So my game currently revolves around you being a soldier POW "caught" by the opposing army. Is that reaching too much, or does it match the theme well enough? I have other ideas but they are all pretty much variations on this theme...

Another idea I had was to make the player character "caught" between both sides of the war. Not quite sure how to pull that off.

Anyway it's looking to be some sort of rpg of sorts so far.

I made some progress. What I hate about game development is how hard it is to get a feeling for where you are. You can program various routines for hours and hours, and feel like you have got nothing done. But then, you keep adding a little bit after a little bit, and suddenly that feeling starts to show itself. Up until a few hours ago, all I saw was random code bits and some mockup art. Now, while technically not much has changed, I do start to see the bits and pieces of a game, or at least the faint shimmering silhouette of one.

With some projects (in fact, the majority of them) I don't really reach that tipping point. The number one reason I have for canceling a project is that I never get that gameplay I have in my imagination to emerge from the bits. I'm not quite there with this one yet. But I have hope.


Here's a screenshot of the current battle engine. It might be hard to see what's going on, but basically all of the army guys are aiming at me (I'm the identical army dude in the upper right corner), and I'm about to die. It's a turn based system, with the option to basically move, aim (pick a target), or shoot at them.

In this scene all of the enemies are armed with knives (because melee ai requires just a bit more work), so they follow you around if you move and try and get in and stab you. The turns for combat works reasonably well, and the interface works great.

One other feature, is that the game transitions between exploration mode and battles fairly seamlessly. Enemies will all be in the same positions they were on the exploration screen, and the scenery will be the same as well. The only difference is the spacing in battle mode is more rigid, while exploration has pixel perfect movement. It's sort of an anachronox/chrono trigger ish system.

Upcoming tasks: map out story, figure out character stats/skills, add randomness to combat, and then take a break from that to do some iterations on the art. After that is just build build build.

So figuratively today was very hit and miss. I DID a lot, but I don't know if very much of what I did actually moved the game towards completion status. So I feel like I exerted a lot of life energy without much to show for it. I don't think anything I did however could have been left undone, so I don't quite see myself going down the wrong road, which I guess is good.

I feel like there are basically two different types of development. There is the creation part, where you are actually implementing new systems, or rewriting significant algorithms. This is what I find the most fun. You sit down with a blank canvas, and an hour or two of dedicated hacking later, it DOES something that it didn't do before. I HAVE MADE FIRE, HEAR ME ROAR.

The other part is refinement. It's a lot of tweaking and integration. You take one of those pieces you built in step 1, and hammer it and shape it endlessly. I often find myself hammering these pieces for very large amounts of time, and not really enjoying myself doing so. In 10 minutes, I can make a character walk around a screen. It starts with nothing, and suddenly we have a cool little demo to play with. But then to figure out how fast he should move, what kind of collisions to use, etc, suddenly it turns into several hours of pretty boring stuff.

I think I will fare better if I spend a lot more time building, and a lot less time refining. My resolution at this point of the competition is to slap myself whenever I find myself hammering on something that is probably good enough. And if it's not good enough, maybe I should create a better piece instead of trying to make a bad piece fit.


Anyway, there is also the literal hit and miss I mentioned. One of the big components of my game is the hit detection during battles. You choose an enemy to aim at, but the actual detection of whether you hit them or not is based on line of sight. The angle you fire is randomized slightly based weapon accuracy and player skills. If the angle doesn't pass through the enemy, it is a miss. If you target someone on the far side of the room, but there are a bunch of people in the way, you're probably going to hit them instead of who you were aiming at. Eventually there is going to be cover to watch out for as well.


As you can see, in the right image, the shot was off target, and didn't connect. In the left image, it grazed the corner of the enemy and did some damage. Currently ai is pretty terrible, and they tend to kill each other off in a big group because they are trying to hit you :) I'm not sure how well I'll be able to overcome that, but I can probably get something that works well enough for what I'm going for. Since my game design has you fighting against hordes and hordes of enemies, even if they occasionally suffer friendly fire, it will probably still be quite a challenge to win.

Wow, how did it get to be day 6 already. I guess it's technically day 5 for a few more hours. I haven't made much progress, not having any time yesterday, but the core of the game engine is getting pretty close. I'm not as happy with combat as I wanted to be at this point, and there are still tons of little things to do programing wise. I haven't had time to do content at all - there is a good chance that this ships with an editor, some test combat scenes, an editor, and no real levels. on the other hand, since there is an editor, if I do get any time to build some stuff out, that part shouldn't be too difficult.

Most recent addition was equipable armor and weapons. I'm pretty happy with how their stats work. Guns alter your accuracy stat as well as setting the maximum damage. Armor has a coverage defining how likely a shot is to hit that armor, which mitigates some of the damage.

Note: Ignore the menu, it doesn't do anything useful, and on linux you may need to fix the line endings on data/items.txt

I think most of the bugs are gone. The game balance is waaaaay off. It was too hard up until recently when I realized that the player's statistics weren't getting copied to the fight scene (so every fight you were basically level 0). Also, there was a bug with the armor statistic in that other than equipped armor, the characters armor level (sort of like defense) wasn't being applied unless the shot happened to hit one of your armor pieces. Those two fixes changed the balance a lot, so it's probably way too easy. You also level up insanely fast, by the end of the (very short) game, you don't even need to aim and you can get a perfect shot every time.

I don't know if I'll expand this post pyweek, but if I do, there's certainly a lot of issues with the combat alone to work through.

Also I never got around to adding spells, which were a major part of the story. Oh well.