Behind The Dumb Episode 6 is out at last! It’s the first half of my timelapse and journal for LD14! Check it out. You don’t need to rate it, because while I did FILM it all during the weekend, I certainly didn’t have it all edited and ready by then. And the second half won’t be out until voting is over anyway. Still, it should be amusing to some.
Tags: behind the dumb, LD #14 - Advancing Wall of Doom - 2009, timelapse
Unfortunalty I wrote the drawing code wrong* (outside of the gui thread) and this is the reason some people are getting the gray screen bug.
It doesn’t depend on your color depthor anything like that AFAIK, I will need to fix it in the code.
The good news is that I know more or less what to fix and just have to find the time to work on it again 
[*] I’m a Java2D noob 
DoomCake was my first LD entry. It was developed in Lua, using the LÖVE 2D engine.
In making this game, I learned a bit about Lua, and a lot about cake. Read on to find out what a sugary zombie invasion might look like…
Introduction
The competition was flagged up to me by someone at work, though I wasn’t sure I’d be able to do it until the day before the start time (which was 4AM local time). I didn’t even decide what development environment I was going to use until the Friday evening, when I spied a Lua book in the office library on the way home from work. I’d seen LÖVE mentioned by a couple of folk here, so I decided I should give that a shot.
I should mention I’d never used Lua before, so when I started reading on Friday evening, I was literally starting at ‘HelloWorld’. I went through a few examples, then started playing around in LÖVE, seeing what it was capable of. Satisfied I could put together a simple sprite-based game, I went to sleep a few hours before the competition start. I figured no point trying to stay up late just to find out the theme, only to ruin myself for the next day.
Trying to learn a new language, whilst against a deadline
Never having used Lua before, I had to learn how to solve a number of problems against the clock. Things I take for granted in my day-to-day C#/.Net life. Things like multi-file projects, debugging and error-handling, collections & queues, and so on. At first it felt like I was reaching for a book five times for every line of code I wrote. Not very productive!
I managed to pull together enough to make my game work, with minimal bugs. However the code-quality is horrendous. Inconsistent styles throughout, as I learned my way around the language. Huge amounts of copy-paste, since I couldn’t get certain code-reuse aspects working properly. Certainly not my prettiest work, from a development point of view.
Quick and easy love
The LÖVE engine made getting assets into the game very simple. Images (static and animated) and audio were easy to bring in. I would recommend it as a good choice, especially if I knew Lua better (or even at all!)
Cake!
The idea for the Wall Of Battenburg Doom came about as I was filling up on sugary snacks on the first day. I went around the house, collecting anything that might work as ingame enemies, with the intention of taking photos of them to make sprites from. (Art isn’t my strongest skill, so I figured that would save me time). However I underestimated the amount of photoshop processing that would be required to turn the pics into something usable.
In the end, I only used two items: The battenburg cake, and some candy teeth. I had an idea that I would create a parade of Marching Sugary Doom. Doughnuts, fizzy drinks, lollipops, chocolate bars, and so on. All of these would be following behind the main wall, covering the ground with sticky, sugary rubbish. This obviously didn’t happen, which lessened the effectiveness of the other idea I had: Creating a Left4Dead parody, specifically the survival-until-rescue part.
The food. It’s… infected!
Imagine you’re the survivor of a zombie attack. Only, instead of zombies, the virus/mutation/whatever is causing sugary snacks to rise up and eat people. What’s more, their weapon of choice is CANDY TEETH! So all the while, we’ve been told not to eat too much sugary stuff, as it will rot our teeth. Now the roles have been twisted around.
Anyway, this side of the game never really worked out, simply due to time constraints. I wanted to have multiple levels, with different types of wall behind the teeth, and various obstacles that the player would have to navigate around. I also wanted to have different coloured pick-ups, with different effects. But sadly time was against me.
Rescue
One of the great parts of L4D is waiting for rescue, against a seemingly never-ending hoard. I blatantly ripped off the rescue-helicopter idea. As the chopper got closer, he would radio in encouragement to hold out just a little longer. If you managed to survive long enough, he would pick you up, and fly you to safety. (Note how he flies you over the wall and beyond, where the legions of marching snacks would be, if I had time)
I wanted to have an actual radio, sat on a table next to the helipad, with the pilot’s speech bubbles coming from it. But again, this didn’t make it in.
A familiar plateful of cake?
The final part of my L4D inspiration was the box art. I’m talking about this
which I copied slightly to make this
Spot the similarity? Apparently it was too subtle, since no-one mentioned it. Oh well…
Tags: doom, post-mortem, postmortem
It sucks to have very few ratings and comments on your game, so here are a few pointers for next time.
Platform
The most rated games are the ones you can play in your browser. Try making your entry in java or in flash and embedding it in a webpage for maximum exposure. If you don’t want to have to suffer the horror that is (Java|AS|Flash) at least make a binary release for Windows since most people seem to have access to that platform (I don’t so please also make an OS X version 😉 ). If you’re making your game in Python (or its pale imitator ruby 😛 , or any other interpreted language for that matter) you might think you don’t need binaries but you’re wrong, people aren’t going to install a different environment for every single game. Same for Löve games.
Dependencies
If you do only release as source, or if you only release for one platform, or dynamically link libraries, keep dependencies to a minimum. If all I have to do to build the game is type “make” there’s a better chance I’ll try it than if I have to install 13 libraries which each in turn depend on half a dozen others.
Screenshot
I try to play all games regardless of the screenshot (in part because my own drawing skills are easily surpassed by those of a drunk monkey with a pen) but if the entry requires more than clicking on a shiny icon to run, I’ll definitely spend more time trying to get an entry with a cool screenshot working.
Rate other games (and leave comments!)
While I am determined to try as many games as I can, I prioritize rating the games from the people who left comments on my games first.
Of course these “rules” aren’t set in stone. You can make a game for the Atari 2600 and still have plenty of ratings. You can also make a game with an uninspiring screenshot and win best overall.
Of course you could also cheat by making a ton of games under different names to rate your own games 😛 .
Sort of a postmortem for my game, Heart. I don’t know, I just wrote about the process and a few later thoughts. You can read it in my blog.
Also, the game is finally finished. Play the release version.
Tags: post-mortem, postmortem, tigs
Fan (?) made youtube video showing why you must play (or avoid?) my LD entry Mind Wall before time runs out!
Outline of 48hrs as I recall it…
Day before compo…
Stayed up too late on Chatroom, just getting hyped about the competition – Should have slept more.
Got up at 4am to check theme – found that it was AWOD couldn’t think of anything and went to bed.
Got up again jotted down game concept in notebook (see below)
Went to work knocking together basic sprite object – I know I should have grabbed a pre-developed set of objects but at least I can refine and reuse these now!
Went for a very free form map, no tiles just objects relying completely on BlitzMax’s in built pixel perfrect collision detection.
In the first day I managed to get the basic game mechanics working block growing, turrets firing, player deploying turrets – it was now almost a game.
Went to sleep thinking about adding towns, people, animals, plants running from the AWOD… Dream on!
Day two
Added screen, levels, win/lose states, basic sound effects
Afternoon – time was running down fast added …
Powerups – providing a limited upgrade to turrets
Base something for the player to defend and powerups to appear next too
Last minute play testing
Quickly zipping and posting it near the deadline – Phew Final post
What went right…
I managed to create a game in 48 hours!
I avoided getting tied into the complexity of a tile based system, thankfully BlitzMax’s collision detection works well!
I kept the ‘artwork’ to the bare basic minimum – inspired by the BUPA adverts (Coloured circles as people)
I kept the design very simple as well.
My Keep It Simple Stupid – KISS approach worked!
What went wrong…
Wrote too much code to start with I need a basic prototyping framework for at least sprites, collision detection, screens, menu ect.
Then I could have spent more time designing filling out the game design further but staying flexible, this might have prevented the last minute addition of powerups and base leaving next to no time for game testing balancing.
Motivation wandered during day two, it was a playable prototype and therefore in theory a game, thankfully I managed to get moving again and ‘polish’ it up a bit, again a framework will make this a none issue.
I quickly realised that there was a performance issue with the block count, once you get over the 1000 mark the framerate really drops so I had to limit that.
I wanted a wrap around world where you can’t run away from the wall of doom, althought this should have been an easy thing to add I did not get around to adding it, possibly resting on my laurels a bit!
In Summary
I really enjoyed this although I wish I had setup a time laps screen capture, probably not a webcam feed ;o).
Well I must judge some more of the other amazing entries, Looking forward to the next competition already!
Regards
Merx
Tags: design of AWOD, postmortem
Introduction
PET THAT KITTEN! was my second attempted Ludum Dare, my first actual entry, and my first XNA game. Going into this I knew I was going to use this competition as a way to get my feet wet with XNA. Considering the added learning curve I aimed for a very tiny scope. I think I went with the idea that crossed most LD’ers minds, turn the wall into an enemy and have it chase the player.
Development Tools
I had worked with a bunch of C# in the past so putting everything together in XNA wasn’t a problem. I had done my art up in Flash CS3. I had planned on using hobnox to generate my sound effects and music.
What Went Right
The scope I had planned out mostly revolved around the fact that I was tackling XNA for the first time. The scope had lots of features I could easily cut out while still leaving me with a “complete” game. I had done a bit of VB.net and C# managed DirectX stuff in college, XNA turned out be extremely similar so only very few features got cut.
Despite the programmer art, I’m pretty pleased with the way the artwork turned out. It was more time consuming than I had liked, but I feel like it breathed a lot of life into an extremely simple concept. I also decided that however crappy the artwork looked at the start, it would make it into the final project, no placeholder art.
What Went Wrong
I was using this Ludum Dare to kick myself in the butt to get something done in XNA. Most of my perceived problems stemmed from this though. I realize the ideal entries are executeables and broswer based games, no one really likes having to install dependencies or go to too much trouble getting your game going. I’m sure the installer for the XNA framework turned at least a few people away.
On that note, I know I could have gottent his entry done in half the time, possibly with more features if I had just done it in Flash. A good chunk of my time was converting assets to spritesheets and then fixing up the spritesheets (you’ll notice in the wall animation that you can see bits of other sprites on some frames).
My time management for this competition wasn’t too bad, I think I spent too much time on getting some of the artwork done, the time would have been better spent with an in game help screen explaining the game objects and scoring system. The confusion brought on by the scoring system (and a scoring bug) was a huge oversight on my part.
Conclusion
As an XNA project I’m pretty happy with PET THAT KITTEN!. It’s a complete mostly functional game, and I managed to plan my scope out pretty well for the time frame. However as a Ludum Dare entry it’s a little less than ideal as I know I would have been able to polish it(Sound effects!) quite a bit more in Flash and avoid installer issues.
This Ludum Dare was a great learning experience for me. Planning out a scope under such a tight deadline, learning XNA, and finding a couple of really nifty tools for mixing music and creating sound effects.
Tags: postmortem
Welp I’m here. I might participate in future compos.
I believe a better name than post mortem would be lessons learned, so let’s see my lessons learned from the competition. I would love to read this before the compo, as I can’t make the time go backawards, let try at least to help other newbies.
1. create the menu navigation at the start of the competition
As I was fighting agains some physics problems during the development process most of the menu was left behind. Then I realized 30 minutes to the end that my menu would be an image =) and a play button…
The lesson on this would be, use time in the middle of the nightmares to do these brainless activities, because what can go wrong while build a menu ? =)
2. draw in the paper build and color in the computer
That helped alot, I save my day manipulate the lines with inkscape, but … in order to slice images I have a nightmare with inkscape, go with Gimp works as a breeze
3. dont split images, let the engine split for you
ahhhh I spend some hours cutting images lol !!, could use something like http://love2d.org/docs/love_graphics_draws_1.html to draw a subsprite of the image …
4. make physics work for you
Oh Well after the physics hit me in the head, I learned some details and was able to make it work in a decent way, special note for the boats that hang around over the sea, for that I created a sequence of x,y that is the path to the boats, and I try to follow the path, I believe if I make the water line as an object and change its group to only colide with the boat would be the best to move the boat around.
5. clean the tables in one place only.
Not sure why, buit when I was removing items from my objects list in teh collision handler the LOVE was just crashing … workaroiund that I found : flag to items to be removed and remove in the update() method and one by one from each table, this is call at the end of the update() for each cleanable table :
[code]— clean the tables
function cleanTable(table2Clean)
for n=1,table.getn(table2Clean),1 do
if table2Clean[n].dirty ~= nill then
table2Clean[n].poly:destroy()
table2Clean[n].body:destroy()
table.remove( table2Clean, n )
break
end
end
end
[/code]
and at the collision handler, I dont remove the pig from the list, just flag it
and I believe that this would be a nice way to animate the pig booom =)
by using a multistate, like alive/almost dead/dying/im outa here
would change the image section, or play an animation … and at the end remove from the list
that would be great …
[code]
function killThePig( id )
love.audio.play( audioCollision, 1 )
pigsKilled = pigsKilled +1
for n=1,table.getn(pigs),1 do
if pigs[n].poly:getData() == id then
pigs[n].dirty = true
break
end
end
end
[/code]
6. reusable files should be in a folder
I’m using for labels the .lua classes that I made called LOVEly-eyes, but I had this problem with the Text object that wasnt transparent … Oh well I went in the code changed the super class, Rectangle to handle this and the Text is transparent by default, cool, but but but … after that I just copied all the files from the LOVEly-eyes folder to my game folder … too bad because I copied a main.lua file together … If I wasnt using subversion would be a nightmare … now LOVELy-eyes are in a separated folder =)
7. use a version control system
I used subversion, saved me when I made an stupid folder copy … revert and just lost some minutes of work… bu remember keep on committing =)
8. put together a zip with everything
Better that just the .love file, create a package with the execs, the best would be create an installer.
9. level up level up !!!
people like rewards, so more than the score I should add level concept, just with faster attack of the pigs, or a different scenario with different speed .. hummmm that would be cool a .lua file for each level
10. collision has 2 sides … A and B
It took me some time to realize that A and B collision data, first they are the DATA from the polygon nothing else, just disconnected data, not a reference, not and pointer … hehehhe string data what makes very nice and unplugged from the code, and you have to test both sides, if wherever A colide on B and the oposite, this was my colision code
[code]function collision(a, b, c)
if string.starts(a, “pig”) and string.starts(b, “battery”) then
killThePig( a )
removeTheBattery( b )
elseif string.starts(b, “pig”) and string.starts(a, “battery”) then
killThePig( b )
removeTheBattery( a )
elseif a == DIODONTIDAE and string.starts(b, “food”) then
eatFood( b )
elseif b == DIODONTIDAE and string.starts(a, “food”) then
eatFood( a )
end
end
function string.starts(String,Start)
return string.sub(String,1,string.len(Start))==Start
end
[/code]
note that I used start() because I add the object id after the type, so I can grabb it from the list, something like “crap_2”, “crap_3”
11. scroll the view is possible Luke… use the force !
Ha ! not the force, at the end I solved the screen scrolling with a simple solution, calculated a shif from the main character and update this shift in the update() method and every single draw has this shift. So the camera is following the character, thats stays in the screen all the time, and to avoid the char to drop outside the world I add invisible walls on left and right side and check if the cameraX is in the possible range,
this in the start of the code
[code]
startCameraX = love.graphics.getWidth( ) / 2
startCameraY = 200
cameraX = -startCameraX
cameraXLimit = {}
cameraXLimit.start = 0
cameraXLimit.finish = -2905
[/code]
+ cameraX on each draw
[code]
love.graphics.draw(diodontidae.image,
diodontidae.theChar:getX() + cameraX,
diodontidae.theChar:getY())
—— draw the batteries
for n=1,table.getn(batteries),1 do
love.graphics.draw(
imageBattery,
batteries[n].body:getX() + cameraX,
batteries[n].body:getY(),
batteries[n].body:getAngle() )
end
[/code]
this on the update()
[code] cameraX = startCameraX – diodontidae.theChar:getX()
— keep the camera in the boundaries
if cameraX > cameraXLimit.start then
cameraX = cameraXLimit.start
elseif cameraX < cameraXLimit.finish then
cameraX = cameraXLimit.finish
end
[/code]
Tags: Diodontidae, lessons learned, Love, lua, postmortem
Behind The Dumb Episode 7 is available now! Thanks to the handy extended voting, it is available before voting is over! Of course, again, I filmed it all during the 48 hours, but didn’t edit and release it anywhere close to that time, so probably not valid for voting. But I hope you enjoy seeing what kind of sandwich I ate, in one of the most epic cliffhangers since Sylvester Stallone’s Cliffhanger.
Tags: behind the dumb, timelapse
UPDATE: Now with (spoilerific) screenshots after the break.
Since several people seemed to find The Triumph of Time pretty hard, I decided to give some pointers as to how to best play the game.
Let me describe the core mechanics of the game. As soon as you have built a pylon next to a star, it will extract particles from it at a fixed rate. The larger the star is, the more particles are extracted per time unit. Building several pylons next to a star will not increase this rate, the particles will just be randomly distributed over all pylons. The particles will then go on to more or less randomly distribute themselves along all barriers which extend from their current pylon.
At a fixed time interval (every two seconds, if I recall correctly), the cloud of antimatter will expand in every direction. It can only be blocked by having a barrier which is powered by particles in its way. However, for every grid cell along the barrier which the cloud tries to enter, one particle on this barrier will be destroyed; when all particles are destroyed, the barrier falls.
So here are some rules of thumb for successful play:
I hope this helps a bit. After the break (SPOILERS): screenshots on how to solve the first four levels.
#gallery-1 {
margin: auto;
}
#gallery-1 .gallery-item {
float: left;
margin-top: 10px;
text-align: center;
width: 33%;
}
#gallery-1 img {
border: 2px solid #cfcfcf;
}
#gallery-1 .gallery-caption {
margin-left: 0;
}
/* see gallery_shortcode() in wp-includes/media.php */
Tags: strategy
it’s smart, srslyok so i’ve noticed the games with the least number of votes are python games that i haven’t been able to run because of some weird python error. i would imagine others have similar issues or perhaps don’t have python installed at all. i’ve just read something interesting, uninstalling old versions of python and reinstalling can help with issues. we’ll see how it goes..
wow it worked! i can finally run python games. one less stumbling block between me and games =]
python 2.6.2: http://www.python.org/download/releases/2.6.2/
pygame 1.8.1: http://www.pygame.org/download.shtml
I started making this timeline a few hours into LD14. I have imagined a program which would produce something similar, although I enjoyed the open ended prototyping and creative potential of doing it with pencil and paper.
.
This is why:
My LD12 entry, Cat Stacker, has grown up and now features assorted fruit. Also, instead of the original tetrisy controls, you can click and drag the current piece around with no time pressure (press A and D to rotate). There are 20 levels and it get’s pretty tough. Good luck! (^ _ ^)/
The later levels are so brutal…Just a reminder that you have 5 days to finish your research for Mini-LD #9. The rules are:
Here is the original announcement. Good luck, and see you on friday!
Tags: announcement, audio, MiniLD
That and I was really, really tired.