours of serial killer nonsense actively detracted from it.
That weird serial killer sequence? It was a twist that came out of nowhere.
It was foreshadowed, but it still felt awkward and forced, because the themes didn't line up.
Creating gameplay based on your theming is a powerful tool for a lot of reasons. First, it gives you a great lead: no matter what you're trying to do in the game world, you know how it should feel and what it should be about, so it's much easier to develop. Second, it draws the player in and keeps them in.
We can craft moments of triumph and pain that work within our arc - either our social anxiety or our heroism. Finding that Rachel died and watching Chloe crumple will gut us either way, but subtle highlights on Max's reaction will tell us whether Max is struggling to support Chloe or rescue Chloe. Those are very different feelings and players will pick up on it, especially as Max's subsequent actions struggle to do one of those things specifically. IE, are we trying to go back in time to rescue Rachel? How about to prevent Chloe from meeting Rachel? How about just supporting Chloe with hugs, pot, and ice cream? How about taking advantage of Chloe to bind her to you emotionally? All of these are valid paths which support different arcs in different ways, and develop Max in different, interesting directions.
Which direction you choose is guided by what path you want to take. And, at the end of the game, the ending sequence will be about resolving that mess.
The canon ending is: sacrifice the town for Chloe, or visa-versa?
However, there are other, equally powerful possible endings. For example: wipe yourself out of existence. Chloe's dad wasn't there cooking lunch for you, never died. Choose which of two or three Chloes is "real", and face down your nightmares of the other Chloes - it's even worse because they forgive you. Rewrite history so many times that there is no "real" Chloe, only your Frankenstein's Chloe made up of a hundred different pasts stamped together with superglue.
Those are all "heroism" endings, but there are just as many "social anxiety" endings. Solve Chloe's problems, or let her work through them? Leave Chloe straight, or just tweak reality slightly so she's your love interest? Etc, etc.
Knowing our goal allows us to build up a path towards it, choosing gameplay, puzzles, B-arcs, and supporting cast that highlight our journey. Usually this is about building up Max's vector until it is out of control, then simply pointing out that it's out of control.
I say "vector" deliberately. This isn't about letting Max use her power to stop a serial killer or whatever. That's not part of either of these personal arcs.
Technically, sure her powers are being used. But not in a way that reflects her personal journey. It's just out of desperation - and live-or-die desperation isn't part of her personal arc, not unless it's on someone else's behalf.
The serial killer arc could have worked if Chloe was the captured victim - Chloe ran off to check the grave without you, and she was gone when you arrived. It would have been interesting to let Max go off the rails trying to rescue Chloe from ever-earlier points in time, watching reality disintegrate.
But I still hold that a serial killer wasn't even needed. Max's powers are the serial killer. Jefferson's role could easily have been to narrate Max's fall from innocence in a much less murdery way, by simply being a smooth-talking pervert. This would have been useful in the part of Max's social anxiety arc where she learns that she can't really know someone by simply collecting their responses.
In the end, Max choosing whether to sacrifice Chloe or the town only makes sense if Max's arc is "Chloe and I will be together at all costs". And that can be established best by creating this universal, stitched-together flow.
That's the best kind of foreshadowing and layered storytelling to me. The two episodes spent on serial killer stuff weren't stitched in properly, and that was the game's clearest failing... despite the fact that it was technically both foreshadowed and layered.
What I'm saying is all of the B-arcs, puzzles, and twists should come out of their thematic connections to the core arc.
...
Now, how about Night in the Woods?
Night in the Woods is really good and the same length as Life is Strange, go play it instead of reading spoilers!
Well I loooove spoilers! Click here to spoil NitW!
Just foreshadowing doesn't make something feel like it belongs.
Thursday, February 02, 2017
Galactic Line Missions
The mission system in The Galactic Line is the foundation that supports the personal-level play. Missions give a reason for the ships to exist and travel, gives goals to the crew, and throws complications in that give texture to their daily lives.
Missions are built on the resource system. So let's talk resources.
The most fundamental resources are the tech level economy points. Currently called "T0", "T1", etc, these replace both generic materials and scientific research. You don't unlock a new kind of superconductor, you get a T2 economy point. And these points get spent on building colonies and space ships. Simple enough.
There are also a variety of resources used to balance the construction of ships and colonies. Things like power, administration points, etc. These create a terrain for the players to build a variety of different kinds of ships and bases in a variety of different circumstances. Mods or custom settings can easily alter the playing field here.
But that's local stuff. Why do ships run around the cosmos? Here are some small possibilities:
1) Local rare resources. However, mining or research colonies are a better long-term investment.
2) Temporary situations, such as negative space wedgies or plagues.
3) Freight. Shipping resources around.
But those are not really enough to sustain a proper mission system. They're very basic and either too boring or too unpredictable.
Instead, the most critical trick is that resources can be local.
Local resources are simply resources that are about or from a specific place. For example, if you scan planet X-9, you get cartography points for planet X-9 specifically. Not generic cartography points.
There are many facilities which convert resources. Normally this is generic resources, like a solar array converting administration points into power. But a residential block converts cartography points into colonists, and a trade center converts cartography points into investment. These conversions maintain the localness: feed X-9 cartography in, you get X-9 colonists and X-9 investments. It takes time: conversions take some time, gathering resources takes time.
If you want to build a colony on X-9, you use X-9 colonists and investments.
This means your chain is to spend time at X-9, then to spend time at a colony world (or start those conversions up and come back for them later), then back to X-9 to found a colony. This has a fair number of gaps in it that can be used by other kinds of missions, allowing for multitasking.
For example, a simpler use of local resources is scanning a sun or solar system in general. Solar readings of the star X take some time to gather. You might be able to do it simultaneously as mapping X-9, depending on your setup. Either way, those localized readings can be used in a variety of ways. For example, they can be baked down on-board, converted into generic astrophysics points over a month or two. This doesn't require you to be anywhere specific, and can therefore run in the background wherever else you go.
Astrophysics points and local solar readings can both be used on any colony with a lab to create T2 economy points. It requires either a lot of astrophysics points, or readings from a lot of different stars. Either approach will pay off in the long run, and can be used at any colony you think needs more T2 economy. Of course, it takes time. You could drop those things off at the colony you're gathering colonists and investments from, do all three missions at the same time. Or you could start the colony missions, leave a crewmember at each, and fly off to take more star readings.
You also don't have to use those mapping points to build a colony. You could convert them into T1 economy points in much the same way. So your overall methods and objectives are up to you.
This is called an "open resource chain": a variety of resources that can be knit together in a variety of ways. This allows for mods and alternate builds to integrate in at any level. The key, from the dev perspective, is that local resources create a reason for star ships to keep moving. Also, during the missions complications arise, which gives players an opportunity to have a personal scenario where they interact with a bunch of people.
The actual interface for this is via "missions". This is a dense and complex environment, especially when the player doesn't know precisely what capabilities are available in which places. Missions offer a method to organize that.
A selection of missions are offered whenever you go someplace new, by simply batching up possible resource transactions. You go to X-9, you see some missions for mapping X-9, some missions for scanning the sun or the solar system. You also see missions for refurbishing your engines, studying for promotion, and a bunch of other ship-centric missions.
You can create more missions if you wish to customize things. It involves interacting with people that represent mission options, I'll cover that some other day.
After mapping, you go back to a colony. The residential district can convert local mapping into colonists, so you get at least one mission about that... and you can choose to look for more if you want other options.
These methods allow the player to have pretty good options offered no matter where they go, and they'll quickly learn "gut instinct" mission chains when they see certain resources can be converted in certain ways.
...
Galactic Line's core gameplay is scenes where you get to interact with people in a rather open way. This gameplay about choosing destinations and missions is a setup to help that happen in a convincing way, because mission complications and alternatives offer a method to unify and connect characters, as well as give them something to care about and a background event the player cares about.
This allows us to sidestep the problem of so many continuous-interaction games like The Sims. The player becomes inured to the soft, indifferent day-to-day pressures and begins to aim at the edges of the simulation. That's why so many people try to murder their sims, or get a ghost baby, or whatever. It's also why Rimworld players tend to vivisect their prisoners.
But if every "day" the player actually plays through is centered around a specific mission complication, the player will have a goal, the characters will have goals, and everyone will have a reason to share the stage and connect with each other.
As a demonstration of this, think about The Sims. Think about those specific events where you were throwing a party. You were probably focused on your goals for the party. For some people, this might have been "drowning everyone in the pool", but for most parties it was about raising friendship ratings, providing enough food, achieving sims goals, and maybe hooking someone up. There's a drive to get the party "running smoothly".
Imagine if playing the sims was done with much lower time pressures, but you didn't play each day. Instead, you had a weekly planner and only played the days when something important happens. It's a bit like that.
In theory.
Missions are built on the resource system. So let's talk resources.
The most fundamental resources are the tech level economy points. Currently called "T0", "T1", etc, these replace both generic materials and scientific research. You don't unlock a new kind of superconductor, you get a T2 economy point. And these points get spent on building colonies and space ships. Simple enough.
There are also a variety of resources used to balance the construction of ships and colonies. Things like power, administration points, etc. These create a terrain for the players to build a variety of different kinds of ships and bases in a variety of different circumstances. Mods or custom settings can easily alter the playing field here.
But that's local stuff. Why do ships run around the cosmos? Here are some small possibilities:
1) Local rare resources. However, mining or research colonies are a better long-term investment.
2) Temporary situations, such as negative space wedgies or plagues.
3) Freight. Shipping resources around.
But those are not really enough to sustain a proper mission system. They're very basic and either too boring or too unpredictable.
Instead, the most critical trick is that resources can be local.
Local resources are simply resources that are about or from a specific place. For example, if you scan planet X-9, you get cartography points for planet X-9 specifically. Not generic cartography points.
There are many facilities which convert resources. Normally this is generic resources, like a solar array converting administration points into power. But a residential block converts cartography points into colonists, and a trade center converts cartography points into investment. These conversions maintain the localness: feed X-9 cartography in, you get X-9 colonists and X-9 investments. It takes time: conversions take some time, gathering resources takes time.
If you want to build a colony on X-9, you use X-9 colonists and investments.
This means your chain is to spend time at X-9, then to spend time at a colony world (or start those conversions up and come back for them later), then back to X-9 to found a colony. This has a fair number of gaps in it that can be used by other kinds of missions, allowing for multitasking.
For example, a simpler use of local resources is scanning a sun or solar system in general. Solar readings of the star X take some time to gather. You might be able to do it simultaneously as mapping X-9, depending on your setup. Either way, those localized readings can be used in a variety of ways. For example, they can be baked down on-board, converted into generic astrophysics points over a month or two. This doesn't require you to be anywhere specific, and can therefore run in the background wherever else you go.
Astrophysics points and local solar readings can both be used on any colony with a lab to create T2 economy points. It requires either a lot of astrophysics points, or readings from a lot of different stars. Either approach will pay off in the long run, and can be used at any colony you think needs more T2 economy. Of course, it takes time. You could drop those things off at the colony you're gathering colonists and investments from, do all three missions at the same time. Or you could start the colony missions, leave a crewmember at each, and fly off to take more star readings.
You also don't have to use those mapping points to build a colony. You could convert them into T1 economy points in much the same way. So your overall methods and objectives are up to you.
This is called an "open resource chain": a variety of resources that can be knit together in a variety of ways. This allows for mods and alternate builds to integrate in at any level. The key, from the dev perspective, is that local resources create a reason for star ships to keep moving. Also, during the missions complications arise, which gives players an opportunity to have a personal scenario where they interact with a bunch of people.
The actual interface for this is via "missions". This is a dense and complex environment, especially when the player doesn't know precisely what capabilities are available in which places. Missions offer a method to organize that.
A selection of missions are offered whenever you go someplace new, by simply batching up possible resource transactions. You go to X-9, you see some missions for mapping X-9, some missions for scanning the sun or the solar system. You also see missions for refurbishing your engines, studying for promotion, and a bunch of other ship-centric missions.
You can create more missions if you wish to customize things. It involves interacting with people that represent mission options, I'll cover that some other day.
After mapping, you go back to a colony. The residential district can convert local mapping into colonists, so you get at least one mission about that... and you can choose to look for more if you want other options.
These methods allow the player to have pretty good options offered no matter where they go, and they'll quickly learn "gut instinct" mission chains when they see certain resources can be converted in certain ways.
...
Galactic Line's core gameplay is scenes where you get to interact with people in a rather open way. This gameplay about choosing destinations and missions is a setup to help that happen in a convincing way, because mission complications and alternatives offer a method to unify and connect characters, as well as give them something to care about and a background event the player cares about.
This allows us to sidestep the problem of so many continuous-interaction games like The Sims. The player becomes inured to the soft, indifferent day-to-day pressures and begins to aim at the edges of the simulation. That's why so many people try to murder their sims, or get a ghost baby, or whatever. It's also why Rimworld players tend to vivisect their prisoners.
But if every "day" the player actually plays through is centered around a specific mission complication, the player will have a goal, the characters will have goals, and everyone will have a reason to share the stage and connect with each other.
As a demonstration of this, think about The Sims. Think about those specific events where you were throwing a party. You were probably focused on your goals for the party. For some people, this might have been "drowning everyone in the pool", but for most parties it was about raising friendship ratings, providing enough food, achieving sims goals, and maybe hooking someone up. There's a drive to get the party "running smoothly".
Imagine if playing the sims was done with much lower time pressures, but you didn't play each day. Instead, you had a weekly planner and only played the days when something important happens. It's a bit like that.
In theory.
Monday, January 30, 2017
Galactic Line's Battle System
I've been vague about the battle system in The Galactic Line for a few reasons. I knew what I wanted it to include, but not how I wanted the pieces to come together. I don't want the game to devolve into yet another space ship battler. But I have worked it out now, and here's a quick description from the point of view of a commander.
Your ship and an enemy are stuck out in an asteroid field somewhere. You don't know where they are. When you zoom out, you can see a hex grid around your ship: a hex for your ship and six more neighbors. Three are above the horizon, three are below. As you rotate, the hexes do not.
As your sensors scan the skies, they pick up traces of your enemy. A bit of leftover exhaust, a bit of radiation, a glint of light. Dunno where they are at the exact moment, but we know what general part of the sky they're in, and some of those hexes darken. "Not that way". Which of the remaining hexes is the right direction? Who can tell?
Eventually it's down to two, then just one. You accelerate in that direction, trying to close. You don't have a solid lock, just a general knowledge that they're in that part of the sky.
Then lasers come crashing in, immediately blowing apart your armor and tearing through your systems. Yeah, it came from that part of the sky, but what the hell?
The sim ends and your competitor asks why you didn't launch any sensors or drones or anything.
So you start over.
This time you realize that those hexes are destinations as well as directions. You launch sensor drones into a hex, and in a few moments they arrive and begin scanning. You launch some missiles and have them wait quietly in another hex. You fire a spray of water in a direction you think the enemy might be: it turns into a cloud of ice, preventing scans from that direction from picking up your heat signature, breaking up your profile. Later that hex goes dark - the enemy isn't that way. You replace it with a pack of missiles, patiently waiting for a chance to attack.
You realize you can zoom out again. All seven hexes shrink into a central hex, and you're faced with six new, bigger hexes. Three high, three low. You send a sensor drone out into one of them, and notice it'll take quite a bit longer to reach.
You zoom out again. Those seven hexes shrink into one central hex, and a new six appear. Now you're looking at a fair chunk of the asteroid field - there's actually an asteroid in one of those hexes. Deploying sensors to it looks like it'll take half a day, and you think about setting course, but get the warning that turning on your engines will more than triple your profile.
And then you notice there's a glowing red perimeter around those hexes. Mouseover: "Enemy scan warning". The enemy has figured out you're in this volume of space, although it's obviously a pretty weak lock given the scales involved.
Now you're starting to get a handle on this. You've deployed some offensive and defensive systems, you're winnowing down the enemy's direction. This time, you get a lock first. You have to decide whether to open fire or refine your firing solutions: if you open fire, the enemy's gonna detect you pretty much immediately.
Well, you fire those missiles you set aside. The enemy shoots most of them down, but you get a much better firing solution from them and fire your own laser barrage. This time you win.
The next fight begins to unfold the same way, but when you start to get a good lock-on, you realize the two of you are within the same hex space. Your systems overlap, things become a mess, and then there are bombers on top of you.
As you progress, you learn a lot more tactics. ECM drones specifically calculate enemy scan patterns and help you avoid detection. Anti-missile interceptors. "Buzzers" that draw off the enemy sensors and incoming missiles. Sensor drones arrayed in "shotgun" lines along different scales of hexes. Weapon drones. Moving drones to new locations, or retrieving them. Radiation storms or ice clouds to mask your profile and the profile of your drones.
All of this is on top of managing your own profile. The higher your profile, the quicker the enemy will narrow down their search. You have these fancy high-tech deflector shields, but if you use them, your profile spikes so high the enemy will immediately detect you. You find out through experimentation that you're better off actually leaving them off even while you're under fire, if the range is long. A low profile helps more than a barrier.
Then you get to fleet logistics, and begin to struggle with the idea of deploying a ship to a distant hex and then having them have local hexes within that hex... and you begin to work with noncombatants and fixed fortifications, requisitioning or hacking for sensor logs, resources, triangulation help.
And that's how The Galactic Line's battles work.
Even a one-on-one, ship-on-ship battle takes place at great ranges, on a strategic map. There are moments when someone warps in right on top of someone else and a slugfest starts, but no matter how epic that feels, ships are expensive. Hugely expensive. And the people who die? They're people. Your people.
You begin to appreciate the art of getting a hard lock and then pinging the target with targeting lasers, forcing them to surrender without firing a shot. You begin to appreciate the art of convincing a local politician to get the enemy fleet to stand down.
Hopefully.
Anyway, the same system is used outside of combat, especially when scanning planets or assisting colonists.
... the same system is used for building colonies, actually. That's what gave me the idea.
I haven't settled on this 100%, but it plays well on paper. I can't really justify arcade-style action.
Your ship and an enemy are stuck out in an asteroid field somewhere. You don't know where they are. When you zoom out, you can see a hex grid around your ship: a hex for your ship and six more neighbors. Three are above the horizon, three are below. As you rotate, the hexes do not.
As your sensors scan the skies, they pick up traces of your enemy. A bit of leftover exhaust, a bit of radiation, a glint of light. Dunno where they are at the exact moment, but we know what general part of the sky they're in, and some of those hexes darken. "Not that way". Which of the remaining hexes is the right direction? Who can tell?
Eventually it's down to two, then just one. You accelerate in that direction, trying to close. You don't have a solid lock, just a general knowledge that they're in that part of the sky.
Then lasers come crashing in, immediately blowing apart your armor and tearing through your systems. Yeah, it came from that part of the sky, but what the hell?
The sim ends and your competitor asks why you didn't launch any sensors or drones or anything.
So you start over.
This time you realize that those hexes are destinations as well as directions. You launch sensor drones into a hex, and in a few moments they arrive and begin scanning. You launch some missiles and have them wait quietly in another hex. You fire a spray of water in a direction you think the enemy might be: it turns into a cloud of ice, preventing scans from that direction from picking up your heat signature, breaking up your profile. Later that hex goes dark - the enemy isn't that way. You replace it with a pack of missiles, patiently waiting for a chance to attack.
You realize you can zoom out again. All seven hexes shrink into a central hex, and you're faced with six new, bigger hexes. Three high, three low. You send a sensor drone out into one of them, and notice it'll take quite a bit longer to reach.
You zoom out again. Those seven hexes shrink into one central hex, and a new six appear. Now you're looking at a fair chunk of the asteroid field - there's actually an asteroid in one of those hexes. Deploying sensors to it looks like it'll take half a day, and you think about setting course, but get the warning that turning on your engines will more than triple your profile.
And then you notice there's a glowing red perimeter around those hexes. Mouseover: "Enemy scan warning". The enemy has figured out you're in this volume of space, although it's obviously a pretty weak lock given the scales involved.
Now you're starting to get a handle on this. You've deployed some offensive and defensive systems, you're winnowing down the enemy's direction. This time, you get a lock first. You have to decide whether to open fire or refine your firing solutions: if you open fire, the enemy's gonna detect you pretty much immediately.
Well, you fire those missiles you set aside. The enemy shoots most of them down, but you get a much better firing solution from them and fire your own laser barrage. This time you win.
The next fight begins to unfold the same way, but when you start to get a good lock-on, you realize the two of you are within the same hex space. Your systems overlap, things become a mess, and then there are bombers on top of you.
As you progress, you learn a lot more tactics. ECM drones specifically calculate enemy scan patterns and help you avoid detection. Anti-missile interceptors. "Buzzers" that draw off the enemy sensors and incoming missiles. Sensor drones arrayed in "shotgun" lines along different scales of hexes. Weapon drones. Moving drones to new locations, or retrieving them. Radiation storms or ice clouds to mask your profile and the profile of your drones.
All of this is on top of managing your own profile. The higher your profile, the quicker the enemy will narrow down their search. You have these fancy high-tech deflector shields, but if you use them, your profile spikes so high the enemy will immediately detect you. You find out through experimentation that you're better off actually leaving them off even while you're under fire, if the range is long. A low profile helps more than a barrier.
Then you get to fleet logistics, and begin to struggle with the idea of deploying a ship to a distant hex and then having them have local hexes within that hex... and you begin to work with noncombatants and fixed fortifications, requisitioning or hacking for sensor logs, resources, triangulation help.
And that's how The Galactic Line's battles work.
Even a one-on-one, ship-on-ship battle takes place at great ranges, on a strategic map. There are moments when someone warps in right on top of someone else and a slugfest starts, but no matter how epic that feels, ships are expensive. Hugely expensive. And the people who die? They're people. Your people.
You begin to appreciate the art of getting a hard lock and then pinging the target with targeting lasers, forcing them to surrender without firing a shot. You begin to appreciate the art of convincing a local politician to get the enemy fleet to stand down.
Hopefully.
Anyway, the same system is used outside of combat, especially when scanning planets or assisting colonists.
... the same system is used for building colonies, actually. That's what gave me the idea.
I haven't settled on this 100%, but it plays well on paper. I can't really justify arcade-style action.
Tuesday, January 10, 2017
Humanizing an Empire
As you may know, I really love games where you build things. And what is the biggest thing you can build?
An empire.
When I was younger, I was obsessed with 4X games like Civilization, Masters of Orion, and so on. As time wore on, their shine wore off. These days, I still play those kinds of games on occasion, but I prefer games where I build smaller things.
The reason is simple: those smaller things are more people-centric.
Recently, games about building empires have tried to become more people-centric. Civ animates their iconic rulers with loving care. Stellaris gives you a bunch of people with personal traits and names and asks you to assign them to colonies, star ships, factions. Everywhere, people are becoming more common.
The problem is that these games aren't about people. They never were, and they can't be: the structure of the game is oriented around nation-building, not personal dramas. This means interactions are all about bowing to the will of the empire.
Games like Crusader Kings II push this as far as can be reasonably expected, tracking hundreds of distinct individuals throughout lifetimes, letting you push them into service, crush them, or exploit them. This appeals to a certain kind of fantasy, I'm sure, but it's not really very good at letting me build something that means something.
This actively interferes with my enjoyment of these games. Civilization VI isn't about building a nation, it's about beating Alexander the Great as he weirdly and obsessively declares war against you every decade for a thousand years. It's humanized to the point where I don't think of that territory as another nation - I think of it as Queen Victoria's house. Vickie's front yard.
We've been humanizing these games for eons. Everyone remembers Gandhi's words backed by nuclear weapons. Everyone remembers the personality-filled cockpit views in Star Control. But these were very shallow, passive efforts - the difference between anthropomorphic tags and being a person.
Basically, when something is tagged, you can easily slip a personality onto it and ascribe a mood or emotion behind its actions. The Ur-Quan dreadnaught fighting the Spathi? It becomes the grumpy, remorseless Ur-Quan hunting down a hilarious, cowardly Spathi.
When something is portrayed as a person, things are different. The easiest example is Stellaris: a few minutes into the game, you meet a pirate faction made up of your own species in order to inject some kind of military action into the early game. You choose an admiral for your fleet. Do you pick Sara? She has a bonus to attack and a neat haircut. Doug? He learns fast and is wearing a space baseball jersey. Veronica? She can move the fleet faster and looks appropriately grumpy.
Whichever you choose, it's a person, not just a vague indicator. Instantly, you start thinking of your fleet ops as part of their career, part of their personal story. So you go to fight the enemy.
... and they have a fleet commander, too. A person. With a unique name, unique appearance, unique powers.
You kill them. You never learn anything about them. Ironically, you never even have enough info to give them a personality. They are the same species as you, so there's nothing distinctive enough to make you assign them the personalities you give to the other aliens you might meet.
This is distracting. Players naturally gravitate towards densely simulated things. You've taught the player that individual people matter, each one is simulated and tracked independently. Then you show them a person, as dense and complex and promising as the one they chose... and just kill them. Wordlessly, effortlessly, pointlessly.
Worse, the alien species cultures are largely randomized. When you do meet an alien, the anthropomorphic tags they bear don't correlate to their actual characteristics. This turns the emotional responses into mud. You've trained the player to respond to individuals, but then refuse to let them interact with individuals. You train the player to not care about species, because their appearance is unrelated to their traits, but then force them to interact with species.
Civ VI doesn't do as many things wrong. It has very, very strongly personified nations, but they have the correct personalities and feel reasonable. Unfortunately, you can't really interact with those people. You can only interact with their nations. Even that is in a vague, impersonal way.
On the other hand...
Let's say I make a game about building cars. There's a bunch of NPCs, they want cars that suit their personalities and needs. I build them a car. They go "whoaaaa! NICE!" They go drive it. They come back. "It was great, did X really well, thanks a ton!"
The scale is small enough that it feels like a personal interaction even when I don't do social things. The response is personal, about something concrete and part of their life. The choices I make are much more centered around facets of their personality, because each NPC can have wildly different taste in cars without becoming unbelievable.
It is already much more deeply personal than Civ VI, even if the NPCs are just a doodled pixel portrait.
Honestly, even if there were no NPCs, building the car is still more personal than building an empire in Civ VI, because I can easily imagine how someone would use and enjoy the car. That's a big part of why Space Engineers is fun! There's no NPCs, no personalities, but imagining how people would live in space, use my ships - that's fun!
In my mind, 4X games can't survive in this awkward place. They will naturally have to become either more humanized... or less.
"More" seems to be the trend, so how would that go?
The two basic paths are the Civ VI path and the Stellaris path, which is why I used them as examples.
The Civ VI path focuses on a few, highly detailed, carefully designed NPCs. Further down this path lies a game where you can interact with those NPCs in more detailed, complex ways. Expanding your empire probably matters less than establishing a good relationship with your neighbors. This will generally take the focus off the militaristic side of things and play up more complex interactions.
The easiest way to do it would be to make the lands of the empire play the role of the home and body of the other ruler. Rather than negotiate with the ruler as a lump sum, you interact with specific cities or lands in a wider variety of ways.
There's not really any way around this. We can't continue having a bleak, war-focused set of interactions if we keep humanizing the other side. I already have no interest in going to war, even with the relatively weak personalities in these games today.
The other direction is to track an impressive number of people. Rather than the ruler of each nation, you track everyone. Politics constantly evolves and changes who is in what positions. The rulers of each nation come and go.
I have a suspicion this is definitely going to spawn a genre, but I don't think I'll like it much. That many people means that each person will be disposable, temporary. That's not going to scratch my itch.
At the very least, I hope all these people have personality tags, so I can easily assign them a mood and emotion as they do things. At the moment it's all undifferentiated mush, and it's very hard to establish any deep connection to the various NPCs. It'd be interesting to try to improve that connection while still having a large number of randomly-generated NPCs. Maybe use some techniques to inject personalities into the flow of the game.
But right this instant, I have a car-building prototype to create.
An empire.
When I was younger, I was obsessed with 4X games like Civilization, Masters of Orion, and so on. As time wore on, their shine wore off. These days, I still play those kinds of games on occasion, but I prefer games where I build smaller things.
The reason is simple: those smaller things are more people-centric.
Recently, games about building empires have tried to become more people-centric. Civ animates their iconic rulers with loving care. Stellaris gives you a bunch of people with personal traits and names and asks you to assign them to colonies, star ships, factions. Everywhere, people are becoming more common.
The problem is that these games aren't about people. They never were, and they can't be: the structure of the game is oriented around nation-building, not personal dramas. This means interactions are all about bowing to the will of the empire.
Games like Crusader Kings II push this as far as can be reasonably expected, tracking hundreds of distinct individuals throughout lifetimes, letting you push them into service, crush them, or exploit them. This appeals to a certain kind of fantasy, I'm sure, but it's not really very good at letting me build something that means something.
This actively interferes with my enjoyment of these games. Civilization VI isn't about building a nation, it's about beating Alexander the Great as he weirdly and obsessively declares war against you every decade for a thousand years. It's humanized to the point where I don't think of that territory as another nation - I think of it as Queen Victoria's house. Vickie's front yard.
We've been humanizing these games for eons. Everyone remembers Gandhi's words backed by nuclear weapons. Everyone remembers the personality-filled cockpit views in Star Control. But these were very shallow, passive efforts - the difference between anthropomorphic tags and being a person.
Basically, when something is tagged, you can easily slip a personality onto it and ascribe a mood or emotion behind its actions. The Ur-Quan dreadnaught fighting the Spathi? It becomes the grumpy, remorseless Ur-Quan hunting down a hilarious, cowardly Spathi.
When something is portrayed as a person, things are different. The easiest example is Stellaris: a few minutes into the game, you meet a pirate faction made up of your own species in order to inject some kind of military action into the early game. You choose an admiral for your fleet. Do you pick Sara? She has a bonus to attack and a neat haircut. Doug? He learns fast and is wearing a space baseball jersey. Veronica? She can move the fleet faster and looks appropriately grumpy.
Whichever you choose, it's a person, not just a vague indicator. Instantly, you start thinking of your fleet ops as part of their career, part of their personal story. So you go to fight the enemy.
... and they have a fleet commander, too. A person. With a unique name, unique appearance, unique powers.
You kill them. You never learn anything about them. Ironically, you never even have enough info to give them a personality. They are the same species as you, so there's nothing distinctive enough to make you assign them the personalities you give to the other aliens you might meet.
This is distracting. Players naturally gravitate towards densely simulated things. You've taught the player that individual people matter, each one is simulated and tracked independently. Then you show them a person, as dense and complex and promising as the one they chose... and just kill them. Wordlessly, effortlessly, pointlessly.
Worse, the alien species cultures are largely randomized. When you do meet an alien, the anthropomorphic tags they bear don't correlate to their actual characteristics. This turns the emotional responses into mud. You've trained the player to respond to individuals, but then refuse to let them interact with individuals. You train the player to not care about species, because their appearance is unrelated to their traits, but then force them to interact with species.
Civ VI doesn't do as many things wrong. It has very, very strongly personified nations, but they have the correct personalities and feel reasonable. Unfortunately, you can't really interact with those people. You can only interact with their nations. Even that is in a vague, impersonal way.
On the other hand...
Let's say I make a game about building cars. There's a bunch of NPCs, they want cars that suit their personalities and needs. I build them a car. They go "whoaaaa! NICE!" They go drive it. They come back. "It was great, did X really well, thanks a ton!"
The scale is small enough that it feels like a personal interaction even when I don't do social things. The response is personal, about something concrete and part of their life. The choices I make are much more centered around facets of their personality, because each NPC can have wildly different taste in cars without becoming unbelievable.
It is already much more deeply personal than Civ VI, even if the NPCs are just a doodled pixel portrait.
Honestly, even if there were no NPCs, building the car is still more personal than building an empire in Civ VI, because I can easily imagine how someone would use and enjoy the car. That's a big part of why Space Engineers is fun! There's no NPCs, no personalities, but imagining how people would live in space, use my ships - that's fun!
In my mind, 4X games can't survive in this awkward place. They will naturally have to become either more humanized... or less.
"More" seems to be the trend, so how would that go?
The two basic paths are the Civ VI path and the Stellaris path, which is why I used them as examples.
The Civ VI path focuses on a few, highly detailed, carefully designed NPCs. Further down this path lies a game where you can interact with those NPCs in more detailed, complex ways. Expanding your empire probably matters less than establishing a good relationship with your neighbors. This will generally take the focus off the militaristic side of things and play up more complex interactions.
The easiest way to do it would be to make the lands of the empire play the role of the home and body of the other ruler. Rather than negotiate with the ruler as a lump sum, you interact with specific cities or lands in a wider variety of ways.
There's not really any way around this. We can't continue having a bleak, war-focused set of interactions if we keep humanizing the other side. I already have no interest in going to war, even with the relatively weak personalities in these games today.
The other direction is to track an impressive number of people. Rather than the ruler of each nation, you track everyone. Politics constantly evolves and changes who is in what positions. The rulers of each nation come and go.
I have a suspicion this is definitely going to spawn a genre, but I don't think I'll like it much. That many people means that each person will be disposable, temporary. That's not going to scratch my itch.
At the very least, I hope all these people have personality tags, so I can easily assign them a mood and emotion as they do things. At the moment it's all undifferentiated mush, and it's very hard to establish any deep connection to the various NPCs. It'd be interesting to try to improve that connection while still having a large number of randomly-generated NPCs. Maybe use some techniques to inject personalities into the flow of the game.
But right this instant, I have a car-building prototype to create.
Thursday, January 05, 2017
Stats, Classes, and Fantasy Races
In Ye Olden Days of RPGs, you would roll stats straight. Only then would you start thinking about who the character was. You would choose a class to leverage whatever decent stats you had, and you might choose a race to push your understrength stats into a viable range. So the races and classes were less about building a character you wanted to play, and more about coping with brutal random chance.
These days, we build our character stats along whatever paths we want. There's no need to cope with randomization, but we still have the same fundamental structure. We've gotten used to it, but it's pretty bad. The added complexity doesn't serve any real purpose aside from making the game's high-level play be almost entirely about knowing which stats and classes and races interact in which ways.
If we want to make our game more approachable, that is a useless kind of skill. We want our players to excel at playing the game, not excel at setting it up.
This is a common concern, and many modern RPGs reflect this tendency. We've begun ripping out stats or classes or both.
But there is power in the gap between stats, classes, and races. The way they augment each other is quite dense and interesting. Just ripping out those cornerstones means you lose all that depth!
Without the resonance between the layers, you end up with a very shallow approach and extremely straightforward builds.
Easy example: Skyrim. No stats or classes, just a few skills. Worse, improving any skill is incredibly expensive, since the enemies scale with the number of skill points you earn, not your maximum combat skills. This means your approaches are going to be both extremely basic and extremely focused. Moreover, many approaches seem viable at first but at midgame peter out entirely - for example, pure archers don't end up working out, and having to fall back on melee means you're better off just building a melee character.
A few recent games have tried to keep the complexity of stats, class, and race, but they tend to end up feeling mishmashy and sodden. Pillars of Eternity is a good example: their attempt to make all builds viable results in all builds feeling largely the same.
Statistically speaking, that's what they have to do. Otherwise you're back where we started: learning to minmax would be the primary player skill, rather than learning to play.
Is there any other approach?
Sure! Here's a simple idea:
Stats and classes, basically normal setup. However, the stats don't give bonuses. Instead, they are capacity.
For example, if you have 8 strength you can carry 8 points of gear. Say, leather armor (6 points) and a broadsword (2 points). 10 intellect? You can focus on 10 points of passive skills. Say, detect hidden (3 points), monster lore for crits vs monsters (4 points) and deflect criticals (3 points). 6 points of spirit? You can equip 6 points of spiritual gear/enchantments/spells. Magic deflection (4 points) and a ring of sparks (2 points).
This is just one idea, but the core here is that stats don't have to offer statistical effects. By splitting the class and stats into orthogonal concerns, we can give both a much wider range. A magician with a high strength might seem like a waste, but in practice it allows her to carry 3 staves and 4 wands. She can just pull the right one out in any given situation!
It also allows for midgame optimizations. You want to shave a point off the strength price of that spear? It might be more important than giving it a magical +2 damage bonus! You want to merge a bunch of magic rings into a single artifact set with a lower total spirit cost but ruin them as separate objects? Sure, sounds good!
Races? Sure. Lower the price for specific categories of gear. Elves get bows one point cheaper. Dwarves get armor one point cheaper. Maybe you have a family heirloom? It's not great, but it's super cheap for you alone. No need to modify the stats!
In addition, I find this just flat-out makes for more memorable characters: their equipment variation will be cleaner and sharper.
If you're only controlling a single character, this kind of approach is a bit iffy. If you make it light enough for a beginner, it will usually be uninteresting after a session or two. Make it heavy enough, and people will forget their setup after a week away. But if you're controlling a party, it's very easy to make it work out.
The lightweight approach combined with gear serving specific roles means that players can tell who a character is and what a character does at a glance. Even after a week away, the gear will still be easy to see and you can slip right back into the flow of things. If you're worried about stagnation, a steady trickle of new gear variants or abilities that change the prices will keep things swirling nicely.
Anyway, I thought I should mention it.
Stats vs classes vs races: it was originally built for a completely different purpose. It's ok to not do things that way. It's ok to reinvent the wheel.
Just make sure your new wheel turns, or you'll end up with an oversimplified pile of mush.
These days, we build our character stats along whatever paths we want. There's no need to cope with randomization, but we still have the same fundamental structure. We've gotten used to it, but it's pretty bad. The added complexity doesn't serve any real purpose aside from making the game's high-level play be almost entirely about knowing which stats and classes and races interact in which ways.
If we want to make our game more approachable, that is a useless kind of skill. We want our players to excel at playing the game, not excel at setting it up.
This is a common concern, and many modern RPGs reflect this tendency. We've begun ripping out stats or classes or both.
But there is power in the gap between stats, classes, and races. The way they augment each other is quite dense and interesting. Just ripping out those cornerstones means you lose all that depth!
Without the resonance between the layers, you end up with a very shallow approach and extremely straightforward builds.
Easy example: Skyrim. No stats or classes, just a few skills. Worse, improving any skill is incredibly expensive, since the enemies scale with the number of skill points you earn, not your maximum combat skills. This means your approaches are going to be both extremely basic and extremely focused. Moreover, many approaches seem viable at first but at midgame peter out entirely - for example, pure archers don't end up working out, and having to fall back on melee means you're better off just building a melee character.
A few recent games have tried to keep the complexity of stats, class, and race, but they tend to end up feeling mishmashy and sodden. Pillars of Eternity is a good example: their attempt to make all builds viable results in all builds feeling largely the same.
Statistically speaking, that's what they have to do. Otherwise you're back where we started: learning to minmax would be the primary player skill, rather than learning to play.
Is there any other approach?
Sure! Here's a simple idea:
Stats and classes, basically normal setup. However, the stats don't give bonuses. Instead, they are capacity.
For example, if you have 8 strength you can carry 8 points of gear. Say, leather armor (6 points) and a broadsword (2 points). 10 intellect? You can focus on 10 points of passive skills. Say, detect hidden (3 points), monster lore for crits vs monsters (4 points) and deflect criticals (3 points). 6 points of spirit? You can equip 6 points of spiritual gear/enchantments/spells. Magic deflection (4 points) and a ring of sparks (2 points).
This is just one idea, but the core here is that stats don't have to offer statistical effects. By splitting the class and stats into orthogonal concerns, we can give both a much wider range. A magician with a high strength might seem like a waste, but in practice it allows her to carry 3 staves and 4 wands. She can just pull the right one out in any given situation!
It also allows for midgame optimizations. You want to shave a point off the strength price of that spear? It might be more important than giving it a magical +2 damage bonus! You want to merge a bunch of magic rings into a single artifact set with a lower total spirit cost but ruin them as separate objects? Sure, sounds good!
Races? Sure. Lower the price for specific categories of gear. Elves get bows one point cheaper. Dwarves get armor one point cheaper. Maybe you have a family heirloom? It's not great, but it's super cheap for you alone. No need to modify the stats!
In addition, I find this just flat-out makes for more memorable characters: their equipment variation will be cleaner and sharper.
If you're only controlling a single character, this kind of approach is a bit iffy. If you make it light enough for a beginner, it will usually be uninteresting after a session or two. Make it heavy enough, and people will forget their setup after a week away. But if you're controlling a party, it's very easy to make it work out.
The lightweight approach combined with gear serving specific roles means that players can tell who a character is and what a character does at a glance. Even after a week away, the gear will still be easy to see and you can slip right back into the flow of things. If you're worried about stagnation, a steady trickle of new gear variants or abilities that change the prices will keep things swirling nicely.
Anyway, I thought I should mention it.
Stats vs classes vs races: it was originally built for a completely different purpose. It's ok to not do things that way. It's ok to reinvent the wheel.
Just make sure your new wheel turns, or you'll end up with an oversimplified pile of mush.
Subscribe to:
Posts (Atom)