A question to level designers

[607]

Hi, it's me again!

When you get an idea for a level you want to make, how do you write it down?
Do you design levels on paper, in the editor or in a paint-like program?
Because I thought that when you first get an idea for a level, it'd be good to first quickly draw it out, without needing to design a fully working level including everything right away.

Also, do you think it'd be bad if I already start designing levels? I mean, I've only solved everything up to Taxing 16 from the original Lemmings, I haven't gotten very far in ONML and I've never played custom levels, so I don't have too much experience. I feel like I could already start to design Fun and Tricky-like levels, but I'd like your thoughts on it.

[namida]

As long as you understand the general concepts of the game, and have at least some experience in playing it, I'd say that's enough. You're probably not going to create the next "No added colors or Lemmings"* when you've just started designing levels, regardless of how much playing experience you have - but that doesn't mean you can't create decent levels; plus the best way to learn something is to practice it.

* "No added colors or Lemmings" is Mayhem 20 of the original game, and according to voteoffs and past discussions here, the most popular level in Orig.

In terms of what I do with ideas - generally, there's only two processes that tend to happen:
A) Level appears in my head. Level is transferred roughly to the editor, then touched up until it works.
B) Editor appears on my screen. Pieces are randomly thrown together until something resembling a level appears. From this, the level is touched up until it becomes something good.
(Of course, some levels are a combination of both of these. Often the main trick will come from method A, while the rest of the level - the context in which the main trick is hidden, especially if it's some kind of crowd control trick - is devised using method B.)

As counterlogical as it may seem, in my experience the better levels I've made tend to come from Method B, including my (so far) most popular level, "Panic Attack" (Genius 8) from Lemmings Plus II. This doesn't mean that method A should be discredited altogether, of course - some great levels have come from that too, such as "An Unexpected Journey" (Genius 18) also from LPII.

A third, but rarer method, that I sometimes use (and could be considered a variant of Method B) is that I'll come up with a general concept for a level, but this will be quite an abstract concept and the actualization of the idea is pretty much "slap together something that works" still. A great example of this is "To the end!" (Psycho 28) from Lemmings Plus I; I had the concept (see below) beforehand, but the actual level it existed in was made up as I went.

(I've hidden the concept here behind spoiler tags, since you mentioned elsewhere that you planned to play LPI at some point soon, and so you might not want to be spoiled on this.)

Spoiler

The concept here was to take the common (particularly in the Master System version) idea of "a level with 20 of most skills, but one or two that there's none of", and make that "none" skill builders, in a very complex worker lemming-like level where builders are very valuable (but it is possible without them).

[mobius]

I don't usually write things down to try and help my memory. So far that hasn't work at all so I've begun writing things down recently.
But when making levels I usually just start making the level right away. Or I'll open a level from the original game and modify it until it's unrecognizable; to give me ideas.

Custom Levels are quite a bit harder than the game so I recommend finishing the games first.
I suggest first playing packs like MazuLems or namida's plus series, since they start out with easy levels. DoveLems also starts out easy, and my pack Mobilems.

If/when you want more challenges I recommend packs by Dodochacalo, Piewu, BulletRide, Insane Steve, Clam. There's cheapo levels by Proxima and Ben Bryant that I personally like. [In other words; if you're still a beginner avoid these packs because they'll seem impossibly hard]

[namida]

Hm... I wouldn't recommend MazuLems to someone who hasn't completed the original game yet. Most of the levels are at least Taxing-difficulty.

Lemmings Plus I is a good one for new players; apart from a couple of levels (such as Wimpy 12, Medi 6, Medi 16, Medi 21) it doesn't really get hard until the 4th rank (out of 5). GeoffLems, cLemmings and GigaLems are all good options too.

If you're planning on using NeoLemmix for your levels, Lemmings Plus Omega may be a good one to play, as it actually makes use of NeoLemmix features (the new skills / objects / etc). Lemmings Plus III also makes use of them apart from the skills, but it's not a pack I'd recommend to beginners as even the first rank can be pretty tricky at times. As far as the gimmicks go, there isn't much "easy" stuff using them; perhaps the earlier levels in their dedicated ranks in the LPII / LPIII Bonus Packs may help, or else you could also check out DynaLem's "Unused Gimmicks" pack for those that aren't used in Lemmings Plus. (And of course, if it's the zombie gimmick you're interested in, look no further than Doomsday Lemmings!)

[geoo]

I started editing levels before completing the game too. They weren't all exactly masterpieces (though at least Clam seems quite fond of them), but I definitely gained experience from creating them.

I usually build levels around a main idea, wherever that may come from. I usually sketch out a rough level layout on paper, though sometimes, when it's easy enough to turn the idea into a level, I open the editor right away. Then comes moving terrain around until the level works, and then comes backroute testing.
As for coming up with those ideas, for me it usually happens when playing other people's levels. I come up with some solution idea that doesn't work in the given level, but makes a good base for a new level. In such a case I save a replay, write a note with a sketch or a level skeleton so I'll remember, or move to the editor right away (rather rare though).

If you haven't edited levels yet, then you also have a choice in which level editor you choose.
There's Lemmix, Lix (the levels you build can only be played in Lix), and jLevelBuilder. Give each of them a whirl and see what you're most comfortable with.

[607]

Another thing is that I am afraid I'll be using the same ideas as other people. I mean, it's a puzzle game, after all, and if I design a puzzle, there is a chance that somebody else has designed a puzzle that it very closely resembles. And I have the idea of "I shouldn't play too many levels, or else I will have seen every good level idea already, and I would be copying them, instead of having the same idea by accident."

[namida]

New ideas show up more often than you think. Especially if you use NeoLemmix or Lix, as they have new skills and other new features that haven't been explored too far yet. Even if one of your ideas isn't new, you're most likely combining it with other ideas in a different arrangement than previous levels have, or hiding it or setting it up in a different way, etc.

Some things I would recommend avoiding though, as they're kinda annoying and have been done to death (and I myself have done some of these in the past xD) - excessive use of hidden traps; invisible/hidden exits; and the "I am A.T." trick*. Of course, if you can come up with an original way to use those, then by all means do so - that's a different matter if they're used in a original way.


"I am A.T." is a level from the Genesis version of Lemmings; the main trick of it has been done to death in custom levels. It involves...

Spoiler

...placing a line of diggers one next to each other, so that they end up creating a somewhat diagonal slope and/or over their combined width destroy a whole wall.

[Nepster]

Don't worry about already existing levels! It is rather rare that one discovers a completely new trick (if you use only the standard features of the original Lemmings).
If you look at the big custom level packs, you will find many levels using similar tricks. And they are still worth playing! Each person has a different style in creating levels and therefore the levels will still differ quite a bit in how the trick is incorporated into the level. My own reaction when seeing one trick another time: "OK, we have this trick again. But this was a very neat way to force the player to use the trick!"
In one point I have to contradict namida: Although hidden traps and invisible exist only annoy the player, do not worry about making another level with the "I am A.T."-trick. There certainly exist many levels featuring a wide variety of variations of this trick. But being rather new to the world of custom levels, you might not know all of them. So it is way better to create such a level and be told that it resembles level abc of xyz, than to keep your idea for you and realize much later (or never at all) that it actually contained something new.

On your original question: I create my levels essentially in the same way as geoo.
If you try out namida's process B), then it is very easy (at least for me) to create levels that rely very much on builders (which can become rather boring soon). So it is probably not a good idea to use complete randomness to put terrain pieces in the level, but to constantly think of ways to overcome the obstacles with only few builders.

And two final pieces of advice:
- Take your time when creating levels! I am probably one of the slowest level creators here, but for a one-screen level, I need about 4-6 hours until the first playable version appears on my screen and about the same amount of time modifying it (to remove non-intended solutions and to ease the execution of the intended ones).
- Do not worry if your first levels are not as good as others in the database. Out of the first 20 levels I created, I considered only two good enough to make them available to others.

[namida]

So it is way better to create such a level and be told that it resembles level abc of xyz, than to keep your idea for you and realize much later (or never at all) that it actually contained something new.

This is an extremely good point.

[Proxima]

The flip side is that existing levels can be a great source of inspiration. Don't copy them exactly, but think about how you can get new ideas from them. Perhaps when you were solving a level, you spent a long time on a really neat idea that almost works -- in your own level, that could be the real solution, and the solution to the source level becomes the red herring. Perhaps you find an interesting solution to a 20-of-everything level -- then you can reduce the skills enough to enforce it, or add obstacles that block off certain routes.

This is how I made "You Only Get One Bash At It" -- I can't discuss it without a certain amount of spoilerage, so I'll hide the rest (but it's not a complete spoiler of the solution).

Spoiler

I started from the solution of mobius' "Exit stage right" and flipped it around so that the solution to mobius' level was the plausible red herring in my level.

The level that resulted turned out to be the hardest, and perhaps also the best, level I've ever made. The fact that it was closely derived from an existing level doesn't hurt it at all.

[Colorful Arty]

I will only write down ideas for levels on paper if I already have an idea in my head and i want to visualize how it will look. (I have done this in class a few times when the lecture got boring) However, to each their own. I probably only drew about five or six levels that went into my SubLems pack.

I will say, the more you play Lemmings (especially custom packs) the better your levels will be. That is because there are so many tricks you can pull off in Lemmings that the original designers probably never even considered, and you can pick up on some really neat architecture tricks. I would encourage you to beat Lemmings and maybe a custom pack (ONML was not terribly great at puzzles in my humble opinion) but it is not bad that you are making custom levels now. It's just that as you learn more about lemmings, you will be able to create more intricate and exciting levels with fun and new solutions.  I speak of my own levels as an example, my more recent levels tend to be far better than my earlier levels.

[grams88]

Go for it guys, it's amazing how many different type of levels you can make like for example the hero one where one lemming does most of the work, We probably shouldn't overdo this one too much. (I like my occassional hero level) There are those levels where you have to be precise. There are some where you have to get your lemmings to high ground, I quite like this one but I know a lot you know the trick where you can get a lemmings to go up very fast.

Sometimes we might make a level that overwhelms us with so many ways to get to the exit.