A massive thank you to the judges for their hard work in playing and reviewing our games! Your efforts are greatly appreciated, and your feedback is invaluable. You picked an excellent winner this year (it was my favourite also).

An even bigger thank you to Appel for organising and running the contest!

I would like to propose that we increase the size limit for next year's contest, so we can fit in all the missing/cut features. How about 4.1k? It has a nice ring to it 

Whilst googling for roguelike inspirations, I spotted this article which lists Legends in its top 10 roguelikes for android:

Thought you might be interested if you haven't already seen it!

Congratulations zeroone! An excellent and deserved winner this year.

I must say, the shear variety of entries was quite incredible. It seems that we covered every possible game genre!

Flywrench was my personal favourite.

Markus describes the technique he used in Minecraft here.

Last year, I played all of the games, ranked them in order of descending awesomeness, then awarded points from my favourite downwards.
I'll do the same this year.

/makes a mental note to write a game about aardvarks next year

I have a Netgear DG834G. I've had it for at least 5 years.
It has only one problem, which it has had since new: when it starts up, it takes an hour before it connects to my ISP. So I (almost) never switch it off.

Awesome!
Do we have a final time for updates? 23:59 CET? GMT? PMT?

This is an excellent game!
I've rage quit repeatedly - curse these sausage fingers - but I keep on coming back to it. So I think you must have the difficulty level just right.
Performance is perfect on my creaky 10 year old PC.

This is very pretty indeed, I love the destructible terrain.
Though, I agree with the previous posters that the hit box for the ghosts could do with being slightly bigger.

I survived for 2 minutes and 9 seconds.
I killed 242 demons.

Great special effect. However I think my eyes are broken now as the whole room is throbbing.

You read my mind! 
I've uploaded another version, which includes a difficulty level. Press any key (er... except alt/shift/control...) to toggle between normal and hard difficulty.
If you don't see the difficulty level in the top right-hand corner, then please clear your java applet cache.

Sounds very interesting... do you have a link to it?

Does anybody else have a problem with applets and caching?

I just submitted a new version of my 4K game. When I tested it in a couple of browsers, they both loaded the old version of the game.
After clearing my java and browser caches, the latest version was loaded.

This isn't really a very good player experience imho: the player must either clear their java/browser cache(s), or risk playing old versions of our games!

Does anybody have a workaround for this issue (e.g. give the jar a different name each time it is uploaded)?

I've just uploaded a new version to the usual place.

This includes tweaks to the balancing and the AI. Hopefully this version will provide more of a challenge!

Here are the changes:

• AI build rate is much faster
• each AI has its own opening strategy for the first few buildings
• each new woodcutter/mine that you build reduces the effectiveness of all buildings of the same type in that village (thanks bernieberg for that idea)
• villages gain background resources 33% quicker

I have a couple more things that I would like to add, but space (in bytes) and time (my free time) are conspiring against me...!

Great stuff, I really appreciate you taking the time to experiment with the code!
I added your logic and played around with it. The game plays out really well if you allow the AI to build up its villages. More troops = more fun, I think. However, if you build 1 soldier quickly, you can take all of the villages on the map before the AI gets the chance to develop.
Nonetheless, I think your logic is a step in the right direction.


Oddly enough, there is a comment in my notes, saying: 'single resource pool versus resource pool per village - latter is more like Civ, encourages each town to be properly developed'... not sure this worked out quite as I originally intended!

I seem to be getting 2 comments regarding the difficulty: from developers (i.e. you guys) - "it's too easy"; from friends and family - "it's a bit hard". Perhaps I need to find friends/family who are more competent game players  

Anyway, I hope to submit a new version with 'improved' AI in the next few days. Time will tell if it really is 'improved' or not!

Now that is a very good idea!
I'm a bit short of bytes, but I'll see what I can do...

Cheers for that, I might tinker with the AI.

FAO everyone else: no peeking at Mike's post

Wow, I'm humbled by all the positive replies, especially the comparison with the excellent Settlers games


Totally agree - it's amazing what procedural generation can do, even with a simple algorithm (as is the case here).
Hopefully, the source code is commented well enough for people to find the 20 or so lines of PG code. Experimenting with this kind of thing is great fun, and is one of the main reasons why I write 4K games.


Good idea - I did start out with this functionality, but removed it to save space. I'll add it back in if I can find the spare bytes!


I fear you're better at playing my game than I am! Care to share your strategy, so I can teach Mr AI?

At one stage in the development, the AI would beat me 90% of the time. Now it wins 33% of the time, depending on the starting location. Either I've become less stupid (unlikely), or the AI has become more stupid (oops)...

Here is my effort for this year's 4K contest - an isometric real-time strategy game called tiny_world:



http://www.java4k.com/index.php?action=games&method=view&gid=463

The game was inspired by this post about random terrain generation, plus a couple of great little strategy games from previous J4K contests, especially this one from Grunnt (I hope you don't mind).

I thought it would be interesting to combine the two ideas - random terrain generation plus simple strategy mechanics -  and see what happened. tiny_game was the result!

A few more obligatory screenshots...

Starting to take over the north of the map:



Er... how did that happen?! I'm about to lose to the AI:



Anyway, I hope you enjoy playing this as much as I enjoyed making it.
As always: comments, feedback etc are much appreciated!

Cheers,
dapy

Great stuff! I love the visuals, and I thought the controls were fine once I got used to them. It would be interesting to compare mouse versus keyboard though!

The issues for me were that I lost the ship a few times amidst the particles, and also the player's shots and meteor debris look very similar, so it is easy to confuse them.

My best score was 80 points 

Great little game!
Performance-wise, the game consumed ~ 95% of my CPU, but it ran perfectly.

Ah, now I understand!
The technique I described is pretty much the same as dishmoth's - it works for 'diamond' isometric, but not for 'staggered'.

The following link explains how you might solve the problem for 'staggered' isometric maps:

http://www.gamedev.net/page/resources/_/technical/game-programming/isometric-n-hexagonal-maps-part-i-r747

Look for the heading 'Mouse Matters'.

Yes - tilex and tiley are the co-ordinates of the clicked tile.

How do you store information about tiles in your game?

In my game, tile data is stored in a 2-dimensional array, something like this:



Therefore I can access the tile data like this:

You can indeed solve this with a little maths.
In my isometric games, I use the following code to calculate where to draw a given tile:


- screenx and screeny are the position on screen
- tilex and tiley are the tile position in the game world
- in your case, TILE_WIDTH = 64 and TILE_HEIGHT = 32
- some games may include a camera offset, but in principal they all work the same way

To determine which tile is selected on screen, you can use simultaneous equations to find tilex and tiley. Basically, make tilex the subject of the first equation, and tiley the subject of the second. I can explain this bit in more detail if you like...
Solving the simultaneous equations results in the following code:

Looks fantastic!
Unfortunately the game locks up immediately, and I get the world's least meaningful stack trace:

My preference would be to keep the existing J4K contest in its current format, and have a separate OpenGL nK contest - perhaps run at a different time of the year, to ensure that the two contests don't compete for entrants.

That will be £50 please.

A big THANK YOU to the judges, and especially to Appel, for their hard work.

Congratulations to jimeowan - The Little Scientist was an excellent and deserving winner!

I take it you've never used IBM's Rational Application Developer then...?

If everybody were as prolific as you were last year, we would exceed that capacity in a couple of years