First Design - First Mistakes

This posts is again attended to anyone that wants to go a game development as a hobby or even a indie, and this is just a piece based on my own experience.


Almost 1 year ago, after deciding to make a game of my own, I have started addressing my design. I had tons of ideas so I started to write them down...

Make a Game Design Document

That's something that will seem obvious for some, and useless to others. Writing all of your ideas, sort them and see if everything fit. This can be used as a first filter for not-so-good ideas.
Moreover, if someday someone joins you in your development, it will be a key asset to share the concept with him/her.
So I knew it was something really important and I made one...Kind of...

I know it may seems waste of time,
but it isn't. I promise.

Detail your Design document as much as you can

Unfortunately, I was not patient enough. I wanted, as many devs I've encountered, to jump into making things as soon as I could. So I did it, but actually it was a bit useless as it was not detailed enough and was basically just a summary of all my ideas, or draft of ideas. So when I had to start coding, I had to figure out how I was going to do things. I was then focusing on programming one feature at a time, and I probably lost the big picture.

Think features as part as a whole

One other thing that feels obvious when told but not so easy when actually making things is indeed to always think to how your feature will fit in your game. Several times, I've coded a feature as I have it in mind, and then had to change something or even rewrite all code so that it could really be used.
And after a while, I just couldn't bare anymore wasting my time...

Re-roll, change, kill...

So after several fails, I just had to face it : I needed to clearly put my design down on the table and do what I had to. For each idea, I asked myself : "Okay. How am I going to do that? What would it look from the player perspective? What exactly will the player experience be?"
Then, of course I realized that most of my ideas were good... But doesn't really fit each others. So I had to rethink the whole thing, kill many useless features, add a few critical ones.
As Antoine de Saint-Exupéry stated : "You know you've achieved perfection in design,not when you have nothing more to add,but when you have nothing more to take away"

Yeah I know... But you have to do it.

It took a while, but now my design is consistent (I think!), looks like a real game now (at least on the paper and the developed features) and not a clumsy patchwork of features anymore.

So here are again the obvious pieces of advice I would give to anyone wanting to go for an hobby or even an development :
-Write down your ideas
-Write a very detailed design document
-Don't be afraid to adapt things, and kills useless features

Bonus : latest mistake to date

Even when you know what you are doing and why, you may still commit mistakes. My last weekend experience can tell :
In my current game, I have a loot system where dropped items stats are defined automatically by player level. And I just figured out that my dumb formula I use to calculate these stats was an exponential one : at really high level (150+), I had things like a gun making 555433345 damages for each bullet....and around 500 bullets fired in a second...
Even if enemies stats are also increasing depending of player level, that seems weird for the less. I could have put a level cap as numbers were okay up to level 50, but I don't like it, and that doesn't fit the philosophy of my game which is basically about grinding.
So I have to design a new system I can tune the way I want. More work, and more wasted time. But that's okay as long as it should make the game better.