EPTIt's semester vacation and in order to not fall into the depths of procrastination, I've decided to take on a little project for the next couple of weeks.
Starting on the 15th this month, I will take about four weeks to brainstorm some ideas and fully implement them in a working game prototype. I will post a blog with progress every working day (I won't do anything on weekends) and by the time the thing is finished (or the time is over) I will upload the prototype for you to play.
Besides outlining what I've done to the gameplay itself, the blogs will focus on the technical aspect of development. You will probably see some class diagrams or links to slides or PDFs that explain some technical concept that I'm going to apply.
Motivation
I've been a hobbyist game-developer for quite some time now. I think in 2007 I finished my first game, which has since been long lost and lies in some HDD graveyard among other broken hard drives of mine. It was a simple breakout clone with some extra features.
Since then, I haven't really finished any games. I always discovered some technical challenge related to game development, implemented it, and that was basically that. I've written maybe a dozen small "prototype-alphas", meaning that I would implement one specific mechanic to see if it would motivate me to work on it for a longer timespan, notice that it didn't and be done with it. Then I've written a couple of tech demos and two frameworks/engines.
After all this time and work, I have nothing to publically show for it. I have accumulated a decent amount of important knowledge, but none of my works are online and free to see for people who are interested in it.
That's what I want to change.
The main goal will be to have something to put on a website and say: "This is mine. I hope you enjoy it and have your 10 minutes of fun with it". That would be great. I want to start building a portfolio. I have bigger goals for games (and had them for a long time now), meaning emotions I want people to feel or philosophical ideas that I want to present, but that's none of my concern right now. The focus is on taking an idea, implementing and polishing it to the end, and then having something to show for it.
The schedule
There's no real schedule. Since I already have the basic gameplay idea that I want to implement (more on that in the next section), I will devote the entire 15th to think about what it is probably doable in these 30 days, cut half of it since programmers always overestimate what is doable, and then flesh out the details of those ideas.
Then, I'm going to pin down what kind of systems I need before starting, and set up the basic infrastructure that is required for me to prototype the game. I already have all technical systems that I'm going to need in a little engine that I wrote a couple of months ago, so most of that work will be to create a couple of new folders in Windows and rename a couple of files.
The Game (oh, yeah, you just lost it)
The basic core of the game that I want to get working first will be simple, one on one, player vs NPC swordfighting.
The player uses their mouse or one of the analog sticks of their gamepad to directly control the weapon in the hands of the player character.
The player reacts to swings from the opponent by using the mouse / analog stick to swing their own weapon against the weapon of the opponent, and swings at weak spots in the defense of the opponent to do damage. The goal is to exploit weaknesses in the opponents fighting behaviour and trying to get in hits without taking any yourself.
The idea is that every movement of the mouse or the analog stick on your gamepad should be reflected as a movement of the sword in the game. If you move your mouse fast, it will do more damage, but the resulting momentum of your weapon will make it harder to swing it back in the other direction, exposing a weakness in your defense that can be used by your opponent. Similarly, making slow and careful movements makes it easy to keep your weapon under control, hence providing optimal defense. However, it also makes it hard to do any damage.
I want to make the combat feel quite fast, so it isn't going to be easy. Optimally there would be enemies who use a certain fighting style, and then you, the player, has to learn and get to know that style and quickly react to movements of your opponents weapon. Optimally it would be akin to learning how to counter builds in SC or SC2, where you can theoretically react very quickly and successfully if you see some build for the first time, but you will be able to counter it a lot more effectively and with less brain power required if you play against it a lot. You become more badass over time.
This will of course immediately introduce some elements of rogue-likes, since learning something will make you die a lot, but I haven't throught that through that much yet. That's what Day 1 is for.
The game is played from the top down perspective, by the way.
From a game-development perspective, the primary challenge of implementing this combat system will be to figure out the exact control scheme. Does the sword automatically swing back to it's default position if the player doesn't move his mouse? Or will it remain static where the player moved it last after he doesn't use the mouse anymore?
Apart from that, the only important point is that I'm going to spend about 0 time on the graphical representation, since it will be a prototype only. The player character and the enemies will likely be colored circles, and the weapons will look like thick, colored lines.
Maybe, and that's a huge maybe, if I get everything working a week early or so, I might spend some time creating rudimentary character and environment art, just for the sake of spending some time on it (I can't draw at all). But even then, I'll probably rather spend the time polishing gameplay more or introducing new elements, like playing around with different weapon types or something.
What I'm starting with
The engine that I'm using to write this thing is self-written and pretty powerful. It's for 2D games only, but still it is capable of some advanced features that you rarely see in 2D games. I originally wrote it for another game prototype that I wanted to develop, which would've been a stealth game, so my engine now supports real time lighting and shadows, custom shaders, fully integrated TILED-Map-Editor, scripting with LUA, basic AI things like pathfinding and state machines, and some other things I can't think of right now. None of which I'm going to need for the prototype that I'm going to make now.
The engine is still useful though since it provides me with basic gamestate and resource management, as well as a simple to use entity-component-system for every object that somehow needs to be represented in the game (I will go into details of this on day two, I think, since I need to make a couple of changes to that system before I can start programming the game), and a lot of bonus stuff that I would otherwise have to do from the ground up (collision detection, math stuff, etc.). Most importantly though, I'm already very familiar with it (since I wrote it), so I won't have to spend time to get going with some other engine.
Oh, the whole thing will be written in C++ and the engine is largely based on the groundworks of the SFML library.
If you've made it this far
Thanks for reading! I see that you are probably interested in this and I hope you will enjoy my future blogs, too!
