SGF utils

As a software developer I spend great part of my time inside a text editor, in particular, Emacs. Fortunately, Go game records in SGF format are nothing but text, so I wrote some utilities to help dealing with these files.

For example:

• I like to rename my SGF files in a consistent way: 

[date]_[black player]_[white player]_[result].sgf 

This involves parsing the SGF file and searching for the fields [DT], [PB], [PW] and [RE].
To do this I simply put the cursor over the file when in Dired mode and type

M-x sgf-rename-file-dired

where M-x means pressing the Alt key and the x.

• Selecting all files in which I won or lost. In Dired mode type

M-x sgf-mark-wins

SGF files in which I won are selected

• Rotate/mirror the board. 

Something weird that I noticed in IGS: if you download the file directly from GoPanda2 (Pandanet client), it's rotated compared to the one you get when downloading from the web.


If curious, you can grab the code from here and here. It's in two files because one of them has the utilities, and the other is an incomplete mode that I started, that simply fulfills my needs, such as syntax highlighting, for example.

IGS Blacklisting

If you were wondering how to blacklist somebody on IGS, this is how: (and yes, it deserves a post since I find it a bit contrived, and discovered it by accident, not when I searched for it).

First click on Member search.

Once you're on the member's page, click on Register in refusals.

The Refuse option (in the first image) show's all the people in your blacklist.

I don't like players who run away by disconnecting from the server when they're loosing, so I add them to this list to avoid playing with them again.

A similar method applies if you wish to add to your favorites list, or request friendship.

Pattern search

I added to the links section on the right, an entry for the excellent Waltheri's go pattern search web application. It's very useful for searching go positions such as joseki and fuseki in professional games. When studying, if you aren't sure how some joseki is played, just create a position on the board and you will find all occurrences of given position in pro games with continuations which have been played.

The site is not only beautiful, but works very well, very responsive, and easy to use.

Vice-versa

The following two games were played almost simultaneously against the same player on DGS. Both games turned out very territorial, no big fights were involved. In the first one, by the endgame I was ahead by about 10 points, but I left a defect that was well exploited by my opponent, who turned the tables and won the game. The same thing happened on the other game but the other way around! (well not exactly, by the endgame I was also ahead but by only 2 points). At least we both got our share, haha.

L&D Fundamentals

The following problem was taken from the book Life and Death from the Elementary Go Series by Kiseido (page 59, problem 4). It has at least two solutions, but one is better than the other.

1st solution: Black lives with 3 points
2nd solution: Black lives with 4 points
It's pretty uncommon that once we find a solution to a problem, we keep on searching for better solutions. So this is to remind myself of that fact. Also, I realize that my search for a solution starts from the opposite direction, instead of first checking if closing the border works (which gives better follow up moves), I first try to make eyes from the "inside". This behavior is a pretty common pattern so that's why I wanted to point it out. And now that we're at it, why not remind ourselves the Fundamentals of Life and Death:

Life

1. Get more room (widen your eye space). 
2. Occupy a central eye making point (vital point). 

Death

1. Reduce the enemy's room (narrow his eye space). 
2. Occupy a central eye-making point.

Walkover

A few times I won games on IGS without playing a single stone. My opponent probably accepted the challenge by accident and then didn't show up. Not only do I find this unfair, but it bothers me that my rank increases for reasons that have nothing to do with merit.
Why would anybody do this on purpose, and so deserves punishment?
Additionally, the game doesn't appear in my list of games, even though it added to my rank. Isn't this inconsistent?

Fortunately when this happens, one must only wait a minute until the server declares a win by timeout.

Hikaru no Go manga

Yay! My nephew came back from the US and brought me the first three volumes of the manga series Hikaru no Go.

Even though I watched the TV episodes, I still find the read very entertaining, and it pushes my desire to play even further. Additionally, I can study the go diagrams in detail, it's astonishing the hard work the authors put into this.
Unfortunately it will take a long time until I complete all 23 volumes, since it's very expensive to get them all the way to Argentina. I know there is a digital version available, but I like the real thing.

Lizards

Lizards have a great lesson to teach us Go players. They have the ability to drop their tail as a defensive mechanism. This is known as autotomy and literally means “self severing” or “self amputating”. A tail that has been autotomized will continue to move for several minutes, giving the lizard time to escape while the predator is focused in on the tail. 

If you don't believe me, check this video from YouTube:

The lesson is quite obvious. Many times when under attack, we hold on to our whole group without sacrificing a part of it, and end up trapped with a huge group.

The following game is an example of this. Things were quite fair by move 100. I was playing white, and had a four stone group deep into black's framework. For some unknown reason, I decided to rescue it by running away. While doing this, I helped black destroy my own potential territory. The runaway would cost me the game. In a probing (desperate?) move, I tried to take the offensive with move 158, that I can't recall as a tesuji, since the damage it produced could have been easily mitigated, if my opponent had only remembered our lizard friends. Instead, it led to a succession of mistakes, one after the other, and I ended up killing a huge group.

Real life tsumego

This time I was playing white and failed to kill the black group. The actual game can be downloaded here.

Real life tsumego

This post is the first in a series of real life tsumegos I encountered in my games, in general those in which I failed to answer correctly. By repeating them here I hope not doing the same mistake again.
This one is so easy but I missed it! In fact I had to think twice before posting, but after all, this is my road to go...

The actual game can be downloaded here.

Addicted!

Recently, I'm playing a lot in turn-based Dragon Go Server. I always used to laugh about people when they couldn't take their eyes off the smartphone, checking out the latest comment from whatever social network they're stuck into. Well, that is me now when checking for my next move. I'm running about 20 simultaneous games, and I guess I'll leave this as my top limit, if not lower it down to 10. The problem gets worse because many of my opponents are as addicted as me, so they play many moves a day. Fortunately, sometimes I get a message like MySQL error: Too many connections. That forces me to wait for a while.

What I also find ironic is that instead of taking myself a lot of time to think about each move, since I have so many games going on, I have little time to spend on each one! (be right back, I have to answer 3 moves...)

A rule I impose myself is not to try sequences on a board before playing a move. They must be performed in my head. I don't know if doing it is acceptable or not, anyway I don't care if my opponents do it, I don't like cheating myself.

As a side note, I don't know why they call it turn-based as opposed to real-time, I prefer calling it correspondence games, after all, a real-time game is also turn-based.

The following game was played in DGS, in a period of 16 days. I find it interesting because, as usual, I lost an important fight at the beginning, Black was building a huge moyo so I had to dive into it, barely managed to run out, and in the process Black was in a nice position to cash in the whole bottom of the board. Then there was an exchange, I let him into my territory but got into his as well, and managed to cut his invading stones. At this point the game was close, but the final blow came when I killed his groups at the bottom, an he resigned. I guess the merit on my part was taking advantage of my opponent's distractions, if he hadn't done them I would have had no chance. Anyway I had nothing to lose, because the game was lost from the beginning.

Shortest game

It's pretty common that beginner's games evolve from one small battle into a huge battle. Neither want's to sacrifice a stone, nor has a whole board vision of where the big points are. With this in mind, the fuseki ends up being irrelevant, since the game result depends on this one huge battle. The following game in which I am Black is an example of this.

Resign?

Look at the following game position taken from a 2 handicap game in which I played as white, and try to determine who is winning. If you were playing Black, would you resign at this point?

Definitely not! Black is ahead by more than 30 points. Nevertheless, my opponent resigned after I cut his triangle marked group, which is now practically dead. Maybe he didn't realize that my square marked dragon is dead, or perhaps he was led by the anger of being cut immediately after missing the opportunity to kill my circle marked group in the lower right.

Calculating the score is a difficult task, specially for beginners. This example is another proof that we should not resign early in the game, unless it's really obvious.

You can download the kifu here.

Attack and Defense Tips

I've just finished reading the excellent book Attack and Defense by Ishida and Davies, which is part of the Elementary Go Series from Kiseido.

My purpose is not give you a book review (which you can find here.), but an excerpt from chapter 10, that is a pretty good summary of the whole book. If you haven't done so, do yourself a favor and read the book, for a detailed explanation of each mentioned item.

Chapter 1. Take account of the balances of territory and power. If you are behind in territory but ahead in power, play aggressively: invade, cut, attack, and fight. If you are ahead in territory but behind in power, play defensively: play safe.

Chapter 2. When attacking, instead of just trying to kill one enemy group, look for dual purpose moves, or moves that attack two groups at once.

Chapter 3. Attack with non-contact plays. Don't touch what you are attacking. Know the eye-stealing and angle tesujis and the capping, peeping and knight's attacks.

Chapter 4. For defense, use ordinary moves if possible before you are attacked. If you find yourself in trouble, use contact plays and shoulder moves to extricate yourself.

Chapter 5. Play forcing moves, then leave them; treat forcing stones as expendable. Look for ways to resist your opponent's forcing moves.

Chapter 6. Use inducing moves to raise the efficiency of your stones.

Chapter 7. In dealing with enemy frameworks, play lightly and flexibly, taking advantage of weaknesses. Consider relations with surrounding areas (junction points, weak groups, etc.).

Chapter 8. Even within the confines of a three-space extension there are many invasion josekis. Know the meaning of the first two moves, at least, so you can choose among them.

Chapter 9. Avoid adversely one-sided and very indirect kos. In an all-dominating ko, ignore any (non-local) ko threat.

Other quotes that I find insightful:

Fight, invade, and try to generate confusion - the strategy to follow when one is territorially behind is exactly the opposite of the strategy when one is ahead. p.12

At one extreme we have the player who is so jealous of his own territory that he protects it against the slightest incursion; who constantly forgoes attack in favor of defense; who loses through timidity. At the other extreme we have the player whose jealousy engulfs the whole board; who tries not to let his opponent have any territory at all; who usually ends up loosing large groups of overextended stones. These people are deluded. They understand the concept of 'territory', but they lack the faintest inkling of the meaning of 'balance'. p.12

Territory tends to fall naturally to the side that holds the balance of power. p.32

A struggle for power is basically a struggle for eye space. p.34