next »

[ichi]

I couldn't find a topic about this, so I thought this might be the best place to ask.
I'm working on a platform game, and the biggest obstacle I didn't see coming is level design - I didn't expect it to be so hard! 
What methods do you use? do you build them from blocks then slap all the graphics on them or do you sketch them on paper first? what do you guys do for inspiration?
I'm talking strictly from a gameplay standpoint, not visually.

[Saint]

I'm no authority on the subject but here goes..

Hmm, it depends a lot on the flexibility of the tools; If they have nice building blocks and allow for rapid prototyping there is less need for an initial sketch.

Usually, I start out with a "theme" for the level - this is very important for the visual setting, yes, but it is also important for the layout of the level as you may want to make it feel right (large flat spaces for grass levels or tall narrow spaces for "sinister tower" levels), or intentionally make something unexpected. From there I figure out what your main objective will be and how you'll get there (if there is such a thing, usually going from point A to B via C, D, E...) and make a quick sketch of the rooms or areas in the path I want the player to take. There are a few do:s and don't:s here (for example, it's often a bad idea to force the player to backtrack entire levels) but most of it is a matter of personal taste. After that, with the topology in front of me, I start fleshing out the map with additional rooms and other ways to reach existing ones - some hidden, some obvious.

From there, it largely depends on what game I'm making how much more planning I do before I start building, but usually I make a first implementation of the static geometry in the level to get a sense of how it looks. Again, depending on the game I may add jumping puzzles, static dangers or savepoints and stuff at this stage. Then comes the detailing.

For specific gameplay details in levels, I try to get inspired by great games in the same or related genres, but also by movies and life itself. Even the most mundane action movies (and anime) usually have a few moments that convert into great twitch gameplay, so keep an open mind. Again, theme can be a big help in finding inspiration.

... Oops, it seems I need to leave for work ten minutes ago. Hope this helps to start a discussion, at least =)

[Jimbob]

Usually I think "What the hell is the player going to have to do now that is different to what has come before" and subsequently "What can I do to see if the player has learnt something from something else I have done before" (or in the case of games like IWBTG and Syobon - how can I screw with these lessons). If you have a series of objects to populate your levels with, or if you look at what actions your character has, then think of the different combinations of both. I mean, most of Mario's actions are applied with 'jump', 'run' or even just 'walk' (heading blocks, bouncing on enemies, picking up/kicking shells).

And from those two points, I start planning out a series of little challenges (even if it's simply a specific arrangement of platforms). The actual layout may not be specifically linear or non-linear, but the player should never be made to feel that they are doing the same thing in the same environment over and over again. Well, that sort of thing doesn't float my boat anyhow...

[joshg]

I guess it's probably safe to say, you should totally finalize any gameplay mechanics your player uses to move around (jumping, wall climbing, whatever) before spending too much time on level design? 

I haven't made a platformer yet but I'd imagine that it'd be a huge pain to map out half a dozen awesome levels and then decide that it really would be cooler if your player could jump a little higher ... thus breaking any jumping puzzles and such.

[Melly]

It does seem wiser to work on the gameplay heavily before doing level design (it's what I did for Square Hero, and it seemed to pay off).

Level design is something tricky, and I'm no expert in the subject, but I'll share what I noticed from experimenting with it, and anyone is free to tell me I'm wrong ;P.

It depends on what you want out of the game, what you want the player to feel, to do. If you want to test your player's skill continuously, create consistently increasing challenges. You can create levels that require a specific type of strategy to succeed, then the next one makes that strategy impossible, and the player keep his mind active trying to find a new solution.

Level design can also affect the behaviour of players in a game. That's not something I'm very good at still, but I believe the layout of your levels can lead your players to do what you want them to. Like, for example, if you start the game with deep pits with spikes on the bottom (that you can't see from the top), you can assume that the player will learn not to explore any pits in fear that he may die, especially in games with limited tries/lives or with far-apart checkpoints. You can use that to hide secrets later on, for example.

The layout of the level can also bring out emotions from the player. Scary games usually have levels compozed of tight, maze-like passages, complete with possible dead-ends and pitch-black corners, that make the player feel confused and alone. It's something that, even if possibly accidentaly, Metroid for the NES did with its maze-like underground world where many passage looked the same as eachother (mostly due to the console and cartridge limitations, that didn't allow for much in the way of different graphic tilesets). An open area feels different to go through from a tightly closed one.

Just see what you want out of your game and try to think how you can design your levels to squeeze the most out of those concepts.

[dustin]

I'd check out this article it's about hiding secrets in platform games but still it's very well written and useful

http://www.sirlin.net/Features/feature_platformsecrets.htm

There are also a ton of other articles in the side bar to read...

[AdamAtomic]

When I'm designing levels I like to think about them from the player's perspective...or rather how do you construct enjoyable sequences of architecture that are fun given the player's movement set?  Here are some notes I made when I was doing level design for a game a while back, dunno if this will help or not!

> Series of speed jumps across hanging cages

> Jump -> Slide -> Fall feels great AND you can keep momentum now!
   > Optionally you can fall very very far but do a somersault and live! (good way to end?)
   > Can start with a wall jump if you want

> Should always start with a kickstop right next to player.

> Double jumps are a nice ending too

> Subtle changes in elevation on long jumps is a cool way to change things up

> Ceiling slides are an advanced maneuver.  Ceiling slides over spikes is just mean!

> The lip flip is fun, provides good closure or pause in flow.
   > Double wall jump feels same - looks cool, has rhythm still.

> At least one course needs to be based around slowly spinning blades where timing is everything...

> What about a big spinning thing, which you have to run over, then you drop down below it and have to wall-run up it in order to advance?  that'd be pretty neat...

> How well do angled launch ramps work?

> landing in a bunch of props or barrels

> big falls while keeping momentum

> you can run off ledges, fall a little, and still have enough momentum to wall run

> Add pacing by providing opportunities for the player to try that area again without restarting and without
having to hunt high and low for a usable kickstop.  Don't kill them just because they don't have it memorized!







> Ok!  Remember though - its only an addictive speed run level if there are ways to sort of cheat or do cool tricks to shave a second off the clock time here and there...like areas where doing it a certain way lets you keep momentum longer, or there is a jump or a corner/mantle you can skip, etc.

> So one easy way to do that is just to have sort of two intertwined paths throughout the course.  Let users choose the path that they think will be fastest, and have the paths cross regularly for maximum effect!  E.g. allow players to slide OR jump over the sliding part...one path ideally is faster but harder!

I ramble a bit in there but I think those sorts of notes can be really helpful - rather than thinking about the level graphically right off the bat, maybe start taking some mechanical notes about cool arrangements of bad guys, and from there move on to doodling in a sketchbook or level editor and start playtesting...THEN start slapping some gfx on there.

[Corpus]

What's a "kickstop"?

[Derek]

Level design was something I struggled with initially with Aquaria, because most of the decisions to be made there weren't obviously right or wrong when I started.  I had some basic ideas (e.g. "this mechanic needs to be introduced here" or "this part needs to feel X"), but everything in-between is kind of up in the air.

In the end, what worked best for me was to just start building and then iterate over and over again.  Each time I'd have a better idea of what worked and what I wanted.  It's just a lot easier to work over something that's there, even if it's wrong, then to work over a blank canvas.  So I try to get to the point where there's something there as quickly as possible.

Regarding what I started with, it wasn't completely random.  Usually I'd have some kind of basic idea in mind, like "open area" or "lots of little nooks" or "central room that branches out into smaller rooms."  Or, if there was a particular emotion I needed to convey, I went with that.

(Note, Alec's awesome tools made it a lot easier for me to work like this.)

But yeah, all of these suggestions are good ones.  Whatever suits you the best, you should go with it.

[___]

When I started the Bonesaw game's levels, I had no clue what to do and I just went for it.  The levels then start to get more and more thought out as the game progresses.  Each world introduced some new platforming and enemies, so I thought of interesting uses for each and puzzles that revolved around them.  Then I just had to think of ways to use each of these cool little ideas.  Usually for a lot of my ideas, some really basic sketches are all I needed.

I think one of the best ways to go about level design is to just start with some very basic challenges, and then creating more complicated challenges by combining the earlier basic ones.  If you havent, I suggest playing some of the Half Life 2 games and Portal with the developer commentary.  They really think their stuff out.

[cgmonkey]

Something I didn't see being mentioned here is the psychological side of level design, how to create an experience for the player more than a "precise course of perfectly measured distances and moving platforms". Of course, I do most of this fluff for myself but in some cases it really sparks some great game mechanic ideas.

A player will automatically feel progression if you him move up or change color from dark to light. If it's vice versa (going form light to dark or down) the player tends to feel less adventorous and more cautious.

Contrast is a great way of making the player alert and a great tool to grab his attention. Remember that the player always want to be led but the less he feels being led, the more fluid the experience feels.

I usually setup a small brief for each level I design, which tend to vary the gameplay, a couple of examples are:

"Create 3 badass scenarios within one level.. then connect them"
"Don't let the player stop and think"
"Let the player explore the surroundings"

From this brief I can usually go ahead and explore the psychological aspect of the level. If I want the player to explore, I want him to become catious -- I would generate a slower gameplay level, making him decend down, darken the colors, make the colors more cold, trying to prepare the player.

Making varied (but consistent) level design is so important to a game, almost more so than the controls.

The most important thing of it all: Always add the graphical fluff last.

Hmm.. There are soo much more stuff but I think it's worth exploring by yourself! I hope my broken english and incohesive thoughts made any sense!^^

[Melly]

I like the idea of using color and contrast. I know what it feels like to be thrown into a much darker level in a game and feeling more cautious.

[cgmonkey]

I like the idea of using color and contrast. I know what it feels like to be thrown into a much darker level in a game and feeling more cautious.

Man I still remember as a kid, the first time you got dropped down in the second level of mario bros.. I was scared shitless!