December 06, 02007

Worked on: GeometryStomp.

Mostly got distracted messing with the pipeline some more, despite the fact that what I have is already fantastic. Why is that so much fun?

Actual game changes:

• Enemy graphics are now generated at runtime through Flash's drawing API, then those prototypes are duplicated when instantiating the actual enemy.
• Framerate no longer set ridiculously high

December 03, 02007

Worked on: Haxe game.

Mostly the pipeline, making it a bit more organized. Also added scite to the game's folder and to source control.

December 02, 02007

Worked on: HaXe game, but mostly pipeline/tools for making games with HaXe+SwfMill.

So this weekend I played around with using HaXe and SwfMill to make swfs without having to use the Flash IDE. I wound up teaching myself a bunch more Python to get a pretty nice pipeline set up (well, the Python libraries really. The language itself is a piece of cake). I can now just drop an image in my "library" directory, and it will automagically be available for attachMovie'ing the next time I run my code.

And through the magic of SciTE, I even have a workable IDE for it. It's just missing some autocompletion.

Hooray Open Source!

The real question though, is whether taking all the time to set this up actually ends up with a better (that is, faster) process for making games than say SDL or standard Flash development gave me.

November 30, 02007

Worked on: Flash game(s)

Been messing with it for the last couple days. Problem is, I keep getting ideas and go try them before I finish the previous idea. This results in close to nothing getting done most of the time.

November 11, 02007

Worked on: HL2 Mod

For the past couple weeks I've continued to mess with Source Engine modding.

Although Hammer isn't very nice for authoring it, it seems that you can do wide open outdoor terrains without too much hassle in Source. Did I say without too much hassle? That was a lie. Authoring it is a real pain. On the bright side, the modding community has hacked together some decent tools to create the maps from heightmap images, but it is kind of a hassle and doesn't really allow for fast iteration on them. That's not really a big issue for me, though, since I'm fine with these bland landscapes for now. The other issue is that with a big open map, vvis (part of Valve's bsp compiling pipeline) takes a really really really long time. So to cope with that, I now have the player scaled down to 1/8th of the normal height, with speed and gravity adjusted to make it feel like the map is huge, rather than you are small. Seems to mostly work, but my maps so far lack anything that would really give you a sense of scale.

So at this point I still have to figure out if mods are something I want to stick with and finish. Listing out pros and cons just for the sake of having a list:

Pros

• Mod community available to help test or possibly even help create levels/models/code for it.
• Valve's nice tools for model and level editing and scripted events.
• Looking pretty. I get to steal a bunch of great textures, effects, models, and sounds basically for free.
• Totally sweet game engine already made. Writing my own animation system or physics engine, while cool, would be a tremendous waste of time that could have been put toward the things that would make my game stand out.
• Did mention all the stuff I wouldn't have to create myself?
• Seriously. I'll never find or make better grass/stone/metal/concrete/brick/etc. textures than those.

Cons

• Folks are going to at least need some source-engine game in order to be able to play it. Portal is like the best thing ever, though, so everyone who cares about games at all should go buy it now if they haven't already.
• The Source engine has it's downsides, too. Makes certain assumptions that may be very hard to try and change. Thus far, I've been able to see how to deal with the ones that affected my plans, but I get the impression that I would start hitting major problems if I tried to make a game out of some of my wilder ideas, like a character whose size can vary wildly form one moment to the next. Granted, it has been adapted for things as unconventional as Portal, but that was folks working full time at Valve. I've already had issues like having to go to 1/8th normal scale. That bodes poorly for doing off-the-wall game ideas in Source.
• Looking samey. Not that it is forced upon you if you try to use Source, but obviously, it is easiest to make a mod that looks a lot like the game you are modding, and will likely have a strong influence on whatever I try to develop with it.
• Won't play in a browser. The fact that Flash runs on practically any machine that get on the internet makes it terribly appealing as a platform on which to develop.
• Hard to show off. This is where the DS Homebrew excels. I can't just pull a laptop out of my picket and play a Half-Life 2 mod, and the installation is not as easy as just browsing to a webpage like with Flash.

October 31, 02007

Worked on: HL2 Mod

This was ridiculously fun. All I did all night was play with how the gravity gun worked. It got particularly great when I made a game of trying to find the most spectacular ways to kill myself. Stuff like dropping cars on my head, launching my corpse up to roof tops, that sort of thing.

Next up: I need to make some custom maps for it.

October 25, 02007

Worked on: Team Fortress 2 Maps

Setup was a bit complex and finding the right Hammer configuration took some trial and error, but I was able to get a map up and going in one night. Now that I've ironed out the Hammer problems map development could go much faster. But I'm not so interested in map-making as I am creating new games.

Note for future: Have another go at making a Half-Life 2 Mod.

October 24, 02007

Worked on: DS Homebrew

Added skeleton of RulesGame. (see RulesGameDesign). Modified launcher. Did NOT actually work on any game-specific stuff :(

October 18, 02007

Worked on: Flash experiments

• Meant as an experiment in gameplay involving lots of ambient items that you do...something...with.
• In the future I need a more focused idea that that before starting a prototype! (not that I didn't know that before...)

October 08, 02007

Worked on: CollisionResponseTestbed

• Fixed some asserts in SpriteManager
• unstacked the walls in their initial positions