Scrolling Survivor : history

Meet Scrolling Survivor, a retro auto-scrolling difficult black-and-white plateformer ! If you're running Windows, you can download it for free here !
However, let's have a look at this project which started back in 2008 and proved to be much more ambitious than expected.

English is not my mother tongue ! You may find some mistakes here and there. Sorry about this !
La version française arrivera plus tard. Ou pas. Ça dépend.

First demo

Back in 2008 ! The idea of making a platformer with black-and-white graphics came from another freeware demo game : Coal. While Coal used vector-ish graphics, Scrolling Survivor uses pixel-art, with every sprite having a 2-pixel border and using shades of grey for background.
Having created some platforms and a potato-like character, I wondered how the gameplay could be more original and interesting.

The very first level.
Then I remembered the auto-scrolling sub-levels in the first Rayman games, where you had to keep moving forward. As a child, these levels always frustrated me as you died immediately when Rayman touched the screen's borders, but maybe this gameplay mechanic could be used and improved.
I gave myself 2 weeks to finish a demo, since I was then going to move from my family home. The demo was made with Clickteam Fusion (which was called MultiMedia Fusion 2 at the time).
Also, the whole chiptune soundtrack was made by a cool composer called Arachno ! His music was first heard in Lyle in Cube Sector (a great indie freeware platformer) and then sent him a mail, which he answered positively.
Here is a list of good ideas that were put in the game during these 2 weeks :
  • Multi-directional scrolling with pauses.
  • Knockback system like in Super Smash Bros.,
  • Scoring system, increasing replayability.
  • 1-up for every checkpoint reached (in a hard platforming game like Rayman 1, when you have 0 lives left, don't you hate reaching a checkpoint for nothing?)
  • Surfboard level !
  • Ultra-hard mountain level in which you have to struggle against the wind ! My sister even had the nice idea to put a wooly hat on the character.

"Snow Storm" level in final version.
However, some important aspects were neglected :
  • The controls were awful ! Seriously, the character was moving too fast, the player had almost no control over jump height, and most of the deaths in playthrough videos are due to this.
  • Some levels were incredibly unfair.
  • Animations were pretty simplistic (the character's walking animation had 3 frames, there was almost no jumping/falling animation).
  • Some parts of the scoring system were too complicated (with a weird time penalty when you died).
  • The tutorial was terrible : huge walls of text that you had to read during gameplay, while playing.
  • When you got hit and knockback'ed, you had the possibility to make a recovering move, just like in Super Smash Bros. By tapping an arrow key twice, you could dash in any direction if you had collected "spheres" before. However, a lot of people died by triggering unwanted recovery moves !

    NO ! COME BACK ! WKMZ@YD%f+HS!#!!
Most of these mistakes could have been avoided if I had much more people testing the demo before release. I made a begginer's error : thinking that if I could beat a level, then everyone could also beat it.

Anyway, the demo was made available on some freeware game websites (World of Goo and Braid were not available at the time, so only a few people knew about "indie games" and probably no one was using Facebook or Twitter for game-design discussions!)

Initial reception

The comments and reactions were much better than expected : people wrote really nice articles and some players even made let's play videos !

The criticism was focused on one aspect : difficulty. Or maybe I thought it was about difficulty, where actually it was about controls (and also level design).
A new version was quickly built, with a cheat code to get more lives, a difficulty choice which changed almost nothing and even some unclockable hints (?). But as you don't make a beat-them-all more difficult by giving more health to enemies (it only makes the game more boring), you don't fix a platformer's difficulty by tweaking some parameters here and there. I'd only realise this some years later...

So what do we do now?

Err... nothing?
Well, even though the reception motivated me a lot at first, I felt quite overwhelmed with all the work left to do. There were only 3 levels in the demo and I wanted to make 15 levels ! Why 15? Because, 15.
Instead of keeping up the good work and update the game every month for instance (as the whole 3-level demo was made in half a month), I thought I had to finish and polish the whole project in one run, which made me quite demotivated.
As you may know, when you finish 90% of a computer project, then you have to finish the other 90% ! But I got tired of this even before finishing the first 90%...
Here are some things that could have been done to keep working :

  • make smaller levels and show them immediately
  • make polls about the demo, ask about difficulty, level design...
  • make a goddamn list of features that I wanted to add, and then stick to this list (seriously, how could I forget to do this ?)
  • add online leaderboards quickly to improve players commitment


Several years passed and I even remade the project from scratch multiple times. First I had the awful idea to use vector graphics instead of pixel art (still using a 640x480 resolution !)

Then the game was remade using Construct Classic instead of Clickteam Fusion, but Construct was quite unstable at the time. And Construct 2 was to be released some months later...

So, back to Clickteam Fusion. An enormous mistake, as this piece of software is now totally outdated. No easy way to design levels, no tilemaps. "Events" system with no sub-events, no for/while/foreach loops, no functions. You have to copy/paste your code into every scene. If you want to export your game to other platforms (Flash, mobile, no Mac/Linux), you have to buy add-ons and most extensions aren't even available for these platforms anyway.

This is what I see when working on Scrolling Survivor.
More than 1,500 lines, and you can scroll 6 times to the right !
Of course, there are workarounds. A lot. Actually you have to use workarounds all the time. Some Fusion-made games (Knytt Underground, Freedom Planet) were released for consoles, but thanks to a case-by-case solution which is not publicly available. You can use tilemaps, but thanks to a sprite font extension (!), only for Windows. You can have loops and functions, but they make your code unreadable.
To be honest, running Clickteam Fusion again to put up a "finished" version of Scrolling Survivor on made me feel like going back to stone age. I've used Clickteam's products a lot, but it's time to move on. If you don't know programming and/or want to make games on your own, go for :
  • Construct 2 if you want to start quick with no coding, don't want to make 3D games and don't mind only making HTML5 games (they run fine on modern PCs, but there are still compatibility issues on mobile).
  • Unity for any other use.
There are many other solutions (like Game Maker), but I haven't tested them enough.

Anyway, the game was remade with :
  • tighter controls, wall-jumps
  • more detailed animations,
  • no difficulty levels but you can keep playing after losing all lives, with a score penalty,
  • sub-levels allowing to make better use of each level's mechanics,
  • ability to fly for a short amount of time when you get hit,
  • online leaderboards !

Good stuff

After some months of work, I managed to show the game at a board game festival in France (where it got played by many enthusiastic young gamers and won the jury prize, woohoo !).
Scrolling Survivor was also submitted to IGF, where it didn't win anything but there was quite positive feedback ! Reading "Overall though, definitely something I'd like to spend more time with. Good-oh!" or "Well, I fucking adored this, I love so many little aspects about it that blow apart how we think about regular platformers" really pumped me up at first.

However, I could not find other ways to get feedback at this time, and my motivation vanished again. I also spent time to submit the game to a website where players were supposed to buy a bundle of early-access games and then get involved in development. The idea seemed nice at first and I tried to keep working, posting about development progress, while listening to feedback... but the website wasn't known enough and only one person sent me a piece of feedback. Too bad !

Finished (?), at last !

Two more years later, I won't probably finish Scrolling Survivor.
Still, there is some polished content : 7 full levels, 17 sub-levels, the nice scoring system ensuring great replayability. If you want to finish every level, there is at least 3 hours or gameplay (much more if you try to unlock every star !).
There are some flaws too, the tutorial and the gamepad controls are far from perfect.

I tried to get back working on the game multiple times, but couldn't find the motivation. Having to make it with an outdated game creation tool didn't help. In addition, I didn't want, or didn't know how to show the game to potential players while it wasn't finished, as the game was started way before early access even existed. And I recently realised that as I would not finish it, keeping the game for me was totally pointless.
So, why not putting the game on the Internet and letting players enjoy Scrolling Survivor? is the perfect place to do this. They accept every game, you can have highly customisable pages, it's wonderful for small indie developers. Adding the game there was pretty quick !

And now Scrolling Survivor is kind of finished, as players can now try it for free. It was long and often frustrating, but I've no more regrets putting so much time in such a project, as I learned a lot and had positive feedback all the time.
Thanks for reading this, and have fun playing Scrolling Survivor !

Any questions? Send me a mail : webmaster (at)