I’m sorry I wasn’t able to post at all during December, December is just a generally busy and stressful month with Halloween and everything. Halloween is all about suspense and spookiness and such so many games are trying to get into the holiday spirit by making you   However I’m building a metroidvania platformer and I have no intention of making it a horror game. However, I may still be able to incorporate suspense into the level design for a purpose other than spooks.

I loved Cat Nigri’s “Necrosphere”, if you are interested in metroidvanias I would strongly recommend looking into it (even though it only has 2 ability upgrades). It (much like my game) ditches the convention of large expanses of empty space, in necrosphere everything is crammed together so you are never more than a few seconds before the next challenge. This happens everywhere except in one place where you do almost nothing for about a whole minute as it builds suspense.

The whole first hour or so of necrosphere you have no way of moving up, you can only move left or right. Then suddenly while walking you see the background change you slowly drop


This immediately made me think I was finally going to get vertical mobility, and as I continued I dropped down the pit after pit reinforcing the idea that I was probably about to finally be able to jump or something. Notice how each ledge is only a couple of blocks beneath the previous ledge, this gives you some clues to how a potential vertical ability will work, it drops subtle hints that the ability will only allow you to move up a certain amount. Even though I encountered no challenge for almost a full minute, I was completely engaged, I was trying to determine through environmental clues how the ability would work, would it give you a jump, a wall run, a jetpack?

So I wanted to incorporate this into my own game, to before getting an ability hint at what the ability is going to be. For example one of the abilities is a blast that lets you reveal secrets 


and so I made it so that shortly before you get it you must pass through this area


the extrusion of the ceiling to almost touching you is enough to make anyone notice it, and the slightly different color of the bottom blocks makes you feel like something is special about this, it also helps that this is the first screen where you will ever see floating enemies and also the first screen where you ever encounter platforms you can drop through so it can help you remember it. On the way to that same blast power up you later run into this:


which is also a very memorable location due to the new enemy type discovered here. This isn’t a location you will need to go back to in order to progress it’s just an optional side path resulting in a coin. The overall purpose of all of this is to get you thinking about what the next power up could be. And you eventually make it to this room


a relatively difficult (harder than it looks) room that may take you a few tries so you’ll spend some time in there. However, the whole time you can see the power up at the top and you can’t help but guess what it’s gonna be.



I really enjoyed the checkpoints from Ori and the blind forest,  for those unfamiliar with the game, your character has the ability to create his own checkpoints.  However, this uses up some of a resource that you use for many things in the game.



This requires you to try to estimate how dangerous the upcoming area will be.  I thought about how I could introduce an interesting system for checkpoints in my game. My game has no sort of resource that I could use for that nor do I want to have the exact same system. I thought about what sort of emotion I wanted to instill within my player with the checkpoint system. I want it to feel like you are taking risks, I thought about shovel knights system where you can destroy checkpoints for money.


This makes it feel like you’ve given up your safety net and now you can’t mess up. I want the same emotion but I don’t have anything you could gain from destroying a checkpoint (except coins but I don’t want to utilize them that way). Then I thought about it again, you could have a resource if the checkpoint itself was a resource. Imagine that throughout the game you have locked checkpoints that behave exactly like checkpoints from any other game. But you also have a “mobile” checkpoint, which you can place then pick up and put down.


if you die while the mobile checkpoint is out you respawn at it. If you are holding it then you respawn at the last locked checkpoint that you activated. If you reach the next locked checkpoint while your mobile one is still out then the mobile one comes back to you.


I think this definitely creates a risk vs reward mechanic because the only way to move your checkpoint through the levels is to pick up your checkpoint and have to go through areas without having a nearby checkpoint.  This also means that even though you will die a lot (so far this game is very difficult) you don’t get set back far and the respawn times are instant (i.e. super meatboy).

However, that might be confusing for the first bit of the game and if I wanted to really make the players feel risky for picking up checkpoints I would need to put quite a bit of space between locked checkpoints (about 3 minutes of gameplay maybe) but that’s a lot to lose for one mistake, so maybe I should stick to the normal checkpoint formula. Comment whether you think that this checkpoint system is worth anything or if I should use normal checkpoints.

Level Design Philosophy

Now that I have the jump, it’s time to build the tutorial.  I need to start making the very beginning levels which will serve as a tutorial for the player (even if they don’t realize it). When designing the tutorial levels you have to make sure each section of the level is:

Great at conveying some sort of message to the player that is relatively easy to observe and isn’t cluttered. Like a screen of a level that introduces some new enemy or obstacle shouldn’t also introduce another mechanic at the same time. You should determine what the message is before you start making the level.

And when I say message I mean a mechanical message like “You can jump higher by holding jump” not a narrative message such as “God is dead, nothing matters, oriental ramen is the best ramen”. Narrative message in level design is an amazing thing but it’s not what I’m talking about.


This level is essentially broken into 2 levels, the top part introduces moving threats that must be dodged. It is immediately moving towards you down a long hall with a high ceiling so it’s easy to jump over and if you don’t jump you will be returned to the nearest checkpoint which is right before you enter the room. So if you fail you’ll be trying again in 2 seconds.


Then there’s the lower part of the level, that might take several tries for a new player. It’s purpose is to introduce platforming challenge. If you keep moving in the same direction after jumping over the enemy, you’ll land on solid ground. Notice how there are no threats in the lower section of the level besides falling at first. However in order to eventually force the player to make the first jump, the enemy will eventually fall back on the player.


I think of levels as having layers, (much like onions, ogres, and Taco Bell’s 7 layer burrito) layers of how players interact with them.  There’s the layer of what is immediately obvious to the player about the room, in this example it would be that there is an object moving towards the player. Then you have the layer of their options to achieve their goal, this room has a high ceiling which allows the player to jump over the object. But even within the category of options there are multiple layers depending on the player’s skill and abilities.


That is how a newcomer to the game would experience this level.


And that is how someone with a double jump would experience it.


And that is how a speedrunner would experience it.

This one screen level has many purposes and many different layers (and all the same principles apply to playing this level backwards). I will continue trying to put this amount of effort into each and every screen.

If you want to see more of my game check out our twitter.


Now I have the secret system down and I’ve developed the first few levels with lots of secret areas to explore. The only problem: I don’t have any rewards for the secret areas. I could have all secret areas end up going into useful paths but then is there really a point in having secret areas? I could add coins to collect and have most secret areas lead to areas full of coins akin to Donkey Kong Country Returns.


 And while I’d love to develop DK’s system further (in order to expand Kong) I have no use for coins. I have no live system (I think gameover is a useless artifact from arcade days), so what would the coins do? Maybe every 100 coins you collect gets you a new power-up. Well, that means that I can’t decide exactly where players get power-ups, which would ruin the whole metroidvania theme.  I decided to instead of having coins be like coins from Mario, I made them into collectibles. By that, I mean that you won’t find trails of coins but only one or 2 coins in each area. But I still needed to think of a use, I think it would fit with the metrioidvania theme if it there were doors that required a certain amount of coins to pass through, this would encourage backtracking even more.


I also thought about how in the newer Mario games, lives aren’t really that important because of how easy the games are, however people will still try to collect every single coin. People are motivated by many different things (I am solely motivated by taco bell’s 7 layer burrito) some people are motivated by the feeling of achievement that comes with getting collectibles. So if people see a golden, super mega, space doubloon they are going to try and get it even if there’s no reason because for them the reward is the reason. So I tried to think of a system that would encourage players to collect coins even when they haven’t seen a locked door for a while. Something that would motivate players to collect coins even without a mechanical reward. I literally got inspiration from one of the previous sentences. “Golden, super mega space doubloon” I can just give each coin a really stupid name that displays when you get it, that would encourage players to find them. They would have to be increasingly stupid names so that you would want to collect as many as you can. So I added a mechanic where when you collect a coin it displays “You collected a coin” and every time you collect another coin it would add one adjective to the coin. I think the randomly added attributes to the coin has the potential to be absurd enough to be funny. I think that procedural and random generation has a pretty large untapped potential for humor in games in general.

Hopefully, players will want to do this:


When the reward is this:





Jump Physics

It’s simple enough to make a very basic jump. When you press jump, you go up at a certain speed, and while in the air you accelerate downwards. I figured it should be easy enough…


Well, that’s definitely not what I’m looking for, this jump is the tofu of jumps (tofu is a vegetable that is unable to control its vertical speed). You have no control of how high you jump and it just doesn’t feel satisfying.  I figured I could give the player control of their jump by cancelling your upwards movement when you release jump.


While this is able to control its vertical momentum, this feels almost aggressively awkward to control.It looks like your character is hitting his head on some invisible ceiling. So what if instead of canceling vertical speed I just set it to a lower value. I decided to try halving the vertical speed once space is released. This makes you slow after releasing jump, instead of stopping completely.


Now that looks nice, and you can control how high you can jump.  But I think it needs to feel more punchy. Right now I think it seems like you’re a piece of paper blown up in the wind. I still want to fall (somewhat) slowly after the peak of my jump, but I don’t want the ascension to feel floaty as well.  My idea is to increase gravity when you first jump, and decrease gravity slowly afterwards.


Now your character feels more solid and I’m finally feeling satisfied with the jump.  Next I added a simple “ghost jump”, which means that a few frames after you walk off a cliff you can still jump.  This is in most platformers and it ensures that the player doesn’t feel cheated by not jumping after leaving ground.


This means whenever I try to jump it checks if I’m on land and also if I was on land recently. However this created a problem. If I jump then press jump again in a few frames it will jump again. So double tapping jump will add your jumps together and go higher. This is because it checks and finds that I was on land recently and jumps again after already jumping.


That’s weird. I should probably remove this glitched double jump thing. Or maybe not, even though it was unintended it isn’t necessarily bad.  Most players will never discover it, and it adds an extra layer of skill for speedrunners because it’s quite difficult to pull off. What do you think? Should I keep it?