Hi, Erik here. A solo game developer doing his ‘thing’. Currently I’m writing this on the evening of my weekend. For me writing this article is a fun way to get me thinking about all the things I accomplished the past to weeks. And for you a way to keep notice of the interesting things I come up with.
I learned from my mistakes
My ‘asking for a buddy’ a few weeks ago is still influencing the things I do. I did some more talks, and this time it was about my way of prototyping. The designer pointed out I was going too fast and not ironing out the necessary things needed before you go to production. With his background coming from EA he said ‘game companies do this all of the time, no hard feelings’.
But I realized that I was making the same mistakes I made with ‘Find the Gnome’ back in 2017/2018.
Let me take you back in time. Back then I had some things worked out, like using gnomes and doing hide-n-seek with funny mouse interactions. But that is just a very small part of the game. I had not looked into level design and story design, and how these would amplify the funny game mechanics. But I started production, happy that I did find some kind of core game loop. So I ended up working myself into the corner, not being able to recover from it. Eventually I came at a point in the production that if I wanted to make the game better, I had to start all over with all the level design, assets and story. But 75% of the time spend was already in those parts of the game…
(Interesting detail: If you play ‘Find the Gnome’ you see that there are some levels ‘under construction’ still to this day. That is where I gave up hope.)
Back to current day. With ‘Manage the Universe’ I have everything still open. I can still make a great game out of it. All assets that are created currently are technical/mechanics only and can easily be dropped or repurposed.
So I made a whole new plan, this is for the prototype phase only:
If you want to know a bit more about how I salvaged the progress, see this in depth article. There I explain what the end-product of the prototype phase should be, and from there on you can work back and create steps on how to get there.
Let me introduce ‘Actionblocks’
Another experimental technique I am currently using is an ‘actionblock’.
What is the official idea: it is an isolated ‘main’ game mechanic, but with enough added small helper mechanics so that you have a mini game around this ‘main’ game mechanic. Make like 40 or 50 of these, and from there on you can easily play with different mechanics and mash up a few of these ‘main’ mechanics and find out if an interesting core game loop emerges.
I did add my own twist to it. I want my game to be more component based even during production. So I see use in this as a combination of prototyping, game-play testing, and component testing.
So, in practice. For a new mechanic I first create an actionblock and work out the details of this mechanic, and then add it to the core game loop. But the two are interconnected in that I use clever interfaces. So if isolated correctly I can than easily tune the specific mechanic in its actionblock without having to fire up the whole game. And preferably do some automated testing with it in the build pipeline.
Actionblock ‘tech tree’
A big portion of my previous 4 weeks went into this actionblock thing, and specifically the ‘tech tree’.
During working on the actionblock of the tech tree I found out I was creating all kind of production ready UI elements. So it was then that I decided to get back to prototyping. Hence the prototype look:
It is all about the tech tree, the additional stuff is there to extremely simplify the remaining game loop. My current goal with this actionblock is to determine how the tech tree interacts with you while doing your usual stuff of exploring the world. It is still a WIP, the idea is that you find items with technology on it and that this unlocks your techs. But sometime you can’t unlock tech because you have to advance other parts first before you can do ‘better’ item finds, and with this actionblock I am trying to work out a good way to get this interesting.
GameFeelings 