[ postmortem #1 | #2 | Play the game ]

The theme for LD22 was “Alone”. I have two strict rules for myself when it comes to the theme:

• some aspect of the core gameplay mechanic (as opposed to just the dialogue/artwork/whatever) must be inspired by the theme
• never modify a game idea for another theme to make it “fit” with the theme that actually gets chosen

I feel like I put a lot of thought into this theme. It wasn’t enough for me to go with a game where you control one character and have very few or no interactions with an NPC. The large majority of games from any given LD are like this, and they could have filled this requirement with some minor rewriting. I rejected my first game idea because it was like this.

So rather than making another game where you just happen to be alone, I decided to change a game where you’re normally not alone into one where you are. The obvious choice for multi-character gameplay is an RPG. I could have just made an RPG where you only control one character, but that’s not very original, especially for Ludum Dare. So I decided to turn the standard formula around: instead of starting alone and building a party, you start with a party that diminishes until you’re alone. I like this idea a lot because:

• it’s original. I can’t think of any game I’ve ever seen with this mechanic.
• it’s inspired directly by the theme, but a creative interpretation rather than an obvious one.
• it adds an interesting challenge to the gameplay (because you don’t control which characters will leave).
• it challenges me, because I still had to make five different characters.

So I’m really happy with how I used the theme. Based on the comments I’ve gotten so far, most people disagree with me. They say the connection to the theme is weak or forced. I guess a game about being alone is more accepted than one about becoming alone. I think I would have gotten a higher Theme score  if I had only one character from the beginning, like most of the games in this competition (and other competitions). Still, I’m glad I didn’t. I’m not disappointed in others’ games, but I would have been disappointed in myself if I had gone the “accessible but obvious” route.

Anyway, I’m surprised some people thought my use of theme seemed tacked on, because I feel like I probably put more thought into the theme than most people did, even if I came to a different conclusion. Still, I think it’s great that people interpret the theme in different ways, so I’m not upset that the judges have a different idea of what’s a good use of theme than I do. I plan to keep trying outside-the-box ideas in the future, I just hope it’ll be clear that’s what I’m doing rather than ignoring the theme!

[ postmortem #1 | Play the game ]

Thte Last Adventurer is an HTML5 canvas game with Javascript and the GameJS library. My previous LD entries were in python, and I was told if I made a browser game then it would get more plays, but so far that’s not the case. I’ve only got 12 ratings so far, which seems pretty low for a web-based game. Oh well.

HTML5 gets used for a lot of demos, and it doesn’t have a good reputation for game making, so I wanted something way more graphically and computationally demanding than Tetris. My game is still a bit rough around the edges, but that’s completely due to the time constraints. I’m sure that HTML5 is more than capable of making an extremely polished game.

I had  a lot of fun working with transformations and transparency. Transparency is used in many places in the game, including shadows, the selection region, the exit portal, and the boss of level 4. I used stretch transformations for anything circular that appears on the ground (shadows, selectors, portal), rotation transformations for the “knock” animation when a character gets hit, and scaling transformations for the level 4 boss and the “quake” spell casting sequence.

The pièce de résistance is the circular gameplay area. An 854×854 square image is pre-rendered. Then, each frame, it’s rotated by the current angle, scaled to an isometric 1:2 ratio, and its drawing region is clipped to an ellipse using the canvas element’s clip method. No caching of transformed images is done, and I think performance holds up well. You can see the smoothness of the rotation when you press LEFT or RIGHT. I should have incorporated that into the gameplay more, but more about that next time!

[ Play the game ]

I’d call this one a success!

Good:

• Game is finished, with a lot of content (for me): five playable characters, three spells, three bosses with different behaviors, an upgrade system!
• I was dedicated. I slept about 3 hours per night and spent almost every waking hour on the game.
• Game runs (relatively) smoothly. Given how little I optimized, and how many graphical tricks I used, I think 20-30fps is pretty good.
• I’m getting the hang of Javascript.
• HTML5 canvas turned out to have a lot of relevant useful features
• Really pleased with how the boss of level 4 turned out (and I got in a Wayne’s World reference)
• Really pleased with how the rotating gameplay area turned out
• Happy with the graphics, sound, and character design for the most part
• Deploying to web turned out to be pretty easy. I’m very happy to have my first web-based LD game

Bad:

• I get so engaged with my game during the competition that I have trouble joining the community. I’ll have to make up for it now that it’s over.
• Navigating an isometric layout is a tricky problem
• Had to leave out in-game instructions
• I wish I had more engaging gameplay, but it’s fine for what it is
• Had to leave out storyline (which is fine because it was too depressing)
• Sprites are cut way down from my original idea, surprising no one
• Had to develop in Firefox because of how Chrome handles trusted domains on local filesystems

Definitely getting nervous. I should be more comfortable with a tool starting LD than I am with GameJS and JavaScript right now. I’ve spent the last day writing a module for handling game entities in a tree, which is the way I usually like. If you’re using GameJS and you want to use it, get it here.

And I saw that you could submit warmup games so I submitted mine here.

Good luck, everyone!

I’m going to make an HTML5 canvas game using the GameJS library. I really want people to be able to play my games easily, so it’s time for me to learn something web-based (my three other LD entries were in python).

I’m pretty nervous, because I just started trying out GameJS for the first time tonight, and I’m not very good at JavaScript yet…. I made an HTML5 game for practice last week, and it was pretty painful. Still, now I know it’s possible! Check out my practice game on your HTML5-capable browser if you like:

[ Intro to Fixology ]

[ Play Abigail's Descent | gameplay footage ]

My game’s a platformer where you descend around the outside of a cylindrical pseudo-3-D tower. It’s nothing groundbreaking (I’ve seen the tower mechanic in at least 2 other games) but I still felt like I put enough into it that it wasn’t completely derivative.

I don’t usually have trouble coming up with an idea that I like. I started on the tower mechanic 21 minutes into the competition, and I had the rest of the gameplay in mind within a couple of hours.

I think I made the right choices about which features to cut and which to keep. Some highlights:

• I’m really happy with the 2.5-D tower. This obviously was not trivial to implement, and it turned out pretty well. I wish I could have made the platforms look better, but nobody has complained yet.
• I like the grayscale characters. It reminds me of Paper Mario. Feedback has been mixed.
• I really like the clouds. Nobody’s commented on them, but I think they do a great job adding to the 3-D effect and they only took 20 minutes to add.
• The dialogue/story is good, but I wish I could have spent more time on it. The knight was originally only going to be in one level, and I had other characters planned.
• I think I got the challenge level just about right. I followed my rule of thumb of making it 1 to 2 levels easier than it seems like it should be. Some players still thought it required pixel-perfect jumps, though in reality every horizontal jump has at least a 14 pixel margin of error.
• People don’t like jumping games, I know. I really like to have as simple of mechanics as possible and make the most out of them. I think if your character can jump, you should have some challenging jumps. I hope that the frequent checkpoints helped minimize the frustration.
• The sound and music are fine. I may some day be able to make my audio rather than just using whatever comes out of sfxr and Autotracker-C. But until that day, I’m grateful for these amazing tools.

I had a great time this LD, and I was pretty happy with how the game turned out! It’s already gotten twice as many ratings as my last LD game did during the whole competition, so I certainly can’t complain about that. So far I’ve gotten very constructive feedback.

Thanks to everyone who took the time!

[ Abigail's Descent | timelapse (youtube) ]

I admit I don’t really “get” the whole timelapse thing – it’s too fast to see anything that interesting. I had my webcam this time (although it stopped working on Sunday) since I was going to be LDing in a variety of cool locations (you’ll have to watch the video to find out where!). I put a big countdown clock on there too, sacrificing a good 30% of my screen real estate for the sake of this video. I still don’t think it’s that interesting!

Anyway, if anyone has any tips for making a better timelapse video, lay them on me!

I’m making a fairy tale of a self-reliant princess who has to rescue herself from a tower. Pretty crazy, right? The mechanics are pretty simple, so now I’ve got to focus on content. Here’s how it’s going:

Despite the mind-blowing 3-D graphics, this is a pygame (SDL) project. Still sticking with what I know here.

I’m using python/pygame again. After my LD #19 entry received only 9 plays, I decided I really need to learn something web-based. I started to learn ActionScript and even made a failed entry at One Month Game, but I don’t feel like I’m comfortable enough with it yet to use it for LD. So I don’t plan to get a lot of feedback again, but I still hope to have a good time!

Unfortunately my home internet is down until 29 Aug, due to a telecom workers’ strike! Since none of my neighbors were nice enough to unsecure their wifi, that means I have to take this production on the road. I plan to write my game from work, the library, and a friend’s house. Never a dull moment. I probably won’t be able to get a Windows port together until after work on Monday. I hope that doesn’t keep too many people from playing.

I hope to stick to my strengths for this game, which are humor, story, and creative use of the theme. It will probably be silent with badly hand-drawn graphics, and lack polish since I do everything from scratch, but I hope to make up for that with interesting game mechanics.

Good luck to everyone!

[ Go to entry | Watch timelapse ]

My game turned out to be a vaguely steampunk-ish exploration/management game in space.

I can’t believe I finished this! Aside from the lack of sound, the game is more polished than most of my Pyweek entries, and those get 7 days. I was very surprised that I managed to implement everything on my list without cutting features. I’m actually kind of impressed with myself.

Having said that, it’s not a great game. It’s not very fun and it’s not very innovative. If you like exploration in games as much as me, there are some cool moments, but you have to work to get them.

The gameplay complexity is added gradually throughout the game. I think I should have front-loaded it, because I expect that people will quit after 10 minutes without seeing the whole game. People who play for less than an hour probably won’t get the full experience, and I know that’s way too long to expect someone to invest in a LD game, especially when there are hundreds to play! Oh well….

This is now the third game I’ve made that’s basically a rip-off of one-quarter of Starcon 2. One more to go.

What did I learn? I need to learn how to generate sound and music for the next LD. In Pyweek you can use sounds that are freely available. I forgot this wasn’t allowed until the last minute, so I had to disable the music I put in. I may upload a Jam version that includes it. Also, python + pygame is great, but it’s time to learn another language for making games.

Thanks to everyone involved in putting this competition together, and good luck to all the contestants!

This was my first Ludum Dare, and I was really looking forward to it. Well, I did have fun making a game, but there’s one major problem: I don’t have Windows, I don’t have access to Windows, and I couldn’t find anyone to port my game to Windows. So I kind of feel left out. Not only can nobody play my game, but I can’t play most of the other people’s games.

I always have such a great time with PyWeek I thought for sure I’d like this too. But I really have to consider whether Ludum Dare is for me. If I needed Windows to participate, I would easily choose not to participate. But maybe that’s not necessary. I could write in C++ instead of Python, and maybe more people could run my source. Or I could learn Flash or something highly portable. I still couldn’t play most of the other games, but it’s a step up. We’ll see.

Anyway, it’s been enlightening. And at the very least, I enjoyed making something, so it’s not for nothing.

I was waiting for a Windows port before doing a postmortem, but I wanted to add a walkthrough for people who don’t want to solve the whole thing.

General hints: There’s three parts: an “orientation” section, followed by a puzzle with three switches (orange pink white), followed by a puzzle with four switches (blue green red yellow). Each of these three parts is separated by a long tunnel, and you never need to backtrack to a previous part. If you think you’re going the wrong way, press Down to turn around. Also, hold Up or Space while walking to speed things up. Because all the steps are reversible, it shouldn’t be possible to get stuck. You might wind up worse off than you started, though.

In the walkthrough below, each line is one room. “Blue 1st right” means turn to face the blue switch, hit it, and then take the first door to the right of the switch. If the line doesn’t have a color, such as “2nd left”, do not turn to face the switch, even if there’s one in the room. Just take the second door to the left of the direction you were facing when you entered. (This is counting closed doors, so if the first door on the left is closed, “2nd left” is the first open door on the left.) Okay?

Part 1:

1st left
Cyan 1st left
Purple 1st left

Part 2:

Orange 1st left
Pink 1st right
White 1st left
Orange 2nd left
White 2nd left
Pink 1st right
White 1st right

Part 3:

1st right
Green 1st right
Red 1st left
Yellow 1st left
Blue 1st right
Yellow 1st right
Green 1st left
Yellow 1st left
Blue 1st right
Yellow 2nd left
1st right

Thanks for playing!

Well, here’s my work space 15 minute before the competition started:

I got gedit and a terminal, so I’m good to go. But looking at the other posts, I got a bit self-conscious about how modest my setup is, so I decided to take this show on the road.

LDing on the subway!

Harvard!

MIT!

Boston Public Library!

Hey, they had a book. Now I can totally make a 3D game!