Second, here’s my timelapses:

Coding etc.

Audio

There’s two because I did the audio on a different computer, and I’m too lazy to edit them together properly.  Read on for my post mortem.

This actually went surprisingly well, especially considering I spent the best part of 7 hours on Saturday afternoon at work.  It was definitely the most enjoyable Ludum Dare I’ve done so far.

What worked:

• Setting my alarm for 2:15 am to see what the theme was.  It may have been a fluke, or it may have been that my head was in a different place having just woken up from deep sleep, but literally within 5 minutes of seeing the theme I knew what I was going to do, and within 5 minutes of that I had the entire game fleshed out in my head (I don’t think I changed anything between then and the finish – I’m pretty sure I even came up with ‘Maggott’s Restorative Ship’s Biscuits at this point).  Another 5 minutes and I was back to sleep.  The whole thing was pretty much perfect.
• The map and background paper.    I did the map in black acrylic ink on cartridge paper, and I think it turned out really well.  The compass is a bit wonky, and I maybe should have drawn those lines you see on old maps to go with it, but that’s just nitpicking.  I was a bit worried about the background paper being too light and not stained enough, but a little fiddling with Paint Shop Pro fixed that.  I’m really pleased with how the whole thing turned out, especially when the coloured segments get drawn on the map as the game progresses.
• Research.  Getting to research old maps and food adverts on google was fun.  I think I did a fairly good job of getting that old-fashioned (Victorian?) look for the game.

What didn’t work:

• The AI.  The CPU players are useless and offer no challenge.  I should really have spent more time on them to make their actions less random, and I should have used A* to get them to move towards unexplored spaces.
• That’s it, as far as I know.  Maybe some comments will point other problems out, but I haven’t seen anything obvious yet.

What may or may not have worked:

• Not having the chance to play the game with other human beings yet, I’ve no idea whether it’s actually a good game.  I think it has the potential to lead to arguments and people storming away in a huff (the mark of all good multiplayer games), but really I don’t know.  The rule set is very simple, so maybe that’ll make it less fun to play, but until I can try it out properly I’ve no idea whether it works or not.

As a bonus for reading this far, here’s a picture of the stuff I scanned/wrote down during the competition:

So there you can see the 3 buttons and the amount of food the player has (top left).  Each movement/garrison costs 1 food.  On the map there’s a garrison to the right of the blue player, which means nobody can move onto that space until its turn limit is up.  You can also see various spaces with food in them.  These appear after someone forages in that space, so you can see how much food can be got from which spaces.

I keep forgetting to post my pictures of my food, but here’s my lunch:

Beans on toast, and a goblet of Irn Bru. You can just about see the original map in the pile of paper to the right.

Next step is to add the victory/game over screen, then the Forage, Garrison and Wait buttons.

Players can move to adjacent spaces, and can’t move onto spaces that are already occupied.  The next step is to draw up graphics so you can see how many spaces the player’s already explored.  It’s fairly time-consuming filling in those thin squiggly lines, so I’m only 2/3 of the way through that.

If I have time I might try and stain the edges using computery magic.

(still waiting on the tea-stained paper to dry so I can use it as the background)

The compass is kind of skewiff but otherwise I’m pretty happy with this.  Now I’m going to do the old ’soak a piece of paper in tea’ trick to create a vintage-looking paper background.  Not started coding yet…

There are 4 players, and an unexplored island divided into tiles.  The player to visit all those tiles (and thus explore the whole island) gets to claim the island for their country, and wins the game.  There are three obstacles in their way:

• Players cannot move into a space another player is already occupying.
• Players can only move if they have enough food rations.  Otherwise they will have to spend a turn foraging.  Each space will have a different number of food units, so some spaces will be more useful in this regard than others.
• Players may erect a garrison on a space which will prevent other players from moving there even if the original player has moved on.  Garrisons last 3 turns.

That’s it.  It’s probably more like a boardgame than a videogame, but I’m trying to keep it as simple as possible to make things easier on myself.  There will be the option of CPU players to make up the numbers, but I can’t guarantee they’ll be particularly smart (not being particularly smart myself).

First step: draw the island map.

Anyway, I’ll most likely be coding in C++ using SDL/OpenGL and PortAudio plus libvorbis, and my template code which you can see here.  If not I may use flash, but I reckon I’ll work faster in C++.

I’ve got a 5-hour shift at work on the Saturday, so I’m aiming for a more minimal design than my previous 2 entries.  We’ll see how that goes.

Caverns of Light Walkthrough

• Fixed the performance on 16bit displays.  Thanks to Endurion for pointing this out.
• Reflectors can no longer be placed on top of each other.  Should make the game slightly easier.
• The game was accidentally set to default to antialiasing being on.  This seemed to cause slowdowns on the linux version, and may cause other issues for folk on other OSes.  It’s completely unnecessary anyway, since the graphics are entirely made up of (pre-antialiased) bitmaps.  If you’ve already run an older version though, you’ll have to either delete the CavernsOfLight.conf file (see the readme for the location – it’s different for each OS), or change the ‘a 6′ line to ‘a 0′.

Quite a few of the comments I’ve had have complained about the game’s difficulty.  My only defense here is that I’ve been playing a lot of Spelunky, Mighty Jill Off, and When Pigs Fly lately, all of which rely on pretty accurate jumping skills.  I think when I have a bit more time I’ll probably do an updated version with 3 different levels: easy; normal (the level currently included with the game), and sadistic, which will be massive and absolutely punishing .

Incidentally, if anyone likes the game enough to make their own level for it, I’d love to hear from you!

I’ll start with the bad, because it’s only little things:

• Stupid jumping bug.  Platformers really need to have solid collision detection, and mine kind of fell down (ha ha!) there.  It’s no good if you’re jumping up to a ledge only to fall through it into the lava below.  It only took 5 minutes to fix when I came back to it this evening though.  The perils of writing a game in only 2 days…
• The music.  I’m not sure it really fits the rest of the game.  I was going for an underground, subterranean theme, but I think it’s just a bit too sinister.  I should have tried to do something more upbeat.
• The theme.  Given all the existing games already set in caverns, I can’t help feeling I would have made a weirder, more interesting game if one of the other themes had won out.  Still, given how Caverns of Light turned out, I’m not too fussed…

The good:

• The graphics.  Doing everything in Inkscape really turned out well, and other than a few slight glitches where certain tiles get drawn over others, I think it looks really polished.
• The ‘connect the lamps to lightboxes’ mechanic.  Particularly the moment when you activate a lightbox and the sky gets lighter as it plays the ‘light-up’ sound.  That’s a great moment.  And I think the mechanic has a lot of potential.  If I’d had more time I would have definitely made the level bigger and more complex.
• Doing the level editor first.  Getting this out the way before I did anything else really made things easier for me than if I’d done it the other way round.  Once it was up and running all I had to do was implement the game logic for navigating round the level.  Incidentally, I didn’t use up anything like the maximum space in the level (the full size of the level is 1024 x 768 tiles, which is pretty huge), so there’s scope for other people to make better ones…
• Building a complete (if a bit slight) platformer in 2 days!  I’m still grinning about this…

Now for the timelapses.  I’ve actually got 2, because I did the music and sound fx on a different computer (my laptop’s not set up for making music), and I didn’t bother joining them together.  First, the main one:

Ludum Dare 15 Timelapse 1

And the music one (this takes place between 10:30 and 12:00 on the Sunday):

Ludum Dare 15 Timelapse 2

So I’ve added music and sound fx, the last of the hazards is in and working (to the left of the above shot), and I’ve sorted out the sky so it’s smooth and gets lighter the more lightboxes you activate.  Now I just need to do some level editing and add a story screen and it’ll be ready for release!

Okay, last post before I switch computers and go and make some music.  Here I’ve added spikes and lava, and collectable reflector crates.  It’s possible to die now too, although there’s no fancy animation – it just decrements your life and puts you back to the start of the level.  I don’t know if I’ll have to time do anything more polished in that regard.

Still to do:

• Moving spiky balls.
• Fix the collision detection.
• The sky should get lighter as you light up more light boxes, plus it should be smooth, not stepped as it is right now.
• Make a proper level for it.
• Story screen.

There’s other things I’d like to do, but only if I have the time.

I’ve added lightboxes and a UI (left to right: num lives, num lightboxes lit up, num reflectors available).  I’ve decided the aim of the game is to light up all the light boxes in the level, which will entail connecting lamps to boxes via reflectors, finding extra crates of reflectors, and avoiding any hazards along the way.  Eventually the sky will get lighter as you light up more lightboxes.

• First, do all the little things the game still needs:

1. Light boxes which send the light beams up to the sky.
2. Light pipes to send the light from light boxes deep in the level up to the sky (and to act as extra obstacles for the player).
3. UI, showing: number of lives; number of light boxes lit up; number of reflectors available.
4. Pickup-able reflector crates.
5. Ability to die, from:
6. Lava.
7. Spikes.
8. Moving spiky ball things (yay originality!).
9. Fix the stupid collision detection!

• 10.30: Do the music and sound fx.
• 11.30: Off to the shops.
• 12-12.30: Back home for lunch and the grand prix.
• The afternoon: Finish putting the level together.
• Late afternoon: Add story screen, build final Windows version.
• 6.00: Dinner.
• 7.00: Build Linux version.  Release.  OSX can wait for another time.

I’ve got the lightbeams up and running, so yay!  There’s still a lot of stuff to do, but I think I’m going to leave it for tomorrow.  I reckon (I hope) I’ve got the hardest parts out of the way now (though the collision detection’s still not right), so hopefully tomorrow will go a bit easier…

I made myself a mozzarella pizza for dinner.  It looks the part, but I think I need to work on my sauce-making skills – there was definitely something missing…

Game-wise, I’ve got the scrolling mostly working except for a weird offset between the player character and the rest of the level that’s suddenly appeared.  I’m going to try and fix that now, then hopefully I can get the light beams up and running before bed.

The collision detection’s pretty ropey, but you can now zoom your little robot about the place like a proper platformer.  Next step is to get the screen to scroll as you move about the level.  Then it’s onto the beams of light themselves…

Hopefully now I can get on with the business of coding the game itself…

It’s still missing the ability to save and move around or zoom the level out, but it’s basically there.  Now I’m going to empty the washing machine and have lunch.

The background’s a placeholder, but other than that everything’s ready (well, the quit button anyway, the other two don’t do anything just yet).

First step I think is to create a level editor though – I’ll see if I can get that done by lunch.

Here’s an initial mockup of the tile graphics:

My very untidy desk: