Blog for Drew Hitchcock at GarageGames.com

Turn-Based Strategy (again)

Drew Hitchcock — 2006-11-22T18:51:25+00:00

(Click for larger image)

For some reason I really like turn-based strategy games. After abandoning my last one, I'm back to working on another for a school project. Luckily, this has been a fairly straightforward project because I've solved a lot of the hard problems in other projects. I've been working alone, but I've still made a lot of progress.

The new game is a turn-based space combat game that is playable over TCP/IP with 2 players.

I made a choice to use Python for the whole project, even netcode, rendering, and other core engine stuff, but eventually it became pretty obvious that I needed to move to a better rendering solution. I had used pyOgre in the past with decent results, so I decided to give it a try again. It was a bit tedious to integrate the existing code with the new rendering system, but the change has been worth it.

(Click for larger image)

So far everything is going pretty well (except for the programmer art). Multiple players can connect to the server and move their ships around in turn. Combat works to a certain extent, but there's only one kind of attack, and there aren't any kind of particle effects or other visual feedback, so it's pretty bland. I have a few weeks left before the thing is due, so I believe I should have enough time to add those things.

(Click for larger image)

The space combat theme seems a little bland to me, so I don't know if i will continue this after the semester is over. I might change the theme and continue to work on it, though, because it essentially has everything that I would need to make a network-capable turn-based strategy game, and it doesn't make sense to throw that away.

EDIT: Every time I post I screw up images in some new, creative way

A fairly uninteresting update

Drew Hitchcock — 2006-07-23T21:17:29+00:00

It's been a really long time since I've posted anything, and my last plan is rediculously out of date, so I thought I'd post a little bit and damage all of your eyes with some programmer art at the same time.

The strategy game design I was working on in my last plan wasn't working at all. The game really wasn't very strategic, was entirely based on luck, and wasn't any fun at all. In addition, the random terrain didn't feel like it had any connection to the action going on in the foreground, and it generated rediculously boring levels. So I decided to scrap it.

Recently I've been trying out little concepts, but I haven't settled into a real project yet. The screenshot above is just a result of some tinkering with TGB. You can fly the rocket around and pop hot air balloons. You only control the rotation of the rocket, so it's pretty tricky to actually hit things. Not very exciting, but I liked the screenshot.

Anyway, I don't have any kind of project going, so if anyone needs a scripter, or wants to work out a code-for-art type deal, let me know. I would obviously be providing the code half :p

Danger! Programmer Art!

Drew Hitchcock — 2006-02-04T07:56:59+00:00

Large Version

I've prototyped the tile-laying and hand/drawing mechanics for the game I've been working on throughout my last few plans. This screenshot shows the player's hand (across the bottom of the screen) along with a few tiles that have been placed on the map by the player.

I haven't really provided a good description of my design yet, but in an extremely general way, the game essentially will involve placing tiles on a map to influence the activities of some AI-controlled factions, which operate in real time. So in essense it's a real-time tile-laying game.

Now that the tile-laying mechanics are more-or-less in place, I can begin working on the real meat of the project, the AI. The AI works at two levels, the low-level unit level, and the high-level faction level. I've never tackled any kind of high-level AI, only low-level things like path-finding, so this should prove to be a pretty interesting challenge.

EDIT: Linked the large screenshot and inlined the medium version.

More T2D Terrain Stuff

Drew Hitchcock — 2006-01-25T06:27:46+00:00

I haven't been able to spend as much time with my random map generation stuff as I would like, but the progress I have made is fairly visible, so I figured I'd post some screenshots. Everyone likes screenshots.

In my last plan I showed the quick proof-of-concept that I threw together to test out my idea. In the first iteration, the terrain textures were blended together in software to make a single texture for the map, which was then slapped onto a quad and rendered. Obviously, this approach would use a lot of texture memory if I wanted to make a very large map that had a lot of detail.

In the latest version, the textures are blended in hardware according to an alpha map. The alpha map is still occupies some texture memory, but much less due to the fact that it's only an 8 bits per pixel image. The dimensions of the alpha map are also much smaller than the map itself, due to the fact that it only contains blending information so it doesn't suffer in quality as much when it's enlarged.

You'll also notice that the terrain textures are now tiled, and have smooth transitions between them.

I'm going to expose some of the generation parameters to script tonight, and do some cleanup of the C++. After that I'm going to put it aside for a while and start working on actual gameplay. If there's one thing I've learned from all of my past projects, it's not to get caught up in one detail too long. An unpolished game is more entertaining than a polished 2d map renderer.

Edit: Closer picture to show detail

Random Terrain in T2D

Drew Hitchcock — 2006-01-19T08:27:26+00:00

It's been a very long time since I've posted a plan or in the forums. I honestly haven't done much worth talking about during most of that time, but recently I've been getting back into the whole game development thing, so I figured I'd post a plan. I think I'm going to try to start posting plans more regularly, you guys have been pretty encouraging in the past, and I'm hoping a little blog post now and then will help keep me motivated, and will help me keep track of my progress.

Anyway, enough blathering. I've been throwing around an idea for a strategy game using T2D, and I was experimenting with some different ways to handle the map rendering. Tilemaps are the obvious solution, but I've found that it's difficult to make nice-looking, arbitrarily-shaped terrains with rectangular tiles without expanding the amount of tiles that my (nonexistant) artist needs to make by a rediculous amount. To complicate things, I want to be able to randomly generate maps on the fly.

The main problem that I ran into was that my random terrain generator would generate tile groupings that would require some pretty insane transitions in order to look right. It worked fine with ugly, blocky maps, but as soon as I tried to smooth out the edges with some transition tiles, I realized how complicated the problem was.

So I decided to switch to a more interesting approach, abandoning my script-only solution for some C++.

A quick proof of concept:

Basically, I'm using libnoise to generate a random square of coherent noise, and blending three textures together according to the output. It looks really grainy right now, and it's not configurable at all from script, but I'm definitely encouraged that I could get this result so easily.

So far I'm really pleased with T2D. It seems to be able to do a lot from the script side, and it also seems to be very easy to modify the C++ source.

Saturday Feb 19 20:46

Drew Hitchcock — 2005-02-19T20:46:35+00:00

A new screenshot and a basic description of my idea.

Here's a newer screenshot of the game I talked about in my last plan.

Based on a couple of the ideas I recieved last time and some of the ideas that I was throwing around in my head before my post, I've come up with a general plan for the gameplay of my game.

Basically, when you line up boxes of the same color in a straight line, they create a sort of laser-barrier between them which blocks the movement of the AI and the player. In each level, the player attempts to dodge or trap the AI, open up the level's exit, and escape. All of these things are accomplished by moving the boxes around.

Right now (as you can probably see from the shot) the boxes can be pushed around, and the barriers are correctly created between them. Next I'm going to attempt to get some "AI" opponents together and to implement things like death and level exits.

By the way, the art is all placeholder art. At this point I have no idea how I want the game to look :)

Saturday Jan 29 22:23

Drew Hitchcock — 2005-01-29T22:23:49+00:00

Some experiments with Ogre and a problem

My main project seems to be on hold at the moment, so a few days ago I decided to give ogre a try. I had no real intentions of making a game with it, but after a few days of learning the ins and outs of the engine, I actually have a pretty good start on one.

My overall impression of ogre so far is pretty positive. It's not a full game engine, of course, but it's pretty powerful and I think it's a great choice for simpler games. It's also pretty quick to learn, provided you already know C++.

Now I have a serious problem, though. I have a pretty good start on a game, but I have no idea what that game is. In other words, I have no design, and no idea where I'm going with this.

I don't want to start something new -- that's what I ALWAYS do -- I would much rather use the pretty good base I have right now and do something cool with it. I also need to figure out what I'm doing pretty quickly before I lose interest and make 1/4 of something else. I'm sure you guys can understand what I'm talking about.

So if you have any ideas about what I could do with this feel free to post them. Right now the capabilites are similar to sokoban, you can push crates around, and movement is constrained to a grid (although the movement between squares is smooth)

I was thinking of adding some patroling enemies and making a kind of stealth game, or maybe adding some color to the crates and doing some kind of warehouse-clearing-color-matching game. Who knows?