14 users online: Anas, Children's Digest 1950-2009, gui, HammerBrother, KAMESUTRA, LordElixs,  Major Flare, Morph Moth, qbear, Sokobansolver, Soul, Soup64, Sthian Felipe, thekiller678 - Guests: 119 - Bots: 150
Users: 55,589 (2,472 active)
Latest user: Soup64

What are people's process on designing levels?

How do people design levels? Not like get ideas or anything but planning out and creating a whole level. I know I have plenty of ideas but whenever I get started my mind either ends up a blank and I end up dropping it or it ends up not very fun to play. Maybe I don't have the creative drive to do it, idk

My main recommendation is, don't overdesign your level early on; you can start by creating the majority of it with just simple tiles like cement blocks. Later, once you've settled on the general structure of your level, you can go back todraw over it with fancier stuff. It'll be faster and easier for you to adjust sections that way.

Also, don't get too bogged down in generating ideas immediately. As you design things, play through whatever you got a few times to check if the direction you're going in actually feels good. It can help narrow your scope by either telling you "these ideas work well, maybe I can do more with them", or "these ideas don't seem to mesh, I'll try something else".

Past that, my only recommendation is "experiment". Don't expect every idea to flash into your head immediately; a lot of the specifics of your level will come to you as you design it. Throw down things even if you're not sure they work, and then see if you can adjust them to the point that they do.

Professional frame-by-frame time wizard. YouTube - Twitter - SMW Glitch List - SMW Randomizer
Mind if I move this to SMW Hacking Discussion? (I ask as I move it) You are asking for help, but this is an interesting topic I think a lot of people could give insight on, and in this help forum threads tend to become buried after they're answered.
Originally posted by Koopster
Mind if I move this to SMW Hacking Discussion? (I ask as I move it) You are asking for help, but this is an interesting topic I think a lot of people could give insight on, and in this help forum threads tend to become buried after they're answered.

Yeah I think that'd be better. This forum looks more like technical stuff.
i just watch whatever streamer doing Lunar Magic and steal all the ideas.

I thought everybody did the same.. no?!

For the real answer.. idk.. i just try to define a method and have a somewhat techinical approach.. its hard to explain.. its easier to just copy someone else
I go to the music portal and find a song that really sticks out to me. I think music is one of if not the most important aspect of atmosphere.
After I get the song I like to see what kind of setting it makes me imagine. I always try to think of at the very least a color and a location/theme. An example would be something like "Purple, ghost house."
Then I try to think of a gimmick for the level that would fit the pacing of the song.

I try to take the actual level design one jump at a time, the first obstacle is always the hardest for me so I just kinda throw stuff down until it sticks. After that I just try to make obstacles that flow and I find fun to play. If I find myself stumped and unable to come up with anything that feels good I take a break for a few hours. Usually when I come back I'm ready and able to slap some stuff down that seems right.
If I come back and I still can't seem to find anything that works I go and work on/start a new level. Obviously if you're not working on a full hack and you're just making one level that wouldn't really work.

I'd really recommend playing some other hacks and paying close attention to their level design choices, not necessarily stealing from them but pay attention to how they connect different parts of the level together in a way that flows and doesn't seem separated. Also pay attention to what you're finding fun about the level, even pay attention to what you don't like about it.

I haven't made any non-kaizo levels so my advice is gonna be pretty skewed and it probably work for standard levels. Sorry if this is a little long winded I just wanted to make sure I could share as much of what's helped me as possible.
My process is something like:

I create some screens (6-7), add some FG, some obstacles and sprites, test the thing several times, modify sprites position and adjust FG position. Then, I get rid off the entire thing, start again (no jokes). I think most creators have the same experience, and then, they ended up fleshing the ideas onto something better than the very first sketch.
The process on designing levels varies from people to people. I play with objects and sprites and then I see if I am satisfied with whatever setups I came with, doesn't matter how many screens I did. Just do it to your heart's content, I guess, there's no "ruleset" for that.
Well my process usually goes like this, be prepared cause this will be long to read lol.

Here are my preparation steps:

1)I dont think of the idea of my levels in the heat of the moment, they just randomly pop in my mind at any time of the day, so I have the habit of writing them down on a paper or a notepad for remembering them later when I open Lunar Magic.

2)I make sure I downloaded everything I'll probably need so then I can start designing the level, this usually includes: custom sprites,blocks,music,graphics,etc

3)At the moment that I start my level I usually put some masses of ground first and then some of the sprites/objects that I pretend to use together in the ideia that I previously had of the level to see how they work together.

4)If some objects/ideias dont match well enough together I try to see which other combinations could work, and which other ways I can try to execute the ideia.
When I start to design my levels, I usually follow the following rules that I created for myself:

1)Each level must have its own ideia/mechanic to set them apart from others and keep them memorable.

2) Each level ideia must follow some steps:
-introduction in a safe environment,
-development of the challenge,
-twist: its basicaly a ramping up of the dificulty it consists in taking a new aproach to the ideia of the level, it can be mixing it together with other obstacles/mechanics introduced in the hack, using the ideia in a diferent take like a vertical level or a chase/autoscroll section for example, or some times instead of that I just introduce a second ideia that goes through the 2 same previous steps above, also some levels may have more than one twist depending of its lenght.
-A conclusion of the ideia to end the level.

3)Sometimes especially when I'm designing long levels I like to put some other mechanics that is unrelated to that level especific theme, these mechanics are usually more spreadly used through the hack on smaller section, their purpose is to break the pacing and add variety and keep the level's main ideias from being repetitive.

4)My levels usually have 0D-12 screens to keep the level from being too long and overstaying their welcome, keeping them from becoming boring.

5)I usually develop first the route of the main path the player will take to clear the level, then I go back and modify the layout a bit to create and think of good places to hide secrets, it can be pipe entrances, secret passages, places hard to reach, some sections that are usually obscured by the camera etc. and I reward the player for reaching these areas by adding collectibles, 1ups,coins,etc.
These secret passages/subareas can also provides smaller optional challenges/minigamesonus areas to the player.
And the reward for finding them must be consistent with the dificulty of finding/completing them.

6)I usually try to put at least 1 1up mushroom in the main route of at least 1 between 3 levels, so the less-experienced players can have some extra tries in that levels

7)If the dragon coins are a important collectible in the hack i usually put them closer all before or after the midway point so the player can collect them in one go even if he dies.

8)The more obscure secrets usually have some indication to help the player find them, it can be a trail of coins, a detail on the scenery,etc.

9)Some times I break these rules on purpose for a single level, it can add some tension in that particular level and subvert the expectations of the player keeping the levels from being formulaic.
Back as a kid I used to actually draw out my levels on paper and then re-create them as best as I could in LM. Nowadays, though, I don't really have a methodical process for it. I think of a gimmick I want to base the level around - maybe it's an environmental hazard, a specific enemy, a time limit, a race, who knows what. Then I just place down flat ground and slopes in a shape I think fits the gimmick. I play through each slight revision and see if I consider it fun. I suppose if I've learned anything in over a decade of using LM, it's that if you can just run to the right and occasionally jump through every level in your game, with almost no effort and nothing actually challenging you and making you do something else, it's not a very good game.
I have seen and played so much SMW that sometimes "levels" play in my head when there's nothing else in there, usually at work since it's such a muscle memory job that it frees my mind. It usually starts with "what if (ridiculous idea?)" Or "this is where (something) should happen". If it's a persistent thought, I go ahead and make it. I always have to set up the approximate palette first. I tend to start the building part by putting any basic contact tiles I need (wall, ground, munchers) somewhere in the level so I can easily repaste them and I put the sprites down that I need for that scenario. Seeing those tiles also helps me get a fairly accurate idea of how the rest of the level will appear and that is a bit motivating. I try to build obstacles around the purpose I have behind the sprite I want to set rather than landscaping first then setting up guards later so that way I can ensure a meaning in each part I build. I don't add fill dirt or decorations until I consider it just about done because if I don't interact with them, it's just extra stuff to move. There sometimes comes a point where I really want the background to be there to remind me of what I am making, so I will just stop designing and play Map16 surgeon for a pretty picture before I can carry on.

After I structure a piece, I usually test it out right away. Elevation of surroundings has to be right; I don't like important places and things being off-camera. Foes have to do an exact thing in an exact place at an exact time. The distance between it and the previous major component must be just right. The setup's screen must be scenic without carry-over from components that don't look good next to it when it is finished. I like symmetry and evenness where it makes sense, but I will drop it if it gets in the way of a smooth gameplay experience. Then I X out of the emulator, shift a thing or two, try again until it's right. I sometimes go slightly ahead with building but I rarely complete a room before refining each segment in order. The brightness and hues might get adjusted periodically since a Lunar Magic still and in motion are sometimes very different. When I test a "finished" level, I want to revise pieces that don't quite do their job right in full context but by that point, whole redesign shouldn't be necessary. I believe I spend 5x more time testing and critically analyzing my levels than I do making them... and that's just vanilla.

When I do multi-room levels, I don't necessarily want to keep the same exact gameplay elements of the first, but I want it to make it feel like it belongs in the same environment and make it either rougher or suddenly relieving as a reward before the end of the level. I would think of Chocolate Secret or any of the Koopalings' castles to call my ideal level progression.

I put on music to help me focus and help set a mood and energy to feed on, and that mood and energy should reflect in the setting I establish and the gameplay. I usually end up using that song in the level. If I don't have the design ideas, I would rather sleep on it for a while than fluff it with filler. I refuse to compensate just to "get it done," because I know I will not like the outcome, and getting rid of the bad would be harder than starting from scratch. I will end a room if it has served its purpose and doesn't need anything else. My last level was made over the course of 20 days with daily work of 1-4 hours per, as a result, before I was happy enough with it to seek public feedback.

I have tried planning levels on paper, but I find drafting them right in Lunar Magic to jump around on seconds later works better for me... not to "meme" but I have trouble making my pre-drafted levels have "soul" versus stuff that floats in my head. I have never been good at following a script, and what I blurt out tends to be much better. The exception: I would rather draft an overworld on paper than wing it any day.

I am terrible at organizing projects of ambition since I really want all the aesthetics and mechanics to be complete before I start any level design. Therefore, I spend more time drawing and porting than any level-making unless it's vanilla. I want to be more serious about finishing something other than contest levels, but I at least want to meet that quality every time I make a level.

Just look above you...
If it's something that can be stopped, then just try to stop it!
I've only designed kaizo levels. However, this topic interests me, so I thought I'd put my thoughts here too.

As I'm working on my first hack, I have discovered my design process isn't exactly the same every time I design a level -- although, it always starts with one of three questions: "How do I want the player to move?"; "How would I describe the space of the level?"; or "How does the the level sound?". Maybe I want the level to instill a sense of urgency in the player, or I want the level to look and feel claustrophobic. Heck, maybe I found a song in the music section here and thought of an entire level (which is something that has happened to me many times). Another way I come up with ideas is from other games that aren't SMW, sometimes not even Mario. I'm influenced by many different games that aren't even platformers, and yet the concepts in those games influenced my design (whether that be aesthetics or gimmicks).

After the initial idea, I think about the sprites I want to use. I, personally, am very particular about limiting the types of sprites I use. This is to keep everything in line with the theme of the level, as long as I can be creative with the obstacles and setups -- akin to a painter working on a piece with limited colors, or a musician writing a piece with a particular chord progression. For me, this information is enough to draft up a level -- or mess around in Lunar Magic until I create a set of obstacles that match what I'm going for. If it's fun and consistent, I keep it and revise it over time; if not, I scrap the level and try again.

Of course, the process of designing a level is much like any other creative process in that it differs from being to being. What truly matters is that you're patient with yourself and that you're having fun creating. After all, anything you create is something that comes from within yourself.
It's said that there are two types of creators. There are the architects, of which this thread has many. They are the ones who don't lay a single tile until they've got an idea, a theme, all the tools, and a draft or sketch. The other type is the gardener, who plants a few seeds, sees what starts growing, and doesn't know, exactly, what will result until it does grow a little. One is proactive, the other is reactive. I count myself among the gardeners.

With a few exceptions, I don't plan anything. Maybe, I go in with some vague idea in my mind, but nothing really precedes the initial laying of objects. In fact, I tend to start a new level by placing some terrain and objects kind of at random, seeing if the arrangement might lead to anything unique, and then adding more objects and maybe an enemy or two.

Beyond that, I build a level in sections, a screen or two at a time. Then, I play test the section and see what I've got. I'll make changes based either on necessity or inspiration, and when it has a flow, I'll continue on, repeating that process until it feels like I've built a level that works, mechanically.

This is when I will start to refine the level. When it is barebones complete, I can play test the whole thing and get an idea of where the level flows well (and where it doesn't), and then I can make changes to improve the overall experience. This part of the process is complete when I feel that the rough parts are ironed out, that there are no undue frustrations or bullshit difficulties.

Now, I will draw graphics and start the polishing process. Oftentimes, having new graphics will necessitate (or inspire) further changes. In any case, I tend to draw detailed graphics and use most of the available graphical space, so sometimes, I might end up adding a lot of new objects overlaying the originals. Then, more testing, to see how it all looks. I draw graphics as organically as I design levels, so I will often do a lot of redrawing and revisions. I will do the same with palletes and even music.

Then, since my hack incorporates text, I write the story pieces for the level, use Map16 to create the tiles, and place them throughout the level in appropriate places.

For me, construction of a single level can be extremely involved. This image will show you how one of my levels evolved over 7 years. It is a big gif, so I'll just link to it: the evolution of 'the dying light'.

Chapter two: Land of No Shame
Coming January 2022
I like all of your responses and I've been taking them as advice on what I should do. I've been trying the idea of "storyboarding", where I draw a simple sketch of what the level looks like (Just the ground, no enemies or anything since that comes later), then recreate it in Lunar Magic so I get an idea of what the level looks like. After that, just build and tweak upon it. It worked well for me so far.

I also found out that working on levels while other people are watching actually works pretty well. People tend to give ideas and where you should place things, incase you might end up hitting a creative dead-end.
Allot in consideration to take is to make them have some sort of challenge while still being fun depending on the theme of the level maybe a maze or gimmicks added.

You are challenged by Champion Daisy!