What Happen?

Why isn't Arity finished yet? Why haven't there been any updates? Has SLG dropped off the face of the earth?

These are all valid questions with one simple answer (and no, someone didn't set up us the bomb).

I'm working on something else. I've got an awesome composer (unless you don't like trance) who also happens to play video games. He's played a lot more of this type of game than I have, so he's also got a lot of useful knowledge and makes a great co-designer. I think that once I get it all laid out, the engine will be fairly simple to make. I think the gameplay might not be entirely original, but I think it will be fun. I think that this game will also see the first iteration of L.A.U.R.A. Tomorrow is Bright!

In celebration of the new and exciting project, I've take time off from doing homework instead of working on the prototype engine to fix up the blog layout (the old one was really ugly) and the fix up the logo for the blog (the old one was really ugly). I might as well also mention that I made a new and slightly more interesting background for the main site a month ago. Get out your text highlighting-enabled browser, because that old broken down projector does a pretty poor job of making the text stand out from the background (EDIT: That's because apparently Chrome doesn't "do" FixedSys).

THEE FYTIE OV THEE FVTVRE

I'm still alive

I've just been spending all my time doing science and programming instead of writing blog posts. But I'll make it up to you, internet void. Here. I haven't posted this anywhere else yet. It's a platformer controls prototype with procedurally generated cave-like terrain (using ChevyRay's algorithm). Controls are wasd to move, s to crouch, w to jump and to climb up nearby short walls, and right click to boost. Let me know what you think (or not since the internet can't talk).

Edit: When I say "what do you think," be honest. I mean it. Be as brutal as you need to be. I am aware that the controls might well be terrible. The goal here is to learn more about how to implement interesting controls effectively, and that means cutting out the ones that suck.

II

Root's Eye View

Seg Fault

Different Raycasting

I starting reading flatland and it inspired me to try making a 2d -> 1d raycaster. If its current for, it uses thickness to communicate depth, but later on I'd like to use 3d glasses (inspired by gamma 3d) to convey depth. This could allow for first person 2d games and I'll no doubt pick it up again later for that purpose. For now I'm trying to keep clear for Tixel when programming starts on that.



For now I'll just hand out the source with a Linux binary.

Source

Page Up/Down - angle
Right/Left Arrow keys - Forward/Backward
Up/Down Arrow keys - Straight up and down

Raycasting

Same cave, different framerate:



Each triangle is like a flashlight. This renders at 4-6ms per frame instead of 100-130ms per frame, plus it adds realistic shadows and lets me easily change the widths and direction of the light.

More

This one uses an actual randomly generated cave segment. It was actually created in several parts in order to test the ideas behind the algorithm. The whole cave would have been 20480x11520 pixels (1280x720 * 16). I only generated a 1280x720 pixel segment of that, and then used a 400x400 pixel segment of that for the lighting.



This is the whole finished cave segment.

Even More Cave Lighting

And this time it's a video!

More Cave Lightning

Later versions of this will use ray casting instead of checking every pixel. This will enable faster rendering and shadows (1600 lines vs 160000 pixels).

Cave Lighting

Frame rate is really low, but it works.



I'll see about the promised Death Geode stuff later today. No promises, though.

gearGen and Linux

I finally got around to filling that empty space on my hard drive that I've had sitting around for six months. I filled it Linux. What this means is that in the future (if I can just figure out how to use KDevelop), I will be able to offer both Linux and windows versions of my games.

It took me about two hours to install Linux. I then spent 8 hours fiddling with xorg.conf and rebooting over and over again before I realized that I could just turn off overscan on my monitor. After that, I'm pretty sure Linux is actually more "user-friendly" than either Windows or Mac. It certainly seems to have a lot more fancy graphical stuff going on. And it seems you can download and install just about anything by typing "sudo apt-get install [program name]." I mean anything. "sudo apt-get install pi" downloads and installs a program to calculate pi.

Anyway, in celebration of my new operating system, and because doing it on a calculator was really tedious, I decided to write a good old-fashioned command line program. It's called gearGen and it prints out x and y coordinates for drawing gears. Those coordinates can then be converted to images using paint or the GIMP. This ought to help a lot with making sprites for a certain game. It turns out that just opening up paint and drawing a 32x32 gear from scratch is not at all easy.





gearGen code: (C=C+1)

Hard Drive Update

Well, I've run a few diagnostic programs and I now know what data can be saved (the important stuff, it turns out), and what sector to stop at. I've got a new 250GB drive on the way, so when that arrives I'll see if I can't use some ghosting utility to copy the first 60% of the failed drive over to the new one, copy the data from the new one over to my backup, and then backup to the new drive.
I'm holding off on Contrecoup for the moment because I don't have 3d glasses yet and I'm still working out how I want to do the collision reaction (kill the player, randomly spin the player, push the player away, etc).
Meanwhile, I still haven't done any work on Purity, but I might do a little of that in the next couple of weeks.
Right now I'm getting ready to prototype a boss fight for another game and maybe work some on the code for the O.D.d.S.A.C.k.S. More data on that game when I release in ten years or so.
The other thing I'm doing is honing my programming skillz on Project Euler. I've finished 5 of the problems and right now my computer is around prime number 1.5 million for another. Maybe I'll go back later and figure out how to solve that one in 1 minutes instead of a few hours.

What you should take away from this, dear internet void, is that I'm still working on making video games and I might even finish one some day.