That's basically the idea of this topic: coming up with the layout for a Marathon map when all you start with is a blank field on which to place things.
My big discovery in trying to slowly piece together something for Marathon: Grendel is that I suck at this part of the process. And it's not the big stuff that gets me - I'm okay at coming up with the basic dynamic for what makes a level work - it's the little stuff; I go to start placing rooms (because you have to have rooms, after all, and maybe even a few corridors to connect those rooms), and my mind goes blank. I simply don't know what to do.
So, here's the big question: what do YOU do at this point? You've got an idea of what your level is going to be like; you can picture the general feel you're going for, and maybe you've even got the core dynamic that will make the level work. Now how do you decide how to lay out the rooms and corridors that will make up the level proper?
Just thought this might make for some good discussion - I've never heard anyone talk about this, and yet it seems so critical to map-making.
Obviously, first comes gameplay: how will the room/corridor/ledge affect the player's movements and strategy? You want something that looks cool, but doesn't snag the player or put them at a great disadvantage. As such, walls and ceilings are usually used as great areas to add architectural details. The convenient thing about Marathon textures is that they are often tiled to fit .25WU or .5WU increments - which makes it easy to add differently-sized columns and light strips and ceiling struts that look like they fit.
Next is just figuring out how to implement these details, which just takes time to learn. The hardest thing for me to learn was making sure the architecture looked meaningful. Early on, I'd think "I want to add a column here" and oftentimes it would look sort of forced or out of place, or like it didn't go anywhere. Make sure structural details follow through; make sure those details have a sort of beginning and end, if that makes any sense.
Really though, it just comes down to trial and error, and experience. Don't be afraid to go two steps back to make the layout of a room work. Be objective, and know when something works/doesn't. And taking a look at existing maps really helps as well. In all honesty, when it gets down to the little details, there's only so much you can do with the Marathon engine (oh, a waterfall here! A light strip! A column of blocks!), so looking at existing maps is a great way to see how other people handled the little things.
And, of course, sometimes it's okay to just have a big blank wall. Some of my older maps are overflowing with detail and it gets a little overbearing at times, I think. It's all a balancing act.
EDIT: Also, don't be afraid to change your original concept of the whole map if it turns out it's not quite what you thought it'd be, or there ends up being a better way. It may take more time, but in the end you'll be rewarded with a better map.
PerseusSpartacus wrote: I go to start placing rooms (because you have to have rooms, after all, and maybe even a few corridors to connect those rooms), and my mind goes blank. I simply don't know what to do.
So, basically you're stuck on the entire concept of map design.
PerseusSpartacus wrote:So, here's the big question: what do YOU do at this point?
You mean at the starting point?
PerseusSpartacus wrote:I've never heard anyone talk about this, and yet it seems so critical to map-making.
Yes, making a map is somewhat critical to the process of making a map. Because, let's be honest, what you're spending several paragraphs saying is: "I haven't done anything and I don't know how to get started. Help."
PerseusSpartacus wrote:Now how do you decide how to lay out the rooms and corridors that will make up the level proper?
So, here's my less snide, possibly more helpful response, it's very long. It might not work for you, different strategies work for different people, but this is how I go about things.
First, you need tangible, concrete goals. As Yogi Berra once said, "If you don't know where you're going, you're never gonna get there." This is really important. If you don't have a concrete goal to work towards then you're just going to end up floundering and dithering about; "the basic dynamic for what makes a level work" is not a concrete goal.
Here are two ways to create concrete goals to work towards:
1) Decide on 1-3 specific combat spaces that you absolutely want to include in your level. Focus on making them. Later, connect these spaces in any way you see fit, hallways, spaces, teleporters, whatever. However, those combat areas are your tangible goal. Just pick something you think would be fun or cool or whatever. If you can't think of anything cool then just make a big room with some pillars and see where that takes you. Or try to recreate a combat space from some other game.
2) Decide on a theme to the combat that takes place in your level. Some gimmick or pattern of encounters. Then go to the above paragraph and think of specific combat spaces that would facilitate that theme.
Second, quickly iterate your ideas. I usually sketch out my ideas on a scrap piece of paper. It takes maybe 10-15 seconds to sketch out a room or an entire level depending on what I'm focusing on.
Not only does this type of iteration help me get a feel for the level and help me work out my ideas quickly, but it also acts later as a concrete goal to work towards. When I'm stuck and don't know what to do next, I look at my sketches and then I have something tangible to work towards.
Three, do not worry about quality or quantity. Do not be a perfectionist. Just focus on getting it done.
Four, be adaptable. As you actually start making the maps you'll have to change a lot from your original ideas. You've just got to stay on your toes and keep an eye out for openings to create interesting moments or areas that you hadn't anticipated.
Here's how that process played out in the second map of my "start of an unnamed mod" which I just uploaded a few days ago to simplici7y, if you feel like reading it:
The level is the tower where futuristic tech is being developed, and this is where the player first gets the fusion pistol. So my goals were, a) have a room that reveals the fusion pistol to the player in a cool way, b) have a dark room of pillars (inspired by a room in Cool Fusion, which scared me a lot as a kid).
I also wanted, c) a large open area where the S'pht's flying abilities would be shown off to the player (because I wanted to underscore the S'phts' very different behavior).
And I also wanted, d) hallways that involved architecture from Beware of Low Flying Defense Drones, because I always thought that entire level looked like a futuristic laboratory.
So there I had my four concrete goals, which was probably one or two too many: a) fusion pistol room, b) cool fusion chamber, c) spht arena, d) futurey hallways (and I had specific, concrete goals for each of these spaces as well, but I'm not going to list those).
I sketched out these ideas on paper and came up with ways to connect the combat spaces/set pieces, experimented with some different arrangements to see what worked, and then I got started making the actual map.
Somewhere in there I noticed that the actual Beware of Low Flying Defense Drones has a lot of areas where you can shoot enemies through windows. I decided I liked that a lot and wanted to use it as a theme in my level. So I made sure there were a lot of spaces (at least on the second floor) where the player is physically separated from the enemies, but where they both can still shoot at each other. This "theme" of attacking into out-of-reach spaces informed a lot of the design on the second and third floor (and actually the second and third floor should have been a separate level from everything that came before it).
I failed to not be a perfectionist. I took way to long to make this level because I just couldn't stop editing it. Some of the combat spaces (the cool fusion chamber, the fusion pistol room) went through ~20 iterations each where I changed different aspects of each room, and ~4 iterations each where I deleted every single polygon and started from scratch. That is just stupid, but I can't stop myself, my perfectionism nags at the back of my head. And I'm still not entirely happy with either of those rooms either.
I hope that's in some way helpful.
Again, as far as I'm concerned, the absolutely number one, most important thing is to set up some specific, concrete goals of what you want to make, and then work towards those goals.
To build mostly on what Windbreaker said, be adaptable. I find a map can look very different when it's actually constructed in 3D than it did in my head. For example, take the room in the distance on the right here from The Goose & The Hog. I knew I wanted this to be the highest of the 4 loosely-defined rooms of the level, and I wanted to connect it on the back side to a split-level hallway via a long window. Those parts turned out as intended.
But originally, I wanted this room to be enclosed, such that players inside could shoot out from some windows, but were well protected. I also wanted a square pillar somewhere in the center of the room, figuring that two players fighting each other in the room wouldn't have enough options without some cover. I tried different-sized windows, and I pushed the pillar around to different parts of the room. But both elements were just obstructing the flow too much. People in the valley below were looking at a dead end, which sucked even if they got the rocket launcher.
I tried putting a small platform in the two corners of the valley, but the paths for people down there were still too predictable. I had to bite the bullet and open up the room above. I knocked out the two walls, and turned the pillar into a little dais, because I still wanted to break up the room visually, and it created a natural spot to put items. It also created a nice little duality with the yellow inset part of the valley.
The key to all this was rapid prototyping. Early and often, I would save the level in Weland and jump into it in Aleph One. Make sure your workflow allows you to do this with a minimal number of steps. I might apply landscape textures, since it changes the feeling a lot when it's 'outside', or paint some big walls where I had a particular texture or light in mind. But beyond that, I make a point of settling all the geometry before texturing and lighting, since virtually any geometry change means having to redo one or the other.
Also, I tend to work fairly organically. I don't build multiple rooms disconnected from each other and connect them later. If the distance between them is off, the tools available don't have the ability to grab a hundred-polygon chunk and move it closer to another chunk.
I find it useful to think in stories. They're not always 1 WU, but I do think in terms of, I want this passage to be one floor higher than the main floor, so a player here has a height advantage. This comes in handy when you want your level to have verticality while also having paths that interconnect in interesting ways.
Crater Creator wrote:Also, I tend to work fairly organically. I don't build multiple rooms disconnected from each other and connect them later. If the distance between them is off, the tools available don't have the ability to grab a hundred-polygon chunk and move it closer to another chunk.
If this is a response to one part of my suggestions then I have to say I didn't write what I was thinking. It's written as, "build disconnected rooms and then build connecting spaces later", but what I had intended to say was more like, "conceptualize the different rooms, conceptualize any way to connect them that works, and then start building them".
Also, the distance being off between rooms is a much bigger problem in multiplayer than in single player.
Interesting. I'd say having a concrete goal is definitely a must. But I do it quite differently; I first picture the level in my head, and always ask myself: where does this door lead? Do I need a corridor here? Is the flow okay? Of course it ends up looking quite different in-game, usually less cool and creative, but that's not your fault (Engine's Limitations).
$lave wrote:Damnit bridgit, you are forgetting how fucking serious business the internet is.