Progress! Also, there is a Fullscreen mode, and a Tutorial, too.

So, yeah, this game’s based on an idea I got some times ago.

The idea is that you’ve only one big pixel to know your surrounding.

Currently, I just have to put sound, window title and actually implement the “easy mode” which will print the pixel value(r,g,b)

The maze is operational, has a minimum look, and you can walk around in it.

I have a very busy weekend, but 26 is my favorite number.  So, I’m going to squeeze in a maze game.  I’m using Unity3D so it should not take too many hours.

So I let my AI roam my procedurally generated maze. Check the video!

The weekend wasn’t on my side sadly but I’m here and aiming for the jam.

In the last hour I have made a tree with the pathways in the maze – levels are procedural ofcourse. Here is a picture of the tree visualized.

Now, on to my first ever legit AI in my life: a simple DFS.

I made a special maze generation algorythm to coop with my first attempt in AI… and I slept all day… BUT I’m on it now!

As we get close, I feel it is a good idea to post an update to my maze library. I’ve built a tool which will help people create mazes.

There’s only one path from start to end and it runs quickly. The points are 1 based (instead of 0 based) with 1,1 being the upper right hand corner. There is an outside border, so the actual size of the arrays generated are 2 spaces bigger than your specified size.

I’ve built a generator:

See this post for the library:
http://www.ludumdare.com/compo/2012/12/11/maze-creation-library-testing/

Here is the application, requires .Net 2.0:
http://www.kins-home.net/ludum/timelapse/MazeGen.zip

So I continue to test my library before release tonight. Does it work with small mazes?

Sure does.

How about medium ones:

Great. But let’s get a challenge going.

A 76×120 maze:

Click to enlarge.

Looks good to go.

URL: http://www.kins-home.net/ludum/timelapse/MazeCreatorLib2.zip

Quick level generation.

I’ll be using this in my next project.

Oh and the times here are based on the debug build on a VM, they improve on real hardware using a release build. For example that big maze can be created in under 830 ms.

Hi!

In the middle of the bad things I’ve been through the weekend. I decided to keep my game simple and tought that I could make a game out of my work. I’m alone at the night every thay and keep telling Windows that he should work becouse he used to. But I have leaved out the programming and stuff like that and added the fun parts and some problems to solve and watch out for.

Basically you sit in front of the monitor and try to find a way out of randomly generated 2D mazes to get score. But you have 4 stats – awakeness, stress, hydatation and bladder – as well to watch out for. If they reach their limit than you loose – fall a sleep, going mad, get dehydratated or be in shame at work place. These stats can be easily controlled with a mouse click. You can make coffee to drink to stay awake or water to get hydrated and you can go to the toilet.

The problem is their effect on eachother. Becouse if stress go high, then you are getting tired quicker and start to see hallucinations that couse more stress. Bladder goes up from water and coffee too, but you need to drink to reduce stress and to avoid dehydratation. And while you can make more coffe, you can’t replenish the water resource.

Hallucinations are coming when you are doing your job, but gets more intensive when you go alone to the toilet or the make coffee. So watch out. The ghosts are building up your stress…

The game made entirely in Processing becouse I needed to learn it. So This was my first time I’ve wrote a line of code in it. Sadly other things have gotten in my way, so the game is about half finished. It need some features and some graphics. There isn’t even support in the code for sound – I’ll add it tough. But the main engine is working pretty well.

This concludes to myself that I’m f***ing awesome and have to work hard on things like game making. Instead of the DAQ programs for foundries and the likes…

Anyway have fun with it. The source is up with the release. Feel free to use it. To play the game, click on the first screen, use wasd to move in the mazes and try to go out at the bottom as many of then as you can. Click on the coffee or the water to drink. If you run out of coffee, then click on it once more to get another one. (This minigame isn’t working yet!) To go to the toilet, click on the Bladder stat. (This is a similar minigame and it’s in the state like the other with the coffee…) There are no scores, but I wanted to do it like “how many mazes in wath amount of time, but maybe there are more possibilities than this. I’ll consider them too.

Here is the entry: http://www.ludumdare.com/compo/ludum-dare-22/?action=preview&uid=5167 Try it out!

My game after 24hrs!

After 24hs, I managed to mostly complete my game engine. There are a few features missing, but I got all the mechanisms that I need to make my game:

• Controls are all in.
• Victory and Defeat conditions are in.
• Level Selection and Unlocking are in.
• Reading levels from text files is in.
• Placeholder graphics and animations are in.

This means that tomorrow I can mostly concentrate on adding levels to make the game fun/challenging, and adding bells and whistles like images and sounds. I feel hopeful that, for someone who has never made a video game before, I will not embarrass myself

My code, on the other had, is a complete mess. Hardcoded stuff all around, hacks to make the player and the world talk to each other. Very few comments. Objects and world tiles mashed together. Ugh Ugh. I must say that these first 24hrs of LD were great for me to learn some lessons in programming.

Anyway, here is the game. There is no content, but you can get a feel of how it works. You are an explorer, who has to find your way out of mazes. You can do it by yourself, or help people stuck in the maze. Some of them will give you powers, but they will also get in your way, and eat your food. Sometimes it is better to go alone…

Maze Explorers!

PS: If anyone could give me some hints of what exactly I should put in my manifest to make java skip all (or some) of those nasty warnings, I would be very very happy

Just releasing a quick library I plan to use in Ludum Dare, consisting of a maze generator and an implementation of A* in java.

https://github.com/xSmallDeadGuyx/SimpleAStar

I’ve got a vague idea on the story and content for my game, but first I have to make a simple maze generator from scratch. Here’s the debug report from my path making tool, which is the first half of my maze generator.

To the left are the current visuals of my game, which so far comprises of an awesome player controlled circle! Truly a wonder to behold.

Baby steps

Hi, it’s been a while since i sit down and do a game myself. So, I really hope i can finish this one in time;)

yue

All done! Already filled with lots of user submissions. I can already tell that making sure submissions work right is going to be murder. But we shall see! It’s pretty awesome already, with multi-part puzzles made by different people.

Play now: http://hamumu.com/maze.php

It requires a Dumb Account to play, but those are free and the most fun thing in the world, so you would be a fool to pass that up.

Well, I spent day 1 playing WoW, watching TV, and trying to come up with an idea with no success.  Now it’s day 2, and things are coming along really smoothly:

There’s more functionality done than you can see in the shot.  You can move around the (two-room at the moment) maze.  If there were items, you could use them, drop them, and combine them, but not pick them up yet.  So as you can see, there are doors leading off into space.  If you go that way, you actually stay in the room you’re in, but an edit section will pop up on the screen for the room you tried to enter.  I haven’t done that yet, but you’ll be able to make your own description and layout there.  Then that place is marked “pending” as it’s emailed to me for verification (to check for language and such).  Pending areas can’t be built in or entered, and there’s a “nifty” graphic indicating them.

This is going to be a sight to see!  I can see great potential for impassable situations, but that’s all theoretical right now.  If I didn’t include items and just had riddles, we’d be good, except for one thing – I just know shortly after startup, we’ll end up in a state where all possible paths are blocked off.  But I can always manually throw in doors if that happens.  I could just verify that there is always at least one door leading into the void, but that’d be a very heavy database churn to do.

This was my entry for Ludum Dare #7, which was the first LD I entered. The theme (growth) eventually gave me the idea of growing a maze.

So, you create the maze as you walk around inside it. When the game begins, the maze is just a set of disconnected squares. Each of these squares can be linked with a set number of its neighbours (how many depends on the square, from none to four), and you create new links by walking from one square to another where there’s no previous link. Once a link is created it can be walked on as much as you want, but a link can’t be removed once created, so you have to be careful when creating your maze so you don’t get stuck.

Once I had that working the deadline was looming close, so I threw in some keys and locks and made the objective to clear all locks of each level, to make the thing resemble an actual game. In the end there was four levels, a random level generator, and also a level editor.

I wrote in the original README that I’d continue to work on the game, something I haven’t done. I still like the general idea behind the game, but it has this tendency to degenerate into just staring at numbers, which isn’t very fun at all, and on top of that it’s easy to get stuck, having to restart the level if you don’t pay attention. Perhaps some of the extra elements I didn’t have time to put in the game for the compo — more tile types, powerups, bombs, enemies — would have made it better (more varied if nothing else), but I think the interface is the main problem. It should be more obvious what tiles can connect, how many exits are left, etc, so there’s less guesswork, no number tracing, just puzzle solving. Since the levels are so “dynamic” getting that to work would be tricky, though.

Download: [ Windows | Linux (x86) + source code ]