This is for discussing level design... real design... and discussing level 'gimmicks' and features that are introduced... and when they should be introduced. For example: water, water's effects, dynamite and boulder puzzles... etc... It should be better to introduce water early on... and later on, there should be a water world. And water levels early on should use it sparingly. I've started a list of the tiles we are definitely going to use, and I've arranged them in the order I think they should be introduced..

Fragile, Indestructible, Steel, and Wood Crates: First level
Arrow, Dynamite (launcher): Early on

So, in the level design department, we'll need to start with a list of level elements and when they will be introduced, then level gimmicks. After the programming is done, level design (secret areas, traps, the level itself, the look of the cave)  follows, then finally the worlds. Then we should plan out the levels before making them, and then we'll make the levels... which will then be edited... and stuff. So, even though Livio didn't give me the go-ahead for this, who's interested in Level Design? ( Livio's classmates are included)

(Is there any sort of list of all the tiles we plan on having?  I'm not sure what all we have planned and there may be something I have forgotten and may change the ideas behind my thinking.)

I am very interested in level design, and hope I can do something of the sort.  I doubt I will be better than most other people, though.

For introduction, I believe that we can't really say just how fast we are going to be teaching things until we actually have the ability to play what we are teaching.  Four different categories of crates and multiple kinds of crates can lead to many different combinations.  While I don't think we need to teach them everything in the tutorial levels, we still need to make sure they have a good understanding of the kinds of traps that will arise in more complex levels.  As in lots of internet games, the player will be learning something new every level.  How much will we introduce to the player, though.  We can't have the tutorials too long, as then the play might be overwhelmed, or we can't have them to short, or else we will have a list of tutorials that is pages long.  Personally I think we need to introduce treasures first thing before anything else, as they are a very important part in the game.  After we do that, wooden crates are next, and then arrows.  At first when I was thinking about it, I thought we should introduce the steel after the wood, but then I realized that the player would see it but not have any way to destroy it, and they might get this feeling that it can't be destroyed.  So once we introduce arrows, we move on to steel, and then dynamite, which I believe to be more complex than arrows.  After that boulders.  Then we move onto water.  I am not sure about this because I am not sure of what we are having for water-affecting crates.

/my thoughts

Level design FTW!!

I think that list in the header would be better off being an unordered list of things to do, rather than a series of stages. I suggest using a more iterative approach where we first ask "what are the biggest risks that we face?" and then we mitigate those risks in order of importance, and as new ones show up.

So here are some risks that we face, off the top of my head:

• We might not introduce elements fast enough, which could annoy players who only want to learn how to make levels, or start playing user levels, as opposed to playing a full campaign. To avoid this, we'll have to make a list of the bare-bones elements needed to understand the fundamentals of most levels (probably the basic classes, the fact that multiple versions of basic tiles exist in multiple classes). We may also have to mark a specific set of levels as "The Tutorial Levels" so that people who only want to learn the basics have somewhere to go. We'll have to also decide whether or not these levels should be a part of the campaign's story or not.

• We might introduce elements too quickly, which might confuse players, or create an uncomfortable pacing. The only real way to mitigate this risk would be through effective playtesting.

• The interest curve for the tutorial levels alone should be effective, and then if these levels will lead to the rest of the campaign, this interest curve should neatly fit into the pacing of the larger experience. I might go about solving this by coming up with interesting traps and puzzles first, and then moving them around between levels and within levels until they yield an effective interest curve, both on the small scale and larger scale.

• We might start designing the tutorial levels too soon, before the final tile sets and mechanics have been finalized. Remember that power cell idea? Right now, I have absolutely no plans of implementing that into the game because I'm mainly focused on getting our core tile set down, but who knows whether or not I'll eventually add it. And if I do, will it be a core part of the game? And if not, don't you think that's a mechanic that's complicated enough to merit its own set of tutorial levels? How will those fit in with the other tutorial levels or the campaign? Will our tutorial levels be presented in such a way where we could modularly add more levels with the addition of new, similarly complicated mechanics?

• What about our technical risks? What if the game runs too slowly with large levels and I can't figure out a way to optimize the code? How will that affect the design of the levels?

• What about artistic risks? What if no one ever finishes Aeon's sprites sheet? Given the nature of how we've accumulated our visual assets, it's likely that the final sprite sheet might not have the coherence of a single drawing style. How will that affect the scope of the experience? It's great to plan to have all these different worlds and terrains, but who's going to draw it all?

• Speaking of scope, what will be the appropriate scope for these levels? They're main purpose is to give people who want to learn the game somewhere to go, and as stated above, the best way to do that would be to simply put up a designated set of levels titled "Tutorial Levels". If that's the case, do we really need a full campaign, then? We could always just make the campaign anyway and maybe put it under some kind of featured levels thing (if we ever add such a feature to the menu), which will then free us to use external assets, because they're just user levels now, as opposed to official, built-in levels.

I'm sure we can think of more things that might ruin the game, but this should be enough to get us started for now.

Sorry for the double post, but people are online, and I just remembered something important.

When designing a game it's useful to apply a theme for the overall experience that you're trying to make. The theme can be as simple as "blowing stuff up is fun", as moving as "the Holocaust was really, really bad", or as deep as "love is stronger than death". The main purpose of providing a theme is that it provides a real sense of focus to the game, where every decision in the game is meant to support the theme in some way. When every aspect of a game follows a good theme, it can create a really powerful experience, regardless of whether or not players understand or recognize the theme.

When I first learned about this a few months ago, I immediately thought, what is the theme behind Aeon? After thinking about if for some time, I settled on the theme that this was "the Interguild's Game", something that could represent this community, its creativity, and its incredible resilience to survive.

This means that the game would have to place its focus on the user levels, because what could better represent our community than our custom levels? This is also why it's so important that the game's customization options allow players to express themselves in their levels, because the community could then leave its mark on the game by having its levels reference other members and by perpetuating running jokes. And this theme also provides a reason why it's important to have customizeable mechanics because the Interguild's Game should at least last as long as the Interguild itself.

But where do the campaign levels fit into this theme? Their main purpose would be to provide new players with something of supposedly decent quality as opposed to a badly designed user level. But a much better solution that is inline with our theme would be to simply have a list of featured levels, or any other mechanism that would allow players to avoid not-so-good levels.

Or what if we changed the story of the campaign to be as light-hearted as the Interguild Comic? Even though that sounds cool, and I'm sure we could fill it with a ton of inside-jokes and Interguild references, it would ultimately be a product of the Interguild's current state in time, rather than a timeless representation of who we are. Not to mention how inaccessible it might be to new members.

I really like the idea of having the Tutorial Levels be the only built-in levels of the game. It gives players a place to go in order to learn more about the game, while keeping the final package focused strictly on the user levels. It's kinda like how Team Fortress 2 has a short list of offline training missions, but the full game can only be found online.

I thought the theme was "Screw TNT"?

The purpose of the campaign is to attract players to the game. To show them what is possible in this game, give them some ideas on the levels and leave them wanting more. Only using featured levels won't work for this. The campaign needs to have increasing difficulty to give the players a reasonable entry level and keep them interested. Semi-randomly selecting levels won't work.
For this to work we will need to keep the campaign short. Such as 3 to 5 levels per character, plus the tutorial and final level. The difficulty also needs to be kept low enough that everybody can beat the game in a single afternoon.
Another thing to keep in mind is how to get new players to even try out the game. On sites such as Kongregate, really the only games that people play are the games where you can earn badges/achievements. These achievements are generally something like "Beat level X". If we don't have a campaign, the chances of the game getting achievements and getting new players are greatly reduced.

As for the tutorial, we have two options in my eyes. We can have each character have their own short tutorial, introducing new mechanics and the character-specific abilities, or we can have one larger tutorial featuring Hugsy, the vanilla character. If we go with the second option, then the first level of each character should contain a very short section showing that character's special abilities, such as high ledge for Cyborg Eagle right at the start.

As interesting as this thought is, can you name me one game similar to this one where the "full game" is the level design aspect?

And yeah. If you want to market the game, we need a more developed campaign than just a few tutorial levels.

There's a difference between themes and the original idea behind the game. Just because we wanted to make a HATPC clone doesn't mean that "HATPC clone" should be the game's theme.

Hmm, I guess you guys are right about the campaign. But if the campaign is supposed to show what you can do with the game, don't you think it's oddly limiting to keep the levels focused on the core tile set and not even touch the rest of the customization options? I guess it all depends on how we present the advanced options. I remember a discussion a few months ago where we decided to treat these options as a side feature instead, with the analogy being that the advanced options allow you go inside the game and mess with its wiring.

I like the idea of keeping the campaign short. There's really no need to make a 60-level epic, which I think was the plan at one point. This will also ensure that each level presents elements and concepts that are in there for a reason, and not just because we were running out of ideas and we needed something to fill up the extra space.

Huh, how about we make a 'Game Mode' (as in an epic 5 world thing) and a 'Tutorial Mode' (a mode where only the concepts are introduced)? That way, if one just wants to make levels, they can choose the Tutorial Mode and if they want to have some fun (heh) they can choose the Game Mode?

Anyways... to attract people to the game... give them cookies. As in, cookies that keep your progress. And... you might want to add a bit of conversation to make the game more attractive... maybe before a level? A (press space to skip) is necessary for that.

Ok, this doesn't make any sense. This is just incoherent. I hope you understand what I meant to say...

I think I told Isa an idea the other day about using floating text in the level as a way of exploring the character's thought process as you move through the level. This way it's woven more into gameplay and it could help us make deeper characters. It could work, as long as the writing doesn't turn out to be terrible...

Oh, I'm good at writing such things. If we decide to do that sort of thing, if you want, I could do it, perhaps?

I'm good at writing, or so I've been told.  I have a hard time believing that.

When you say floating text, you mean independent text, right?  Not built into the level itself?

And Livio.  Why not make a 60-level epic?  Sure, it's about the community, but since we are planning on submitting more of a demo version to sites like Kongregate and such, wouldn't it encourage people to visit the site and make their own levels?  After seeing a short campaign, they might start to think it's pretty cool and wonder what they could come up with, but if we make it longer and show off the deeper mechanics of the game, then I think it'd allow the people who want more to get it.  I'm not going to delete everything I said, but I just realized that if people want more, then we want them to come to the site.  Either way, I think we want it to be a good length, longer than short.  We won't run out of ideas, Livio.  We are still coming up with unique things in HatPC, and Aeon will be more diverse than that.

/random jumble of randomness that doesn't make sense

Do we want the text to be straightforward or poetic and mysterious?

When I said "writing" I was thinking more along the lines of storytelling, not the kind of technical writing that they teach you at school. This is a hard skill to do right, and so it would be very risky to include that kind of aspect in our game.

First of all, it will take us a really long time to make 60 good levels. With so many levels to make, I feel like we would start to value each level much less and we'd get sloppy. It'd be a better idea if we could first prove that we can make a really good short campaign before we attempt to make a long one.

Second, while it might be cool to have a ton of levels all forming one big, epic experience, at some point you have to think, does the game really need to be that long? If the point of the campaign is to leave the player wanting more, enough to browse the user levels, then having a campaign that overstays its welcome really defeats that purpose.

I reckon 5 levels per character.

If we want to find the ideal amount of levels, I imagine we'd need to first figure out what concepts we want to teach the player and make a dividing line between the tutorial levels and the other levels. Then we can list any other additional concepts that we would like to teach that didn't make it into the tutorial. And finally, we can come up with individual trap ideas that manipulate the mechanics of the game in creative ways. This will bring with it even more concepts to teach the player.

From there we can start grouping things together based on how practical it would be to introduce what in which level. Designing the levels will be tricky, though, because there wont be a single linear order that the player will play all of the levels through. We may want to rethink this feature, given the length of the campaign.

While it's important that the player is left wanting more, the player should not be left expecting more, especially if our campaign turns out to be short. For example, Portal's levels were labeled "01/19", which meant that by the time you reached level 19 after one or two hours, you wouldn't be so disappointed at how short the game was, and you would be even more surprised to see how much longer it really lasted. Assuming we're still going to go with an overworld map, as opposed to a simple list of levels, then we might be able to display the player's progress by a meter that reads "Distance from Ship" or something (they are trying to reach a ship, aren't they?), and the player could watch this meter animate every time they finish a level.

Looking at your step plan in the header post, how can we even discuss a difficulty curve without being able to test it yet? Or do you intend for us to say stuff like "the levels get harder later in the game"?

Skipping ahead to Step 3, I have an example for how the levels could be set up:

Hugsy the Hamster Set


Floyd the Wolf Set


Felix the Eagle Set


Jean the Cat Set


Ivan the Otter Set


Rita the Armadillo Set


Final Set


Bonus Set



With this set-up, there will be exactly 2 levels in the game (1 Tutorial, 5x3 character sets, 1 Final and 3 Bonus). There is almost no overlap in the level styles between these levels, while still including all the important ones. You could also be introduced to all concepts in the game in under an hour by playing the tutorial and first level of each set. The levels themselves are also varied enough to keep the player interested all the way through and should give the player plenty of inspiration for their own levels.
If people can't get enough, we can always release a Player's Pack with 10 levels for each character later.

I just read up on this topic. I feel like I like having about 5 levels per character. But going with this creates some more interesting problems.

In fact, I think that jellsprout kind of overlooked the importance of the difficulty curve in his post. The difficulty curve isn't just "well they get harder the more we go on"

Going with what you did there, jell, we have 5 sets with 3 characters, and it sounds like we play them in whatever order we want. Okay, so what does this mean for the difficulty curve?

Well it means that we go straight from easy to medium, and medium to reasonable but difficult, which feels like it might be a bit of a drastic change for someone new to the game.  That would just learning hatpc, and going from level 4, to level 10, to level 15. If we move the number up to 5 we can at least make the difficulty curve smoother. But then there's also the problem that after said newplayer finally beats the first set they play, having evolved in their caveplaying skills, they start a new set and the opening level is now underwhelming, because they've just beaten a level noticeably more difficult than it.

And even if we make it so that you unlock one set after another, and the difficulty increases in each set, then we get the problem that all of none of X's levels are challenging for experienced players. It would be like if Armin did the first 10 levels in hatic, and hannah the last 10, and someone is like, "Well it would have been nice to have armin play in more challenging level somewhere along the line." It is true that this is kind of what the user levels are supposed to be for, but the campaign levels are supposed to be near perfectly designed, user levels are not usually designed as well.

Maybe we want to think about the possibility of merging the sets into a linear campaign. Each time you get introduced to a new character, there's an optional tutorial level introducing them and then 2 levels dedicated to them. Once all of them get introduced, the rest of the levels are pretty much randomly scattered, except for the last 5 levels which are given to each character in reverse order to make up for the fact. And then once you beat them all you can play sets consecutively without interruptions from other sets or something. I just worry that the difficulty curve will be either to drastic or non-existent. Anyways if we merged them, it would go something like this:


I'm just throwing the idea out there to.

For the Bonus levels, (insane) should glitching be exploited like in some other games? I say no.

I used to have this crazy idea of a SMW-like map but each character starts at a different place on the map and each works their way toward the centre. Eventually some characters will meet up and you can replay the levels again with other characters to maybe find secret exits that could potentially allow you to meet up and reach the centre faster.

I like SMW's path choosing aspect and the choice of multiple levels at some points.

@Quirvy:
I personally find ~30 levels to be a bit too much. From experience I can tell that if a person can't complete a game in one afternoon, they generally won't bother after that. If the levels will take an average of about 15 minutes, the game will take a total of about 8 hours to complete. That is about equal to the shorter console games, such as Prince of Persia and the original Halo.
We can try 5 levels per character, but I don't know if we can have enough variety in the levels then. I already had enough trouble coming up with my 26 concepts. I don't think this is of much concern at the moment, as we can always just start at 3 levels per character and then add 2 more if we feel it is necessary.

Tutorials are generally terrible, especially if there are 6 of them. I figured we could include the character specific tutorials in their first levels. They don't need to show exactly how each special ability can be abused, the levels themselves are for that, but just one long jump to show Felix's gliding ability, one pit of quicksand to show Rita's rolling ability, etc.

I figured the difficulty curve would also be present inside the levels. Floyd 1 would start simple but get more challenging toward the end and then Floyd 2 would start at the same difficulty as Floyd 1.

And your post has given me a new idea on how to do the levels. How about you start with the first five levels unlocked, but you need to beat all five levels to unlock the second level for each character.
So similar to your set-up, but levels 1, 2, 3, 4 and 5 can be done in whatever order you want, but must all 5 be done before you unlock levels 6, 7, 8, 9 and 10.

@CSD
I also imagined a world map such as that, but I don't like the idea of having characters play the levels of different characters. That would just make all levels a hell of a lot more difficult to design and is generally pointless. As Livio said, the main purpose of the main game should be to attract new players to the game and inspire them to make their own levels. Designing each of our levels to be beaten by every character doesn't really help that purpose. In my set-up we still do one level that can be completed by each character, but doing every level like that is a bit redundant.

@Jazz
If we find any odd glitches before release, we should obviously remove them. But if we find a glitch that could become a cool feature, such as wall jumping, then we could use them. The bonus set should at least be optional to be played, so we can make the techniques required there far more difficult than in most levels.

Who said the levels had to be possible with every character? Put a "recommended characters" screen or something before you start a level, and if you take a different character and you find out the level is impossible, just exit the level... ?

But on the other hand choosing some characters might allow for the player to cheat their way past a level, so it would still force us to test the levels to make sure that it isn't easier to beat it with the non-suggested characters.

That's not very nice to the users.

The idea is you would only go back to other levels like that if you're looking for secrets. It's OK to be a bit of a jerk when you've got secrets to hide. I just think it'd be cool to have a lot of secrets hidden around our game, and I think allowing you to replay levels with other characters could be a good way to do it.