Die to Stay Alive: A Student Project
One of the best and ambitious video game I worked on was Die to stay Alive, my final project as student – made by 4 Game Designers, 3 Artists and 1 Sound Designer. It was a good exercise because beyond its concept and qualities, it taught me important tips about pure development and team working. This post is kind of a post-mortem.
Here is a description I’m getting used to give for talking about it:
In a 2D Puzzle/Platformer, you control alternately a bunch of monkeys enclosed in a complex, directed by an evil scientist, to serve as guinea pigs. Your objective: escape to recover freedom. But to do this, you have to sacrifice each monkey cleverly to create a path to the exit. However, be careful to not sacrifice all your stock of lives if you don’t want guide your people toward extinction.
As showing things is always more efficient than describing them, here is a visual translation:
Above: the prototype trailer, with some tutorial features missing.
Summarizing the video content, the player can sacrifice each monkey from his/her stock and use the dead body with his/her next life to be able to solve puzzles. In order to do that, there are two possibilities:
Firstly, the player can press a dedicated button so the current monkey dies, transforming himself into a block – you don’t think that’s very affording? Neither do I, but I’ll develop that part much below on this post. Of course, the player loses a life point but can use this block to solve all or a part of a puzzle.
Above: a character transforming himslef into a dead block.
Secondly, the player can execute a kamikaze action choosing a well-placed trap on the game field. If the trap hits the player – for instance, a hammer smashing the avatar – the living monkey transforms himself temporarily into another character with new powers. Why temporarily? Simply because after a while, the monkey dies staying on the field just as if the player would have used the ‘sacrifice button’ described above. If we reuse the example of the hammer trap smashing a living monkey, this monkey would become flat like a pancake, so he would be able to use ventilators to fly! Our basic intention was to create some famous traps we all have seen as children – hammers, flamethrowers, etc. in order to get affordable features to use for solving mad puzzles.
Above: a character becoming flat, with one usable situation.
Above: a flat character dying slowly. Cruel, isn’t it?
The concept really excited us – it was in 2011. As a matter of fact, I encourage you to test the prototype before reading the next part! Please note that as we developed it on a very old engine named Virtools, you must be on Windows to be able to install the plugin. You can also download the standalone version on this page if there is any problem while installing the plugin.
There are several good points offered by such a concept:
Gameplay diversity increases well by adding one trap/transformation. In fact, as we can design a level with the traps we want, every single trap/transformation can be combined with another one, letting us create fun and diversified patterns.
The big content created that inherits from the good point above. In fact, the possibility to design a lot of game patterns also let us creating a lot of levels.
The replay value is important as there isn't a unique way to finish each level.
Sensation while playing is cool! We spent a certain amount of time polishing that detail, because embodying a flat monkey isn't the same as embodying a flamed monkey running fast!
The “Merrie Melodies” inspiration used for designing traps is very affording for the player as he quickly understand the traps, the abilities offered, the risks, etc.
It represents another way of seeing a puzzle/platformer game. In fact, there are traps and abilities known from every player because already used in another game. However, as we tried to use them differently, we were able to create some surprises.
Thanks to the level design, a trap can be as useful as dangerous, making the player hesitate. It’s good for both physical and mental skills, so it’s good the game!
Of course, we learnt a lot about the mistakes we did on this project. There are two of them I really want to talk about:
The level design structure.
The art direction puts in relief several design mistakes about the game.
First of all, the ethical problem. In fact, killing monkeys can be disturbing for some players, and it’s not something matching our first willing of creating a game playable by all core gamers. It’s also a problem while knowing that at the beginning, the concept was to embody a bunch of crash-test robots. Finally, monkeys are cool in their design, but not as affording as crash-test robots created to be destroyed. I really think they would have fit better the initial concept.
Above: the final design of the monkeys.
The art direction puts in relief another problem centered on the level design this time: player’s spatial awareness. In fact, all the levels were artistically designed as if the player would explore them, like in a platform-adventure game. Nonetheless, we wanted to get a puzzle-platform game, and this type of game often needs a purified dress code. As a result, we figured out that some players thought some parts of a level were interactive while they only were decorative.
Last but not least, there is currently a gap between the visual and the function. Some design assets didn't match their function at all. The better example would be the "monkey spawn” below:
Above: on the left on the screenshot, the evil monkey spawn. On the right, a hammer trap.
Its design is cool, but can’t match its function as it looks dangerous with deadly peaks all around the metallic structure! I think it’s representative of the difference of vision the artists and game designers had on this project – and why cohesion is crucial!
In this tornado of criticism, artists are not alone. We made errors in level design that impact the game experience quite much. Here is the main problem we encountered in terms of architecture:
Above: you can’t read the scheme? It’s normal, the level is too long!
Every level is too long. In fact, each one includes several puzzles to complete before reaching the exit. That structure enters directly in conflict with the defeat condition: if player’s stock of lives reaches 0, he/she has to retry the level from the beginning. As we designed a puzzle game before a platformer one, the player wants to test different solutions, making errors to find the “eureka!” moment. Unfortunately, it’s frustrating to be obliged to retry parts we have already finished because the architecture doesn't fit an elementary rule! Here is a correction I made on the scheme above:
Above: the same level with some important cuts.
By using the experience we had with playtests, I can now consider one level can’t be longer than what you see on this draw. It’s approximately 1/3 of the level we put in the prototype, and it represents more a macro puzzle than a long level to explore, as ‘Splosion Man or the last Donkey Kong could give as game experience.
Die to stay Alive is a cool prototype, and I took a great pleasure developing it with my friends. Even if there are a lot of problems in the delivered experience, it has the potential for creating mad and affording puzzles if we take time to fix its rookie mistakes.