That’s what I spent the last week doing – generating random maps and waiting for the generator to blow up.  The good news is that 99% of  maps are clean now.  There are a few sneaky bugs left, but I’ll catch them eventually.  The result is that random maps look better than ever now – take a look.

The other thing that I’ve whipped up last week was the replay system.  The game is getting complicated enough now that it’s hard to debug.  I’d be playing a random map and at some point the game would crash.  How am I supposed to diagnose the problem if I can’t make it happen again?

To address this, I’ve made the game record all input – mouse movement, clicks and button presses.  I’ve been careful to write the game in such a way that it’s 100% deterministic – if you do the same things to it, you’ll get the same outcome.  This means that I can now load my replay files and reproduce the problems exactly.  This will save me a ton of time and make the final game much better.

The question, of course, if whether this will make its way into the final game as a “record game” feature.  That’s not as straightforward as rigging a debug system and it might take me a couple of weeks to polish it up to acceptable levels.  I don’t know if I have the time to do that, but it’s on a “wanted” list.

A couple of days ago I finally made the maps have a random shape.  Here are what they look like now:

Another thing – I was so annoyed with the temporary interface that I changed it to something better.  The old one took up a large chunk of screen space and made it hard to take screenshots.

If you look carefully, you’ll notice some new elements on the map, besides aliens and humans.  I’ve been working on the campaign code, and that’s a part of it.  I’m going to talk about them more later.  In short, their purpose is to spice up the game and add some variety.

Another detail that you may have noticed is that there are no walls in the middle of the map.  That’s something that I’m reworking right now.  The old walls were all the same size and I want to add some differences in scale.  There are also some bugs in the polygon code, so I’m in the process of hunting them down.

Well, today I’m going to talk about a problem that I don’t have a solution to. Walls! I hate walls!

A couple of days ago I decided to improve the look of walls in the game.  I had a couple of ideas for this, but so far I haven’t come up with a solution that I like.  Let me show you some of the valiant attempts.

My first try was to add some texture to the walls, like I did with the background.  It’s a bit more interesting, but I still get a profound sense of blah from that picture. Yuk!

My next approach was to try some shading that I quickly threw together.  It helps a little bit, but I’m still unhappy with the look.  It somehow doesn’t fit.

I spent several hours fiddling with the shading settings, trying various combinations of smooth edges and sudden transitions.  Unfortunately, nothing really worked.  I think that I’m going to put the problem aside for now and work on something else.  This is probably the 3rd or 4th time that I’m shelving the wall work.

Hopefully I’ll get an idea that works soon.  Know any game that did top-down walls well?  Keep in mind that the walls have to be polygonal.

Some people have mentioned that I seem to be focusing on the game aesthetics a lot recently.  The reason is that I’m pushing to release a trailer, and I can’t do that unless the graphics are fundamentally settled.

What’s the big deal?  About a month ago, after I announced the game, I started talking about it to the gaming press.  The response was very muted.  Michael Rose from www.indiegames.com was awesome and interviewed me, but other than that nobody seemed to care.

My suspicion was that it was due to the game’s shabby, cobbled-together look.  At the time, I didn’t invest any effort into graphical improvements.  I decided that I had to get the game noticed early on, or it would disappear into the void of the ‘net.  That’s why I’m doing so much visual stuff lately.  Once the trailer is out of the way, I’ll switch back to improving the gameplay.

Alien Blood

I’ve been working on alien and building death graphics for the last couple of days.  That means large clouds of alien blood.  I’m actually a bit divided on putting this much blood in the game, but I’ll talk about that later.  For now, here are some exploding hives.

One thing that felt off about the game is that hives dying didn’t feel satisfying.  This was a major turning point in a round and it was flat.  I made them explode in a big way now, with red streamers and everything.

The other thing that was bugging me was the hive damage response when they were still alive.  They would show no sign that they were hurt.  I added squirting jets to show the damage.

Too Much?

As I’ve been working on all of these death effects, I’ve started to wonder if it all might be a bit too violent.  There are certainly reasons for concern.  Firstly, I’ve noticed that I’m trying to compensate for my lack of art skill with large amounts of red decals.  That seems a bit questionable from the art direction perspective.  Also, the grungy look of the background and effects is starting to conflict with the simplified look of the marines, aliens and buildings.  I’ll need to find a unified art style somehow.

The other concern is quasi-ethical.  I doubt that anyone will go on a killing spree after playing my game, but I find spectacular violence for its own sake somewhat distasteful.  Not to mention that all the exploding blood is starting to be distracting for the player.

The other side of the argument is that this is a game where you mow down large crowds of mindless drones.  That’s not too erudite however you consider it, so the graphics are appropriate for the subject matter.  Perhaps I should just roll with it and make it as ridiculous and over the top as I can.

In the end I might do several things to deal with the issue.  Firstly, this is probably the high point of blood in the game.  I’m done with death animations, so it’s not getting any more graphic.  I might tone down the art for the effects, making the blood more cartoony or changing the colour to be more muted and non-red.  The other option is to provide an “extra blood” switch in the game for people who want the spectacle and make it toned-down for everyone else.

What do you think?  Does the game look too gory now?

The other reason is that I’ve been working on various graphical improvements to the game.  Let’s go over them so that I can brag.

Explosions

Before, explosions used to be just puffs of coloured smoke, like this:

I spent a couple of days last week tweaking how they look.  Now I have a nice fireball, sparks and shockwave effect.

Lighting

I wanted the game world to have more variation in brightness, instead of being the same tone all over.  I already had a simple shadow system hacked in, but I also wanted lights.  Thankfully, I played around with something like this in a previous (unreleased) project.

I also added the small detail of muzzle flashes for the marines and turrets.  They help make the firefights more alive.  Notice how the areas around the cannon and the explosion are subtly lit.

Background Texture

One thing that’s always bugged me about the new art style is how flat the background was.  I was also worried about the colours – grey just seemed bland to me.  We’re still working to improve the colour – that’s why I’m experimenting with blue – but I think that I’ve solved the background problem.

I did two things to make the map more interesting.  I added textures to the ground and I also added extra texture around the walls.  Check out this zoomed-out view.

Future Improvements

As I’ve mentioned, there are still a couple things to figure out.  I still think that the colours of the game need an overhaul.  Carlos and I are working on that right now.  I’m shooting for a darker, more grungy look for the enemies.  The second thing that I want to do is to make the walls more interesting.  I’ll probably use the same technique that I did for the background.

Once I have those two things out of the way, I’ll probably release a trailer for the game.  That won’t mean the end of graphical improvements, but at least the large parts of it will be complete.

One of the fun things about Infested Planet is mowing down huge crowds of enemies.  Playing off of that,  I decided to make things a bit more over-the-top.  A couple of days ago I added a decal system that allows me to “paint” the floor of the level.

Now dead zombies don’t just vanish in a red puff.  This allows you to see where really big battles happened and makes the game look a little bit better.

The first one is that I’ve just overhauled the explosion code.  One unfortunate consequence of that has been mysterious memory problems.  Linux has Valgrind, so I thought that I’d give it a shot.

The second, more positive reason, is that I’ve always wanted to release my games on Windows, Mac and Linux.  The code was written with the idea in mind, so the only obstacle is lack of time.  I’m not committing to a Linux Infested Planet yet, but it became a whole lot more probable over the last week.

I’m not done making the graphics better – we’re maybe 50% of the way there.  Still I thought that it’s interesting to show you where we’re at right now.  First, I’d like to remind you of what the game looked like originally:

Yay for programmer art!  I thought that I did ok with AotPZ, but it’s not nearly well enough.  Let’s fast forward to today (keep in mind that a lot of the art is still unfinished, like the bottom bar):

Just yesterday, I put in a lot of the new sprites into the game.  Carlos has been working hard on them for the past week.  You can see the new hive and tower images on the above screenshot.

I’ve made so that you can have different types of walls.  It really makes the game more lively.

If you look carefully, you’ll see that there are shadows under the buildings and units.  This was something that I threw together in an hour yesterday, just to see how it looks.  I like it quite a bit, actually; I might spend more time to make it look better.

Well, that’s what I’ve been up to for the past week.  After I finish off the map graphics, I’ll move onto effects like explosions and animation.

Of course, the graphics are gorgeous and it’s free.  Bleh!  How am I supposed to compete with this?

Is this the untimely doom of Rocket Bear Games? <dramatic music in the background>

To be honest, I’m not too worried, for a couple of reasons.

Awesome Name

Ok, this one is only half-serious.  I feel that “Infested Planet” is a lot more creative.  Carlos came up with that when we were brainstorming.  Of course, I also had to talk him out of using “Swarm” somewhere in the title.  Whew.

Time

By the time my game is finished and released, Alien Swarm will probably be a long-forgotten memory.  People will be hungry for more alien-zombie killing action!

More Enemies

I think that the first level of my game has 10x more enemies than what their videos show.  I doubt they could draw the gigantic crowds that I do with the fancy graphics engine of theirs.

Less Strategy

There’s nothing wrong with action games (I love TF2, for example), but Infested Planet is definitely an RTS.  Alien Swarm seems to be an action game with minor tactical elements.  You can see this in how they talk about it (4-player co-op).  If you watch the videos, the game definitely looks to be on the twitch-reflex side of the fence.

No Mutations

Of course, I don’t know this for sure, since the game comes out tomorrow, but I doubt that Alien Swarm has my mutation system.  That’s what makes Infested Planet fun and replayable.

Campaign Mode

I crush Alien Swarm in this department.  They seem to have focused on multiplayer, and neglected any sort of single player.  That’s too bad, because most people don’t play multiplayer.

Infested Planet will have a replayable campaign mode that you can run through multiple times.  I have all kinds of neat ideas for it.

Summary

In summary, Infested Planet and Alien Swarm are two different games.  Heh, who would have guessed?  I’m going to be checking the game out tomorrow, and you should too if it looks interesting.  You can’t beat the price.

Who knows, maybe this will even help me.  I’m going to be stealing visual effects from them, and maybe there will be a surge of interest in killing large hordes of baddies.