About LastResortGames (twitter: @lastresortgames)


Ludum Dare 35
Ludum Dare 34
MiniLD #58
MiniLD #56
Ludum Dare 31
Ludum Dare 30
Ludum Dare 29
Ludum Dare 28
Ludum Dare 27

LastResortGames's Trophies

LastResortGames's Archive

Post-mortem: Reaper

Posted by (twitter: @lastresortgames)
Wednesday, April 30th, 2014 5:06 am

You can find our game here.

Sorry, please just play the windows version. The flash versions exist because I needed to export something and I thought they worked… but apparently not.
So, another Ludum Dare down. I am not really sure where to start. So I will start with why we ended up using Haxeflixel this time around.


Karl and I have been making a lot of games in Unity lately. Our last three Ludum Dare entries were all made with Unity, as well as our other four main projects, and a few other projects we don’t talk about (you all know the kind). Having a tool that is mostly easy to get started with and has cross-platform support is invaluable. But, I have been wanting to branch out. There are just very few options that I haven’t already tried, that are also cross-platform. Then comes along Lars Doucet, maker of Defender’s Quest. I started to follow him on twitter after seeing him speak at GDC and expectedly he posts about some interesting stuff involving games, his gaming philosophies and how he makes them. His use of and involvement with Haxe/Haxeflixel and its community started to come up a lot recently (or I at least just started to notice). I figured since I already wanted to branch out and it was getting a glowing recommendation I would look into it. It is open source, has cross-platform support and flash-esque syntax, something I hadn’t used in a while but know well enough. And most importantly it looked like something you can learn to use in a weekend, which proved mostly true. So I figured, why not and decided that this weekend I would use Haxeflixel over Unity. Once I decided, I got it all installed and running. This was all actually incredibly easy. The instructions were clearly laid out and easy to follow, especially considering it involved using the command line. And that brings us here.

Day 0:

The theme was announced, and Karl and I started brainstorming. Since we both enjoyed Super Motherload we figured, how about if you are digging UP. This brought me toward you being a ghost, and were buried alive and need to get back to your body for revenge. Then Karl was talking about being a business man in hell buying and selling souls as a means to get back to the surface. Then we kind of combined them and stuck with the undead theme and ended up where we are now with Reaper. You are a Reaper whose job it is to collect payment from those who have sold their soul. You have to run around the level trying to figure out who your target was, and kill them in order for their soul to be collected. We had discussed a few different methods of identification but we just stuck with a simple scan for the sake of time. There were supposed to be power ups but of course, your scope is always too big. The level design and gameplay was kind of inspired by Infectonator (another game you should totally play on Kongregate). At least I thought it was. By midnight Friday I realized I had no idea what I was talking about.

The Good:

  • For the sake of the Ludum Dare, Karl and I actually seem to work best without much pre-planning. This doesn’t seem to be a common sentiment. Well at least people feel like they should say this whether they believe it or not. We figured out a direction we wanted to go and just went answered any questions as they came up.
  • I still had to get a feel for how Haxe and FlashDevelop worked together so a lot of Friday was getting a character running around the screen and level switching.

The Bad and The Ugly:

  • Nothing of note just yet.

Day 1:

This day was off to a rocky start. I forget exactly what I was working on but all I know is that it was not interesting. But the first little bits of a new project rarely are. I was having trouble getting the other people to move around and figuring out how to save information across levels. Conveniently enough there is a file that is made when you start a new Haxeflixel project called Reg.hx with “Handy, pre-built Registry class that can be used to store references to objects and other things for quick-access” and “Generic levels Array that can be used for cross-state stuff”. You guessed it, cross state information goes in here, so of course I probably abused it but it was totally worth it.

By the end of the day I had two levels built and neither had a working person in them but both had a stationary Target. So I could technically win the level. So we had a win condition but, at least at the moment, I just wasn’t sure how I was going to make it interesting. I could kind of see the potential if I squinted my eyes and turned my head a little bit but it just wasn’t there. But I powered on because it’s a game jam and that’s what you do.

The Good:

  • Got in some gameplay and added in easy way to build levels.
  • You can win the game.
  • Got a proper art direction.
  • I am satisfied with my choice of trying out Haxeflixel. (despite The Bad below) The learning curve was not too steep and I was able to get information from my head onto the screen exactly as fast as I wanted when I knew how.

The Bad:

  • Sorry Haxeflixel, but the documentation just isn’t quite there yet. I had a lot of trouble finding the nitty gritty information I needed. It was all technically available but in hard to find places. There is a javadoc like setup but its unsearchable and if you don’t know what package something you want is in, you have to search through everything. And for some specific setup like the Project.xml file I could only find snippets. The demos are VERY helpful but they just don’t cover all of the basics yet. (Sorry if it is there. It is very likely that Google just wasn’t my friend.)
  • We were having trouble figuring out what we wanted the game to feel like.

The Ugly:

  • The game was not looking like it was going to be any fun.

Day 2:

Sunday started out ok. I went out to lunch and designed the rest of the levels in the game. When I got back I built them out and started to get the actual people running around properly. This was still fighting me every step of the way but I am pretty sure I got them working in a few hours. The day was still rather slow but the game was shaping up a little bit. What did make the day interesting was the music Karl wrote. For some reason the title theme he wrote just sounded incredibly Noir. There was no particular reason for it and we hadn’t discussed sending the game in that direction but it just worked. So we went with it and we now had three Noir music tracks for our not quite Noir game.

The Good:

  • Accidentally Noir.
  • Level design was complete.
  • The Game looked less un-fun than it did Saturday.

The Bad:

  • There were still no people or demons in the level and the targets did not move. I would have to come home early to finish before the deadline.
  • We still didn’t know how to lose or what the end game was.

The Ugly:

  • Nothing of note.

Day 3.8 (2.8?):

Of course the last 4 hours are when everything comes together. As soon as I got home I set up the path’s the people would take. It was probably the hardest part of the whole weekend. I forgot to take into account overuse of certain pathways and making those same pathways into a two way road with one lane. People would be running into each other in a reverse tug of war until one group got close enough to their goal to continue on. Some groups were timed so that this never happened so all of the people on the map would be in groups of 10 on opposite corners fighting to move forward. Basically it sucked, but I was able to mostly fix this issue in time.

So around the 2 hours left mark I was finally able to properly test all of the levels. I had done the first two levels a million times by that point so I skipped a few. So I start the level and start scanning. I find a demon and try and kill it and I miss. For some reason everybody in the game is INCREDIBLY sensitive so they all start to run away. I realize that I still had no idea who the target was so I have to run around the whole map identifying everything as fast I can. I manage to kill quite a few people until I finally get the target and the win screen comes up. At the moment I sat back and realized, “Holy crap. I just had fun.”

The Good:

  • The game is fun! At least I had fun. So I felt a lot better at the end of the day.
  • Got all of the pathing done. Torturous but the system wasn’t completely terrible.
  • Colorblind Mode. It was a last minute decision but it’s something that comes up every Dare that our game has never really required (not much was ever color coded). But this time I realized it was and I figured it would be appreciated. Hopefully it can actually help and doesn’t just make us look like idiots.

The Bad:

  • Making paths is hard.

The Ugly:

  • Always test exporting before you need to. I exported the same flash file a dozen times. It was the wrong name, put it into the wrong folder, it looks terrible in a browser and I realized I had no idea how to embed it in a web page. That was rather unfortunate. I was also not able to export to windows because it does not support .mp3 files. Suprise! So no .mp3’s means no sound, so I decided against the windows version. All at T-15 minutes. It turns out I could have gotten a windows build working in that time since all I needed was .ogg’s but I was kind of freaking out a little bit so it was a better choice to just let it go and make a windows build Tuesday.


Something that I thought was really cool is that the Haxeflixel guys keep a list of games that used Haxe for their Ludum Dare submission. You can find it here:

I just think it’s kind of nifty.

So overall I am happy with how the game turned out. I was worried Saturday but that was fixed Monday night. The music seems to be popular so far. It turns out I was able to quickly make a windows executeable without changing very much at all. Haxeflixel has very easy to use conditionals for when you build to certain platforms that took about two seconds to replace to add ogg’s to the windows version and keep the mp3’s for the flash version. There was a bug that was specific to the windows version but it was also because I was doing stupid stuff in the code and the flash export let me get away with it. Hopefully having a windows build will increase the odds of people knowing what they are supposed to do and allow them to do it.

I overall DO recommend Haxeflixel. It was a bumpy start and I did have trouble finding stuff that I needed but I was able to find it most of the time. I did only spend a weekend with it so I obviously am not going to know where everything is but I also was able to make a functioning game having never worked with it before. So next Dare you should totally try it out.

And I am also kind of curious if anybody reads these. So I want to see if we get any ratings after we post this.



I am going to talk about the “artistic direction” that I went with for this game.  So Friday night, Alex is showing me Infectonator and I’m like “okay” but we end up doing full top-down anyway.  Well, it’s hard to express people from a bird’s-eye view, so I figured we may as well just use colored circles.  It’s just a game jam, after all.

Since the people are all colored circles, may as well make all the other assets simple, as well.  Circles and squares, circles and squares.  But, in the end, Alex didn’t end up programming any circle props into the game.  I also made a car and truck (rectangles!) but vehicles were not implemented.  Had they been, they probably would’ve just been obstacles/death hazards.  I still think it would’ve been an interesting idea, something to spice up certain levels.

The concept for the title screen just came to me.  It was going to be a grayscale image with the people tinted to match the colors of the game.  But originally, I wanted a certain angle from behind a corner, looking at a crosswalk in the background.  In the foreground, in front of that corner, would be the player avatar.  Unfortunately, such a specific camera angle does not exist in the first 20 minutes of Google image search, so I had to make due.  Flickr creative commons search eventually gave me a picture I thought was suitable.  When it came to the art for this game, I always kept in mind to “keep it simple, stupid.”

Musically, I just wrote whatever came to my head.  There wasn’t really any particular musical theme I was going for.  I mean, it’s about reapers, and demons, but in a city?  What am I supposed to do?  The first track I started to write, I liked, but it definitely didn’t fit.  It was kind of a weird rock track with a Flight of the Bumblebees motif?  (I just now remembered we used Flight of the Bumblebees as the BGM for a different game jam, Life & File.)

So the next track ended up being the title track, and it was kind of electro-Noir?  And I was just like, “okay” and Alex was all, “okay.”  The menu theme was written next but it’s really just an 8 bar pad loop with a bass synth, so that one isn’t even Noir.  But the level track, it’s kinda frantic break-beat electro-Noir again, so I think that came out well.

Noting the comments, there’s been quite a few kind words about the music already, which I appreciate, but is it because the music is actually okay, or is it because not a lot of game jam games have music? Either way, thank you.

With my portion of the game discussed, I’d like to critique the game.  This is probably a habit to break, but because Alex and myself use different platforms, I hardly ever play the games we make while Alex is making them.  I’ve actually just now for the first time played the game before writing this.  Of course, I made sure to 100% clear the game (beat the 8 stages with gold medals).


  • Somehow, the game is fun.  From the start, I wasn’t sure I would like the game, but it actually is pretty fun.
  • I want to make a point that adding the names of the people you’ve killed (or the person you were killed by) is great, I think that’s a good feature.
  • There is artistic consistency.  Well, I think there is.  It’s simple and clean; that’s the way that it’s making me feel tonight. It’s hard to let it go.
  • The music is okay?


  • The controls are legitimately hard to use.  I made the call for an Asteroids control scheme, but now having played it, it’s just too much.
  • There are 8 levels, but they’re all basically the same.  There wasn’t that much variance between them as far as I could tell.  This might be because of time constraints; it seems Alex barely got anything working right until Sunday, so it must’ve been a struggle.  I’m sorry I didn’t notice.
  • However, this is also why there are no sound effects.  But now I wonder if having the crowd ambiance overtop the music would’ve been too much or not…
  • When you collect all 8 medals, there’s no reward.  Not even a congratulatory screen.  I should know, because I suggested adding medals to the game at the last minute without implementing a reward.  But, having cleared the game, and receiving no reward, I understand that that’s wrong.  It feels bad.  Anything would be better than what we have right now, because the game doesn’t even acknowledge you’ve cleared it.


  • Alex ruined the Flash export.  It’s horrible.  He knows.  I know.  But it’s still kind of embarrassing, even though I had nothing to do with that.
  • A lot of stuff breaks when you end a level (either by killing the target or being killed by a demon).  The level is over, so it doesn’t really matter, but maybe it would’ve been better to go to a black screen with the death-list overlaid on top.

– Karl

Postscript: I’ve been spelling it as “Ludem” for a while now.  Whoops.

I’m in.

Posted by (twitter: @lastresortgames)
Wednesday, December 11th, 2013 1:27 pm

I will be doing my first solo jam as my partner is busy this weekend. Hopefully I can handle making a full game as well as art and sound that won’t make anybodies eyes and ears bleed. It is going to be an interesting test in making a small enough yet quality game that I can finish in 48 hours.

Turnip Archer:Post-mortem

Posted by (twitter: @lastresortgames)
Saturday, September 14th, 2013 9:58 am

So this is a bit late in the game but I figured I should write one of these. I wanted to make about the same thing with some more RPG elements. I was able to talk it out and realized that the fast paced gameplay we wanted would only get bogged down by having to make decisions and enemies having a real amount of health. We considered the Bloons approach where certain enemies just take damage and change into lower level enemies but we ended up running out of time for that. So overall I am satisfied with the final product although I don’t think the game really has the level of ‘fun’ we had intended. This is also our second Ludum Dare and I think we improved how we handled it.

What went right:

  • We managed to finish at a comfortable time. We spent most of the weekend on it and then had plenty of time left over Monday to polish what we could.
  • I learned a little bit more about Unity. Like Prefabs are just GameObjects and can be ‘Loaded’ as such dynamically. This was a fact that eluded me for a longer time than I care to admit.
  • Karl and I as a team were able to further flesh out how we like to get things done in general. We have kind of been going willy nilly lately and it was a satisfying change.

Graphics: I am satisfied with the art style that we figured out. Karl did a great job as usual although we did end up re-using assets from a previous project for the Spider’s.

Music: Also on Karl. I think it worked out nicely. He always manages to surprise me.

Back end: I think I could have cleaned it up a little bit but overall I think I maintained a coherent structure.

What went wrong:

Theme: I was very unsatisfied with how we used the theme. Mid Saturday I had read about people discussing how 10 is any number you want it to be if you change the base and I realized that I had limited myself a bit too much when trying to think of a game. I still personally found the theme very restricting even considering that fact though. This obviously affected our game and design in a negative way. Next time I will more actively give the theme a better chance.

Level/Round Design: Overall, personally I am not very happy with how the waves ended up working out. 10 seconds was just too fast but it was our main tie into the theme. Again I wish I had thought about the theme further. I don’t actually play games like mine very often so I had a lot of trouble thinking about a good way to prevent wave overlap while still keeping the game at a satisfying difficulty and pace. I think my ineptitude at creating the rounds led to a fun experience anyway but I was not very confident that it would upon release.


Late Saturday I was realized that I just wasn’t happy with the direction the game was taking but it was already too late. I just didn’t think the final product was actually going to be fun. We kept on trucking and Turnip Archer came out. By Monday when we submitted it I realized the game was just fun in a different way than I had expected. It could still be better and the rounds could have been cleaned up but it’s a game and people seem to have overall positive feelings about it. And I was able to publish it to Kongregate which is a personal first that I am very proud of.

If you want to try it out you can find it here.

[cache: storing page]