Level Design

Discussion and analysis of graphics, story, levels, and so on.
User avatar
Freeyorp101
Posts: 159
Joined: Thu Nov 24, 2005 2:12 am
Location: New Zealand

Post by Freeyorp101 »

The problem with forground secrets is that if you don't foregound the surrounding solids, you can see keen poke his head or feet out of the passage, which isn't a good look. and if you have rounded corners, you need to mask them, which uses up tile space heaps.

but yes, foreground secrets are fun.
User avatar
XkyRauh
Posts: 1114
Joined: Sun Aug 31, 2003 9:14 pm
Location: San Diego, California

Post by XkyRauh »

Benvolio wrote: The best example of this I can think of is in level 5 of Cosmo's Cosmic Adventures 1, in which the place where the US lander with the burger is, is in a place I usually would ignore, and that psychologically it's probably the place people are least likely to look.
Heh--I must play games quite differently than you--I didn't even realize that the burger was intended to be 'hidden.' I thought it was right out in the open, waiting to be grabbed. On the other hand, later episodes' levels with their friggin' "hit your head on these tiny little colored switches located in the oddest places ever" levels take me 20+ minutes to complete. :(

The tricky part about placing secrets in Keen1-3 levels is that it's difficult to reward the player with lives in place of score, the way you can in Keen4-6. In Keen 4-6, a secret can contain the Lifewater/Vitalin/Vivas, or a plain ol' Extra Life. Those are ways to reward the player without necessarily boosting their score. But in 1-3, whether or not the player finds a secret means a big difference in their score.

That said, is there a way one might rationalize the number of points in a secret area? For example, say a given level is small-ish in size and contains about 16000 points, not counting secret areas. What amount of points is appropriate for a secret area? A quarter of that--4000? That's not even a single teddy bear--but collecting all the points and the secret would offer an extra life. Does the level provide enough challenge to balance that?

On larger levels, I've always felt that it's less important to gauge the total number of points in a level--particularly, if the level is considered "challenging." You'd be surprised how often a single danger at the start of a level can deplete a player's stored up lives--so offering a reward after the initial challenge can be a good thing.

I guess what that comes down to, more, is... how do you classify your levels, when you create them? I tend to create 3 types of levels--"throwaway," "challenging," and "key." Key levels are for ship parts, Tantalus Rays, etc... Challenging levels are optional but large, and throwaway levels are like the Martian domes in Keen1. :) That helps me decide how to distribute points and secret areas.
User avatar
levellord
Crazy pAtChEr
Posts: 1401
Joined: Thu Nov 20, 2003 11:35 pm
Location: NewZealand
Contact:

Post by levellord »

I define my levels into two types, vital, and non vital. Subclases of these are big/small and empty/full.

The tricky part about placing secrets in Keen1-3 levels is that it's difficult to reward the player with lives in place of score, the way you can in Keen4-6. In Keen 4-6, a secret can contain the Lifewater/Vitalin/Vivas, or a plain ol' Extra Life. Those are ways to reward the player without necessarily boosting their score. But in 1-3, whether or not the player finds a secret means a big difference in their score.
Which is why we have patches for a.) Changing the point scores/points needed for a life and b.) Making points give Keen a life.
User avatar
Freeyorp101
Posts: 159
Joined: Thu Nov 24, 2005 2:12 am
Location: New Zealand

Post by Freeyorp101 »

^ or pogo for an extra life...
User avatar
CommanderSpleen
Posts: 1017
Joined: Sun Aug 31, 2003 12:11 pm
Location: The Land of Sparkly Things
Contact:

Post by CommanderSpleen »

Level design is possibly one of my weakest points when it comes to Keen modding. I mostly just slapped them together for the Monky Business beta. Sure, there was some slight structure to them, but there wasn't really any purpose to each level.

One basic rule I did try sticking with, however, was a form of gravity for the slabs of sparkly platforms that form the bulk of most levels. For a couple of levels I also had a basic concept behind it (ie. the flashing pyramid and the block mazes).

Of course, I had to factor in the doors at the start of the levels because of the way the keycard patch worked, but now that we can get rid of keycards when the player dies this is no longer necessary.

Those designs were far too evil in many instances, too. I wonder how many people have actually completed the mod. I want to focus more on giving the levels a sense of meaning, like the way Keen4-6 have levels that fit in with their appearance on the map.

There aren't many 'unnecessary' areas in my levels, nor secrets, and you need to complete pretty much all of the levels, and the map level is stiflingly linear.

I've never really been a fan of 'optional' levels, especially when the high score table is rendered meaningless by the way the lives and score systems work together (I'm still hoping to find a patch to implement some sort of level-context score counter).

An idea I'm pondering at the moment is a multiple-paths approach, where the player is given a choice as to which (locked) area of the map to explore next, as well as which levels to complete in order to open the next area. It will require some careful planning and maybe some more experimental patches, but it should improve the replay value.

The idea also has potential for level design. Do I tackle the puzzle that activates this bridge to cross these spike pits, or search for the keycard that unlocks the door leading to an alternative route? And which one of these exits to I want to go for? Argh! *cowers in corner, muttering*

Obviously it'll need some balancing, and some trivial areas that are just there for the sake of it will show up.

I tend to make my levels big and then try to work on too broad a scale (making the outline first, without any overall planning). The complex platform tiles (which have inside and outside corners) also make this a lot more difficult. Maybe I should make the platform designs simpler, using these details only in special circumstances.

The various levels I've worked on since releasing the beta have looked more like caves than anything. I sort of draw chunks of a single-tile platform, carve a bunch of caverns and link them up, then overlay the walls with the platform tiles that take adjacency into account. This might make for some interesting levels eventually, but at the moment I should try something different.
Xky wrote:I find it much easier to design levels based on the idea of "scenes" or "rooms." Work with an area that is about one screen in size, and make it look snazzy enough to deserve a screenshot. Then, move to another part of the level--someplace hopefully at a different height, maybe even a different size, and do the same thing. Once you've got two scenes you like, figure a way to weave them together. Slope a hill, make a series of platforms, dig a trench, put in a river, sprinkle enemies, whatever it takes... larger levels need more defining scenes, while smaller levels can focus with just "Point A" and "Point B."
That is a brilliant idea. I must employ it.

Anyway, there's a rant for you to decypher for now. If only more modders were available at once on MSN or something, I'm sure we could get much more interesting discussion flowing.

Grr. I wanna get my tileset and Vortlev done so I can get stuck back into some modding again.

Hmm, an idea. A thread in which we grab a couple of levels from different mods and discuss what makes them interesting, how they could be better, what difficulties each of us has actually playing them, etc, etc. But which levels/mods to start with?
User avatar
XkyRauh
Posts: 1114
Joined: Sun Aug 31, 2003 9:14 pm
Location: San Diego, California

levels!

Post by XkyRauh »

CommanderSpleen wrote:Hmm, an idea. A thread in which we grab a couple of levels from different mods and discuss what makes them interesting, how they could be better, what difficulties each of us has actually playing them, etc, etc. But which levels/mods to start with?
I took a stab at the original Keen2, back when I was working XkyKeen2, in this thread, and more specifically in this post.

I think it'd be fun to tear apart an older mod--XkyKeen1 is full of crap I'm eager to improve upon, and I'd love more specific feedback about what works and what doesn't. Why not start chronologically, with Yorphius II or XkyKeen1? :)

Edited to reply to LevelLord's post below: ShadowKeen1 or PatchworkMod1! Yeah!
Last edited by XkyRauh on Mon May 01, 2006 3:38 pm, edited 1 time in total.
User avatar
levellord
Crazy pAtChEr
Posts: 1401
Joined: Thu Nov 20, 2003 11:35 pm
Location: NewZealand
Contact:

Post by levellord »

Nobody's commenting on my early work there was so much wrong with it you'd never stop talking, although I'd like to see the first patchwork mod discussed, I personally liked the levels in it a lot.
Stealthy71088
Posts: 583
Joined: Thu Mar 17, 2005 11:54 pm
Location: NY

Post by Stealthy71088 »

I'm all for this thread. It'll probably help us all improve our modding skills a lot. Patchwork mod is made by a bunch of different people, so there will probably be a lot of different styles. At the same time, going through the mod classics would probably be fun too. If I have the time, (which i seriously doubt- unless it became a group project) perhaps I could type up whatever we write about into a nice little document or something to make it easier to reference.

Spleen, when do you think Monky Business will be finished, if ever? The beta was cool, although incredibly difficult in a few places. (I still haven't beaten it.)
Benvolio
Posts: 228
Joined: Sun Aug 29, 2004 4:44 pm
Location: Ireland
Contact:

Post by Benvolio »

Yeah, a new topic on the subject, or even a whole new forum section.

Xtraverse's first three mods are also some of the best from early days. (I hope it's still early days now, there's not much recent mods).
I tend to create 3 types of levels--"throwaway," "challenging," and "key."
I tend to create one type of level--- "challenging". Then, a week before the mod is finished, I decide which levels to make "key". You could say all my levels are "throwaway", but that's just negativity.
Level design is possibly one of my weakest points when it comes to Keen modding.
Not wanting to be to critical or offensive, I would agree. But the great graphics make up for it... any sign of those mods shown in the previews of Orbkeen-Beta?
User avatar
CommanderSpleen
Posts: 1017
Joined: Sun Aug 31, 2003 12:11 pm
Location: The Land of Sparkly Things
Contact:

Post by CommanderSpleen »

Stealthy wrote:Spleen, when do you think Monky Business will be finished, if ever? The beta was cool, although incredibly difficult in a few places. (I still haven't beaten it.)
I can't really say yet. Probably not this year, but we shall see. Once I get Vortlev functioning as I want, I'll get stuck back into it for sure. However, there's a lot of work to do in making the atmosphere more immersive. I intend to use up as much of the tileset as I can manage, and I'm barely half way so far.

I'm also considering a small tile-editing utility, but the ability to instantly reload the graphics in Vortlev (currently assigned to T) should be sufficient for the moment--at least I won't have to keep loading up the mod just to see what one new tile looks like.
Xky wrote:I took a stab at the original Keen2, back when I was working XkyKeen2
Indeed, I do recall that post. That sort of analysis applied to a mod would be extremely handy. I'd really like to get some more thorough feedback about Monky Business. However, I'm all for starting with Yorphius, XkyKeen1 or one of LL's.

An extension of this concept could be to redesign some of the levels taking into account the problems and suggested improvements.
Benvolio wrote:or even a whole new forum section.
Now you're talking. We should probably start small and see where the idea takes us, though. If it works well, a sub-forum would be awesome.
Post Reply