Lessons learned from the js13kGames competition

1st October 2012 in js13kGames

The js13kGames competition ended, winners have been announced, so it’s high time to wrap it up and summarize all the things. It was a lot of work, way more than I anticipated, but at the end I think it succeeded and I’m quite happy about the results.

I’ve received great feedback about the compo from both sponsors and participants. I’m almost sure I will run the second edition. There are already other interesting options coming related to the competition, so it could be a lot of fun in the next few months.

js13k-ogs If I run the next edition of the compo I have to prepare better – the first time, without any experience, was a hard one. I was building the neccessary functionalities on the website in real time when they were needed. My semi-automatic submit form should be prepared better, but it did the job. At the end of the submision time I was busy preparing the slides about the compo for the onGameStart conference and then enjoying the conference itself, so I didn’t manage to post the winners on the website right after I presented them during my talk – they were posted few days later.

js13k-tshirt The next time I have to find the sponsors though, because I invested my own money into it: bought and prepared 20 special t-shirts that I gave away during the conference, now I’ll have to send a huge pile of books from Poland (Europe) to the USA, Canada and Brasil. I don’t want to loose enthusiasm because of investing more and more money, but I also don’t want to be accused of wanting to earn money on the compo. It was created for pure fun of coding, to learn some gamedev skills and promote HTML5 gaming as a whole.

There were so many awesome games in the competition, it was wery hard for the judges to pick the best 10. There should be at least 20 of them mentioned, because many great ones didn’t even make it to the top10. There is a long list of different prizes, but every single one of them have small amount of copies, so it was hard to distribute them equally – this should be arranged better.

Overall I’m very happy, the compo turned out to be a big success. The people were writing blog posts about their progress, discussing issues on chats, the entries were being featured on many websites. I even had my first interview published. The number of submitted games greatly exceeded my expectations. If I start the preparations earlier and think before I do, then there’s a big chance it will be easier in the future.

Any suggestions will be highly appreciated – help me make the forthcoming editions better. Use the comment form below and write whatever you think will be useful the next time. Thanks!



1st October 2012, 13:52

I didn’t manage to actually submit anything to the competition but I definitely will. What excites me a lot about it is the fact that stuff like this along with onGameStart puts the Polish JavaScript Community on the map. I myself work abroad but I still count myself as part of it. So well done to you guys.


1st October 2012, 14:10

Although ROTOGAMEsq, my entry, didn’t make it into the top 10, I enjoyed the competition tremendously! Compared to other programming competitions, I especially liked the fact that the uncompressed source of all entries was made available.

What you could improve? IMHO not much. Well, it would be nice if there was some feedback from the judges concerning each game. But then contestants may start to argue… Or, just offer comment functionality on the page for each entry, for anyone.


1st October 2012, 15:29

Vic – thanks. Can’t wait to see your entry :)

Felix – that could be an overkill for the judges to write something about every entry, but the idea about the comments sounds interesting. Thanks for suggesting that.


1st October 2012, 20:25

I loved pretty much everything. The ZIPped rule (instead of people going crazy with eval or PNG compression). The judges (instead of a public voting). Just two wishes. First, it would be cool if we could update our entries. Second, I wish the judges wrote a little bit about at least their top 10 entries. Like the 140byt.es stuff this compo is about learning. So many of the participants (like me) aren’t professional game developers. I think we can learn from the judges when they tell us why they liked an entry.


1st October 2012, 20:38

It was possible to send the updated, zipped package only thru email, because I didn’t want the people to send unfinished games earlier and then adding missing features. I’ll think about the updates though, maybe limited to a fixed number per game.

Looks like I have to squeeze the judges more and have the descriptions at least for the chosen games ^^

Michal Budzynski

2nd October 2012, 14:05

Great idea guy, I’ll try to write something about my favorite top entries today/tomorrow/next week/when I’ll find some time for it. Thanks for attending!


2nd October 2012, 18:03

Great competition! I had a lot of fun doing it.
Besides what you mentioned it seems that posting games as they came in really messed with the dynamic of the compo — early posted games had an advantage with the Facebook/Twitter/Goog+ awards and being able to stand out more in general and people who post games nearer the end have an advantage because they can see their competition while they’re still working on their game.
My 2c anyway!


3rd October 2012, 09:26

Doug – it was on purpose, because I wanted people to post their games earlier (that’s why the idea of Social Special prizes), not on the last day. It didn’t succeeded, because more that the half of them came by few hours before the deadline and I had a tough time (had to take few days off from work) to manage all of them.
It was your choice – you could post the game earlier and have the bigger chance of winning the Specials, or work till the end, look at others, learn from them and post the game the last day. I didn’t find any obvious rip offs or copying whole fragments of other games though.


3rd October 2012, 15:27

I cannot see how this compo could have been pulled off better, especially given you’ve been doing this for the first time and all. Big ups to the organisers, hoping to see / take part in next editions!

What I did like in particular:
1. Making an entry – took me a while since I code like a bitch at best. (Sorry for such an impudent advertisment of my skills) Preparing my entry meant I didn’t have to think about regular work so much, so that was a big big plus. I like it but it gets boring sometimes :P
2. Testing other entries and comming up with ideas of what could have been done better about them. (I like to analyse stuff a lot :p)
3. Actually playing some entries. (More on this later)

Now let’s spread some justified and objective hate.

The problems I had with the compo is the quality of work submitted. Most of the entries are not really valid 13k entries. 13k zipped means you can produce roughly 70 kb of unminified code. Most of the entries are far below that. By looking at the requirements (no restricted theme, timeframe, relatively big filesize) one would think some decent entries were to be expected to pop up eventually. We waited.. and waited.. And waited.. And almost nothing happened. 90% of the games submitted were invalid junk made in one day or so and without much thought. So selfish! ;P

I ‘played’ all of the entries and if I would to choose the top 10 I’d fail at around 5 or so. Don’t get me started on the spirit of the compos in general and all that. The compo lasted for a month for fucks sake, there were interesting prizes announced, yet lots of garbage was sent and posted. And I don’t blame the judges for saying it was great. You got to be polite as an organizer. Accept all entries, tell they’re awesome. Encourage others to make something. Make them send you more low quality crap ‘because if they liked that entry, they will also probably like what I can come up in ten minutes, too’. Or something. I don’t know. I wouldn’t :P I’d be frustrated, I’d visit every developer that sent me a shitty entry and kick their ass :P

Now, a few words about the entries.
First Place: Space PI – One of my favourites – has some replay value, slightly imbalanced though. I hate how it eventually transforms into a deadly clickfest in the last couple of levels no matter what upgrades you choose on your way. Not sure if it was intentional.
Second Place: Sucker – Being an author of this entry I’m quite satisfied where it ended. Needless to say I love the gameplay. It’s just bloody addictive :P
Third: Mindless – Hmmm that’s the most advanced 3D tehtest this compo experienced, however, it’s not really a game.
Fourh: Simland – Pretty awesome idea. However, having not to much history of playing such games the game itself wouldn’t hint me on what to do in order to build up thriving economy. I got discouraged pretty quickly after my people hated and killed me a few times that I tried ;(
Fifth: 13 Tanks: Great concept, pity it didn’t get developed any further.
6: Johnny Smiter: 13th Knight – Not a game.
7: Timber Terry – Why do you have to chop the fucking tree every time you fall is a sweet mystery to me haha :p
8: At Sea – I got lucky in this one and caught a $1.000.000 worth fish pretty quickly.. What was it called.. I can’t recall. Something starting with a… AHAB. Anyway I upgraded my ship, cruised a little bit with my magic wand and what experience that was!:P
9: Hum: In my minds’ eye I can see gay smartphone nerds sitting all over and tapping their screens in order to feed their bird if you know what I mean.
10: Sorades 13k: Boring as fuck in the beginning and impossible to beat in the end. And I’m a proud official world record holder on twitter as far as I know hehe :p

Twitter Special: Johnny Smiter: 13th Knight – Not a game
Facebook Special: Blob Rising, haven’t played, however, with such amount of likes I wonder how it’s possible for it not to be in the top 10. The score is either cheated or the judges couldn’t cope with amount of junk they got sent and got all the places mixed up :P
G+ Special: Pocket Rocket – Same as above. Pretty hard if you play with a tank rocket which I was unfortunate to pick in the beginning plus has a small game design flaw that frustrated me a couple of when I kept spawning in obstacles later in the game without a chance to react obviously :p

A cherry: Freebird: I dare anybody to beat this game without cheating. If you do I’ll send you a beer :P Or you can drink it with me. Whatever works! ;)


3rd October 2012, 23:15

Seen from outside, the compo looked very well organized. Way better than others I have seen in the past, even though that wasn’t their first edition. So you managed to make it a success on the first attempt. More words of praise for that !

As leekish said, 13k zipped is a lot, and some entries could have been polished by putting the remaining space to use : intro or ingame music, sound effects, improved graphics … I plead guilty on that too, as I managed to pack my entry into 10k (submitted version is 12.5k, I knew I could shave some extra bytes but the deadline was closing on quickly). That being said, repeatedly squeezing your code until it shrinks below the limit represents a great deal of efforts in a 1k compo. I would not imagine pulling that on 70kb of code, so having it zipped sounds reasonable.

I started by reading the rules throughly. And interpreting them a bit too much : “extra points if your game supports any mobile devices” had me believe that there were some kinds of scorecards, that the jury would be filling based on pre-defined criteria : +2 for running on Galaxy, +2 for IPhone, +3 for having sound, +1 for having configurable controls, -5 because it is a crude copy of Tetris / PacMan / whatever. Following that (wrong) idea, I focused on adding more polish (sound fx, transitions, hall of fame, end sequence, resolution-independant display, ..), probably at the expense of gameplay. Next time I shall remember to ditch my idea of redoing Need for Speed in WebGL if the judges do not like racing games :)

Continuing on that idea, I definitely agree with Felix and Thiemo that feedback from the judges would be a great help. Especially for those of us who did not make it to the top 10 and want to improve – so we know what aspects to focus on. +1 for opening comments on individual games once the results are out.

In addition of popularity contests, I would suggest fun titles such as “Best demo” (great gfx/animations but not really a game), “Unpacker frenzy” (13k packed, 1 Gb live), or “8-bit special” (looks like a C64/CPC/MSX game).

Last point – maybe a bit towards levelling the playfield, but I reckon that providing a common API for sending scores to the hosting website could be a nice addition. So an entry page would also feature a hall of fame, showing the best scores obtained by players on the game.

Leave Comment