Code Cowboy

Where has the time gone?

2010-06-19T19:40:00.000-07:00

I originally started this blog when I was writing a chess engine in C#. I thought it would be fun to keep a record of how I figured that out, fun for me anyway. That was 3 years ago...and I failed. I never did complete a working chess engine. Two things contributed my failure:

I went to grad school.
I got a Facebook account.

For the last three years I have written no code for fun, just for work and school. Any spare time I had was frittered away on facebook playing Bejeweled Blitz, Tetris, and various word games.

A few weeks ago my computer broke which is always a fun time because I get to buy new stuff and fix it. This computer is getting old and it is hard to find parts on Newegg for this thing, but I got a new video card, some memory, and a 24 inch monitor. This thing won't run Windows 7 so I decided to go with Ubuntu Linux. Long story short as I was going through the hard drive and backing things up to external storage and I found the source code for that chess engine and it kind of dawned on me that I have stopped writing code for fun.

So I am going to get back to having fun again. I have some plans. I'm going to finish that chess engine, eventually, I hope. I stopped playing chess too, which there is a funny story that goes along with that, but I'll save that for a later post. My first fun project has to do with playing Facebook games or more about how to cheat at them.

Rapid chess improvement, my ass!

2007-02-14T13:00:00.000-08:00

I recently completed reading the book "Rapid Chess Improvement" by Michael de la Maza. It provides a study plan for the "Serious Adult Player", that claims to be able to help a player gain 400 UCF rating points in a year. After reading the book I would like to meet the "Serious Adult Players" that can actually follow the plan, becuase I don't think it is a feasible plan for any adult who might be married, a parent, and works full time.

First I'd like to say what it is I liked about the book. I found the book to be written clearly and well focused. It was a quick read that I think any english speaking beginner should be able to understand. de la Maza has an enternaining style to his writing which helps the reader digest the material.

The main problem I have with the book is the plan itself. First the book outlines some simple exercises to improve "Chess Vision", and then describes the "Seven Circles" plan in which you complete about 1200 tactical chess problems in 64 days, then repeat the feat in 32 days, then 16, 8, 4, 2, and then finally do all 1200 problems in 1 day. The premise of the study plan is that lower class players will get the most return for their efforts out of studying tactics rather than the traditional path of studying opening strategy, middle game tactics, and endgame all together. It makes sense, and I believe everything the author writes about how the plan works. Where I disagree with the author is that any adult can do this, or maybe the definition of 'serious adult player'. If you want chess to become your life, then this plan is for you.

I'm a serious adult, I'm just not a crazy adult. I attempted to complete the first circle, which is supposed to be the simplest circle to complete and failed on my first try. This plan requires that you follow it religiously, and make up for lost time for days that you don't do problems. It became clear to me as I progressed through my problems that I was not on track to finish in 64 days. My main problem is that I can't do problems every day, and this program requires that you devote several hours a day to doing problems for over 100 days straight. In fact the last few circles would require that I take time off from my job, leave home, and be hermit for a week doing nothing but chess problems. It reminds me of studying for college finals, which I did not enjoy when I had to do that. The author makes parallels between perparing for chess and preparing for your career. If you want to be good at something, you have to devote the time. I guess I just don't have the time to do it rapidly.

My next problem with the book is that it is expensive for what you get. I realize chess stuff tends to be this way, because of the limited market. As you can probably guess the plan description in not enough information to fill a book, so the author throws in some useless anecdotal support of how the plan worked for him and other people. In the end you pay $20 for a magazine article and some worthless stories.

Looking at the next 10 years of my life, I don't see 100 straight days opening up anytime soon. I guess I'll continue to suck at chess for a while, and take the long way to improving.

An Entertaining Book on Chess!

2007-02-07T14:11:00.000-08:00

Over the weekend I finished the book "The Immortal Game: A History of Chess, or How 32 Carved Pieces on a Board Illuminated Our Understanding of War, Art, Science and the Human Brain " by David Shenk. This was a fun book to read for me because it kind of melds two of my favorite subjects together: chess and history. This is not your ordinary book on chess, because it does really discuss how to play chess, instead it focuses on the history of chess and how it fits into human history. There was an entire chapter devoted to totalitarian governments of the 20th century and how they tried to use chess politically. Fascinating stuff. This book won't make you chess smart, but I would still recommend it to anyone interested in chess.

UCI protocol implementation is working

2007-02-02T11:49:00.000-08:00

Played a game against ACE using the Arena 1.99beta3 UI. Below is the outcome of the game.

[Event "Computer chess game"]
[Site "R0126928"]
[Date "2007.02.02"]
[Round "?"]
[White "adam"]
[Black "ACE"]
[Result "1-0"]
[Time "13:44:25"]
[WhiteElo "2350"]
[TimeControl "300"]
[Termination "normal"]
[PlyCount "27"]
[WhiteType "human"]
[BlackType "program"]
1. e4 d5 2. e5 b6 3. d4 Bf5 4. Nf3 Qd6 5. exd6 Nf6 6. Ne5 c6 7. Bg5 Nbd7 8.Nxd7 h5 9. Nxf8 Kd8 10. Ne6 Kc8 11. Bxf6 Bxc2 12. Qxc2 Rh6 13. Qxc6 Kb8 14.Qc7 {Mate} 1-0

I mated on move 14. This is only significant in that I can now use GUIs to test the engine. Now it's time to get to work on the brains of this thing.

A Milestone Reached

2007-02-01T08:00:00.000-08:00

It has been about week since I started coding and I have actual results. The code for the board representation is mostly complete. The console interface is also mostly complete. I added a stub of an engine that plays random valid moves and now I can play a game of chess using ACE against a ridiculous computer opponent. As horrible as the games are, it is satisfying to have functional results.

Getting to this point is an important milestone that most of my personal projects never get to. I usually get bored or something distracts me and I never get to the point where I actually have something (other than a pile of code) to show for my efforts. I can show this to my wife and tell her this is reason why I have not been getting to bed at a reasonable hour for the past week. This milestone adds fuel for my interest and will keep me going at it, for a little while at least.

Another reason why this an important milestone is that the code is now testable through the interface. I've already identified and fixed several code defects. I saw an interesting bug I have not been able to recreate. While playing against stupid computer opponent (SCO) I placed my bishop on the C6 square directly in front of his pawn on the C7 square. SCO promptly played C7C6 to steal my bishop and a little of the pride I had been feeling up to that point.

Getting to this point was a tougher task than I expected it to be. It took more code than I expected to write to complete my board representation. Perhaps my code is not as efficient as it could have been, but I think chess is actually more complex that it seems. A chess board is simple to describe in words, chess rules are simple and few, and yet I found it difficult to represent them in a software abstraction. Maybe I'm not as smart as I thunk I is.

Early Decisions

2007-01-26T13:54:00.000-08:00

OK, so I got excited and started coding. I figured I could knock out some simple stuff first without much planning, and not get into trouble. I started with what seemed logical and came up with a simple console and board representation. After I created objects for the game board and the pieces it was starting to become obvious that going any further would be a mistake. I need a plan, so I started reading about what other people have done.

Board Represention
I read a good paper on chess board representations by Bob Hyatt here http://www.cis.uab.edu/info/faculty/hyatt/boardrep.html

As it turns out I wasn't too far off in programming an offset board representation. So I've made the decision to continue with the offset representation. I also read another paper on BITBOARD representations (but I can't remember where). I'm just not ready to go there, and I don't think it is necessary to achieve my goals.

Engine Protocol
I've also decided to implement the UCI protocol for the engine. I originally thought I would be implemented an xboard interface, but after gathering documention on the two protocols I changed my mind for the following reasons:

I found the UCI documentation less ambiguous. I think I'll have quicker success with UCI.
I purchased Fritz 10 and Frtiz only plays UCI. If I want to use Fritz to test my engine I'm going to need to use UCI.

Why Chess? Why C#?

2007-01-24T08:40:00.000-08:00

The current purpose of the blog is to log my experiences of developing a chess engine in the C# language. This might seem odd to anyone who knows about computer chess engines, since there are already hundreds of great chess engines freely available. There are even chess engines already written in the C# language. I realize that what I am doing is not original, but that is not really my goal. Here are my goals:

Develop a chess engine that plays well enough to defeat me regularly.
Learn C# while doing it.

Why Chess? I like chess. I like chess a lot. I just can't play very well. I started playing in 1998 after buying a copy of Chessmaster. Since then I start every year with a resolution to improve my Chess game. I have never followed though on that. I kind of hoping that writing a chess engine will give me more insight to the game of chess.

Why C#? I can't say I really like C#, but I started a programming job recently and C# is the main language used, and I don't feel I know the language as well as I should. When I started learning the Java language 5 years ago I started a personal programming project of writing and Asteroids clone to get into the language. It really helped to have a project that I enjoyed on a personal level to help me learn that language and within 6 months I was a Sun Certified Java 2 Programmer. I am hoping this chess project will give me similar results in learning C#.