Knights of the Cardboard Castle

This post is a detailed walkthrough of my bot’s internal logic, the source code of which is attached. It’s written in Java using Notepad++. I probably should have used a real IDE as it would have helped save time finding compile errors. It’ll also contain some improvements that I’d have added if I had the time and some notes on some of the cool bots other competitors have submitted and their strategies.

Source code in zipped file

Read the rest of this entry »

The 2010 Google AI Challenge is now officially over and I’m very glad to have finished just inside the top 50. Not bad for someone with no formal education in programming and next to zero knowledge of Java before this. I learned a bit of JavaScript when I worked as a content writer for one of the very many companies in Malaysia that tried to make its own fortune during the original Dot Com boom. I basically learned Java just for the purpose of entering this contest and picked it because it seemed like the best supported language and uses syntax that’s not too dissimilar to that of JavaScript.

I never expected to do as well as I did. I am by far the highest ranking participant from Malaysia, though this is because there were so few participants. I tried to spread the word around on Facebook and LYN, and I think a few people joined because of that, but none of them made a serious attempt at writing a decent bot. Rather surprisingly, I was also the second highest ranking participant from any Asian country. Top Asian goes to oldman of China who showed up in the top 100 a few days after I wrote this post on my other blog. Other than us, there was one Korean in the top 100 and no less than five Indians.

Read the rest of this entry »

It took some time, but we finally got around to playing Britannia recently. I first read up on the rules over a year ago when we planned on playing this with Han but we never did manage to get our schedules right so we missed out on it. Nevertheless, I never forgot this game and was glad for the chance to play it when Sean brought it up again. The problem with this title was always that it needed exactly four players and that its expected playtime would exceed four hours. In our case, we started rules explanation at about half past seven in the evening and finished well past one the next morning.

The thing with Britannia is that there are only a handful of rules. Combat is handled with a single die roll per unit, much like the Axis & Allies games, and most nations have only a single type of unit. Controlling more territory does give you more units over time, but once again the mechanics are simple. Controlling a normal territory gives you two Population Points. A territory on difficult terrain gives you one. For every six points you accumulate, you get a new infantry. Movement is probably the most complex. Basic movement is easy, two territories if moving through normal terrain, one if moving to a difficult terrain. Things like boats and moving across straits add more options but aren’t too bad.

Read the rest of this entry »

We first played Coloretto some time ago when I expressed interest in Zooloretto. Sean however insisted that playing the original Coloretto would be sufficient to cover the essence of both games and the newer and bigger one simply added a ton of extraneous stuff that doesn’t fundamentally change the game. So we duly played the older, found it okay but unspectacular, and like almost all filler games, quickly forgot about it.

We recently got a chance to play Zooloretto when Chee Wee had more free time to join us in our boardgame evenings and despite Sean’s reluctance to play it. All in all, we enjoyed it and liked it quite a bit more than Coloretto because it’s nice to have a solid theme and playing with animal tiles is more fun than playing with abstract cards. But Sean is still right that the core gameplay remains the same. My thoughts:

• It’s effectively a very random game and what strategy exists is pretty shallow and obvious. It’s still a lot of fun in the way that waiting to see the results of a die roll is always tense and exciting. The core of the game is about how far you’re willing to push your luck. Do you take a single animal that you need right away or gamble that more of what you need will turn up later, unburdened by tiles that are useless to you?
• This is also a game about screwing over your opponents, so it might go down badly with players who are easily offended. You can pick a truck on the turn that you decide to draw a tile, so you’re forced to put the tile on in the most disadvantageous way possible for the players after you. So instead of thinking how best you can help yourself, you spend most of your time in this game thinking about how you can hurt your opponents the most. Very nasty.
• One important difference from Coloretto is that you have to decide which enclosure you want to store your animals in, which is equivalent to declaring in advance how many of that type of tile you intend to collect. Obviously, you’d want to put animals that no one else has started collecting in a big enclosure, and put animals that you’re going to have to share with other players in a smaller one.
• I was surprised to see how important the snack stands and money tiles are. They’re universally useful to everyone, by contrast to the normal animal tiles can be very useful to the right players and very bad for others. So implementing them in this way seems to spoil the elegance of the game a bit.

Overall, I find this to be a lightweight game that’s okay in small doses but would likely boring quickly. I suppose it’ll always be good as a game to play with kids. I still like this more than Coloretto because it’s fun to play with more bits, but I concede that it’s hard to justify the price and quantity of components needed for this level of complexity.

As a final note, I was originally interested in this because I find that I quite like the “tycoon” style themes in which each player has build something up on his or her own player mat. Examples might be Agricola, Vegas Showdown, Shipyard as reviewed by Hiew etc. When I suggested the movie studio theme Hiew said that this already existed in the form of Reiner Knizia’s Dream Factory and Allen suggested Restaurant Row as another obvious example of the genre. Sadly, Zooloretto doesn’t really have any management mechanics so it doesn’t belong in the genre so I contend that a good zoo management game has yet to be made.

Duels of the Planeswalkers is a stripped down video game version of Magic: The Gathering made to be simple enough to appeal to the console crowd. The game uses preconstructed decks and there’s no real storyline or anything. The single-player campaign consists of a series of duels against the AI.You start out with a small selection of decks against easy enemies, but over time, you gain access to more decks and face off against enemies with better decks as well.

There’s no real deck construction in here. As you win games, you unlock additional cards that are automatically added to the current deck you’re using. You can choose to include or exclude these unlocked cards from the deck but you’re never allowed to remove any cards from the base set. You’re not even allowed to tweak the land distribution. Land cards are automatically added to the deck depending on how many extra cards you choose to include in it. My thoughts:

Read the rest of this entry »

Considering that this game has a 2002 release date, it’s perplexing how little attention this game has on BGG, judging by the number of posts in its forum. Perhaps it’s because it doesn’t yet have an English version? Interestingly as it was also designed by Michael Schacht who we have to thank for the excellent China, this game has a Chinese theme as well. Here, the players seem to take on the roles of merchants who are trading on the junk-based floating markets of Asia. The object of the game is simply to earn the most money possible.

Personally, I had a hard time learning this game. It’s abstract, yes, but at least in something like China, it’s possible to imagine what the pieces and situations might correlate to in the real world. Here, it’s pretty hard to figure out it is that you’re supposedly doing beyond playing a game with some pretty arbitrary rules. I know that technically this is true for all games, but I find it much easier to understand what’s going on if your actions and roles actually make some sort of real-world sense.

Read the rest of this entry »

As I’ve mentioned before, I tend to do very badly in area majority games. As these games are played on a map, the positioning of the various playing pieces relative to each other becomes very important. Usually, you need to strike a balance between occupying areas to score points, denying points to opponents and putting your pieces in a better position for future rounds. Due to the many different possibilities, working out the best course of action at any given time can be very computationally intensive. Having the ability to efficiently process visual information helps a lot, and the people I play with regularly can attest to the fact that I’m terrible at this.

This doesn’t stop me from admiring their designs and it strikes me that China is one of the simplest, most elegant designs I’ve played in a while. Its rules take so little time to explain that you might mistake it for gateway game but in actual play, the strategy turns out to be shockingly deep. My thoughts:

Read the rest of this entry »

Like Amun-Re, Taj Mahal is one of the more elaborate Knizia games I’ve played. Like so many of his games, this one has an auction system at its heart. In this case, however, instead of money, each player draws and places cards with symbols on them denoting what they can get from each auction. To make things more complex, the cards come in five suits and once you’ve played a card of a color, you can only follow up your “bid” with cards of the same color. The exception are the white colored cards which act as wild cards and can be combined with any other suit.

A player can choose to fold instead of adding cards to his bid. When he does so, he gets whatever goods corresponding to the symbols that he has the most of on the table. Then all of his cards are removed from the table. There are basically two things to aim for here. The big prize is having the most elephants, which gets you a tile that shows multiple goods on it. This gets increasingly valuable as the game goes on. The other prize are having the most symbols for each of the characters. This allows you to place a palace on the board and you score points by building chains of them across provinces.

Read the rest of this entry »