Why I’m Making Sunblaze, a Celeste-Inspired Game

12 min readNov 25, 2020

First, if you haven’t played Celeste, do yourself a favor and play it right now. It’s the best platform game ever made IMO, and well worth playing. No, really.

The game I’m working on — Sunblaze — has a lot of similarities with Celeste. Some intentional, some accidental. There’s a lot to set it apart as well. In this article I want to share the story of why I made this game.

It’s important to note that in my opinion (and in the opinion of other people who have played both games), Sunblaze actually feels quite different to Celeste when playing, but this doesn’t come across in videos or screenshots very well. That’s not not to say that the inspiration isn’t there, it obviously is.

If you’re interested in how the physics work in Sunblaze, check out this article: https://gamesfromearth.medium.com/a-simple-2d-physics-system-for-platform-games-f430718ea77f

TL/DR

The short of it is that I had some ideas for a game, and Celeste-style gameplay seemed like the best choice for it. After I prototyped it, I noticed it was actually a lot of fun, so I turned it into a game.

Part of it is also that as a player, I want to experience more games like Celeste. I want it to be a genre. I don’t think we should discard a style of gameplay because one game was made in that style, no matter how good that game is. I believe there’s always room for iteration.

I sent a demo version to someone who was quite adamant on Reddit about Sunblaze being nothing but a clone and a complete rip-off. After playing it, he told me that he was wrong, and that the game indeed feels like its own thing while playing.

Games taking inspiration from other games isn’t new. For instance, the indie title Freedom Planet borrows a lot from Sonic The Hedgehog.

Freedom Planet, an awesome Sonic-inspired indie game.

Conception

My initial idea for Sunblaze came years ago. I envisioned a game based around a female superhero sidekick getting trained by a mentor (and I wanted to name the game Sidekick at that point).

The next part of the idea came when I saw this video :

This video predates Celeste by a couple of years. The part around 5:20 gave me the idea of having my game take place inside one constantly evolving level inside a training simulator. At this point I imagined my game to be more of a bullet-hell, as shown in the video, but I couldn’t figure out how to do level design for that. But some remnants of that are still present in the 3rd section of my game, where there are a lot of lasers to dodge.

After playing Celeste, I realized my game idea needed a dash. At this point I thought this wouldn’t be too much of an issue—other games before Celeste have had a dash mechanic, and I figured I’d make my dash more like the one from League of Evil, where you can only dash horizontally. So I started prototyping.

Choices

But then I hit a problem. Celeste is a game that does a lot, and it makes all the right choices.

I was often facing the decision of either doing something in an inferior way, or doing it the Celeste way. For that reason I stopped working on it for a couple of months. But I showed it to a friend of mine, and he really liked the room transition effect and the smoothness of the gameplay, and he convinced me to keep working on it. So I did.

One example of having to chose between the way Celeste handles things or doing it in an inferior way, is the color change to indicate whether or not you still have your dash. The way I saw it, I had 4 options :

Don’t indicate this at all. This was a terrible option.
Show it in the HUD somewhere. Not as bad as option 1, but still pretty bad, because the player can’t easily keep an eye on it.
Do it exactly like Celeste (changing the hair color).
Do it in a similar way, but slightly different.

I opted for option 4. My initial idea was to make it look like the character was on fire when she had her dash. The pixel artist who made the character art even made a version of that, by having fireballs come from the character’s hands.

But these looked too much like weapons, and it was weird that you couldn’t shoot them.

Next I tried overlaying some particle effects to make her look like she was on fire, but that didn’t work out either.

So instead I made it so her outfit changes color.

This is close to how Celeste handles it, but still different. Some people will probably think it’s too similar, but anything else that I came up with made the game worse, and my reasoning was that in the end, I didn’t want to make bad choices for the sake of being different.

Another example of this is the hires menus. The reason why I did this is simple : I’m not good at pixel art, but I can design hires UI myself. So it’s mostly a budgetary choice, and it also makes internationalization easier. But without Celeste, I probably wouldn’t have considered the option of having hires UI in a pixel-art game.

Accidents

Something that happened a couple of times was that I stumbled on something that I wasn’t planning for, only to realize it was something Celeste also does.

One example of this is the “hyper dash”. Of course I knew this was in Celeste, but I didn’t plan on programming it, it just happened while I was trying to do something else. The Celeste devs might have stumbled on this the same way I did, but I can’t say that for certain.

When you press “jump” while dashing, it looks weird if you completely cancel out the horizontal movement and start moving up immediately. It’s jarring. So my plan was to have just a little bit of leftover momentum from the dash, just to smoothen the transition. So I programmed that, and when testing my code, it turned out that the values I had entered for this gave me an accidental hyper dash. Basically there was too much leftover momentum from the dash, propelling the character into a curve.

After playing around with it for a bit, I reaized that in my game you could do this in mid-air because of the double jump. I really liked doing this, and also thought it would be great for speedrunners to use this move they were familiar with.

I tormented over whether to keep this in for a long time. But in the end, I felt like keeping it made the game better, even though it makes it even more Celeste-like than it already was. Having the ability to do it mid-air makes it slightly different, but still.

Unique Ideas That Weren’t

At one point, I realized I needed some more variation in the gameplay. So I had the idea of adding a zig-zag mechanic, based on an old mobile game I made called Zig Zag Boom (which I made with Mikael Tyrsen).

The idea was that you would enter a state where you would move diagonally, and tap a button to change the direction. This is an original idea for a platformer, right? It’s not in Celeste, let’s go for it!

Initially I figured I’d have special areas or blocks for this mechanic. Upon touching these blocks, you shoot through them diagonally and tap to zigzag. Oops, that sounds a bit like the dream blocks from Celeste, only with added control. Gotta think of something else.

I had already nicknamed my character Sunblaze, so I thought it made sense to have her turn into a fireball, and use the zigzag mechanic with that. The fireball also seemed like a nice throwback to my old game.

So I implemented it, and when it was done, I realized that it looked like the feather mechanic from Celeste. It controls completely differently (single button versus stick), but you also turn into something yellow that moves automatically.

So I basically added an idea from an old game of mine, and somehow I was ripping off Celeste again :)

I chalk this up to Celeste doing so much.

So I added some more unique elements to this mechanic. What I came up with was that in fireball mode, you can’t touch water, but you can move through flames, making it so you have to plan your path more carefully.

To make things worse, I needed something to create paths for the fireball mechanic. Using the square tiles the rest of the game uses didn’t work well with the diagonal movement and didn’t give me enough granularity to make levels, so I starting using small buzzsaws from another chapter to create paths.

I soon realized that this was similar to how Celeste uses spiky crystals and dust bunnies for paths, but again, the alternative of using straight tiles didn’t work well. I do have to mention that I considered some other alternatives like generating some kind of mesh or having diagonal branches of some kind, but all of those would require a ton of work and adding new functionality to my level editor. Plus I would need to work out how to handle collisions for them. Clusters of small circular obstacles were just the easiest, most obvious choice.

The holes in the buzzsaws triggered my (very slight) trypophobia, and so I thought it would be cool to have the theme styled after that, with small holes everywhere. So that’s what we went for.

I did try to make these spiky balls (trypoplasma as I call them) behave differently from Celeste by sometimes attaching them to falling blocks so they have a more interactive aspect to them, but I concede that this is minor, and not actually used all that much.

Legality and Ethics

Legally it is perfectly fine to use gameplay concepts from other games. If that weren’t the case, we’d have no first person shooters besides the ones id Software decides to make, and no platform games made by companies other than Nintendo. Celeste also borrows plenty of ideas from Mario 3 and other games, but it does combine them into a package that’s more unique than what Sunblaze turned out to be.

So if not a legal matter, it becomes a matter of ethics. There’s three aspects I want to talk about :

How I feel about it
How the Celeste devs feel about it
The public opinion

1. My Thoughts

Personally, I really don’t like straight up cloning. I have had games of mine cloned in the past (Boom Dots, Zig Zag Boom, Down The Mountain, …), even to the point where they stole assets from these games. That’s not good.

I once published a game called Risky Room that turned out to be a straight up clone of another game. The mechanics, level design, art style, everything was exactly the same. To be clear, I didn’t make this clone, I published it for another developer and I wasn’t aware that it was a clone.

So once I found out, I tracked down the original dev behind the game and paid them most of the money we had earned from the game. I didn’t legally have to do that, but I felt it was wrong to have profited from someone else’s work without their consent, even though it wasn’t intentional on my part.

Sunblaze is different in my opinion. It isn’t a straight up clone. There’s no denying the inspiration, but it has its own feel to it (see the “Setting It Apart” section for more on that), custom levels, different art style, original music, etc.

It’s an interesting question though — at which point is a game too similar to another? I personally think that Sunblaze is different enough to be acceptable. Your mileage may vary.

2. The Celeste Devs

The next question is, how do the Celeste devs feel about it?

During a live stream where Maddy and Noel were playing a big Celeste community mod, I asked them how they felt about games being inspired by Celeste. They said that they think it’s cool, as long as no art is stolen.

I want to make it clear that this in no way amounts to them endorsing my game.

I don’t even think they’re aware of it at this point. But I asked, and it seems like they don’t mind.

3. Public Opinion

The final aspect is the public perception. From reactions on Reddit, I can gauge that about 3/4th of the people think it’s okay, and 1/4th think it’s not. This is significant enough to take into account.

On one hand I just want to make the game I want to make (which I did), and other people have no say over how I decide to spend my time or how I try to make my money. But I also want to sell enough copies to make back my investment in time and money, and to be able to pay dividends to my team. So for that reason, public perception is important. Also, nobody likes getting backlash.

I don’t know if this article will sway any minds, but I still wanted to share my thoughts.

Setting It Apart

With all these similarities, I do constantly feel the need to set it apart as much as I can. One of the ways in which I did this was by making the level design a lot more puzzle-based, especially later on in the game.

Most of the level gimmicks / obstacles are different. There’s lasers, explosives, thwomps, buzzsaws, rails to hang on, slimey blobs, etc.

All of the game’s controls function differently from Celeste’s. Like having a double jump instead of a varying height jump, ledge hanging and climb points instead of a climb button, etc.

I’ll admit that these changes sound minor on paper, and in videos it still looks similar. And it is, but I do believe that it amounts to quite a different gameplay experience, and that players will see that when they actually play the game.

To test this theory, I sent a demo version to someone who was quite adamant on Reddit about Sunblaze being nothing but a clone and a complete rip-off. After playing it, he told me that he was wrong, and that the game indeed feels like its own thing while playing. He even said he prefers Sunblaze over Celeste because of the simplified controls, especially when using a keyboard, which I think is quite flattering. I knew my game was decent but I wouldn’t have imagined someone preferring it over my favorite platform game of all time.

That doesn’t mean I’m done. I’m currently trying to find ways to set it apart even more, especially for gameplay videos. We’re still experimenting with different hair colors and different visual effects for the dash. I’m not sure how effective this will be, and if I’ll be able to come up with suitable replacements. But I’m trying.

I am also open for suggestions on how to set it apart more.

In any case, I hope you will give our game a chance before dismissing it as just a rip off. I think there are enough fresh ideas to warrant its existence, and I hope you will agree after you get a chance to play it.