Dear Lifein19x19 Forum,

I'm a former chess player, and I want to learn Go well, and I want to get good fast.

I've read "Go a complete introduction" by Cho Chikun, some further sources on the web, and watched a few master games on youtube.

I tried playing at online-go against the computer on 9x9, but I lose against the 20k bot. And here comes my problem: I don't have a clue what to play after the second move, and I don't know where to look it up. There are almost no sources on 9x9 openings, and this freaks me out - I don't know what is a good move to play! In chess, you have names for the openings. You can look up very extensive analysis on those openings, and you know exactly which moves are good, which are bad, and how you can play. You can find videos of master players who explain all of this to you, you can find large databases with statistics and lots of games. There is chess.com premium membership which has puzzles for tactics, interactive lessons in strategy, and a very deep database for openings. In Go, I can't find anything.

So here are my questions:

Where can I learn how to play perfect 9x9 openings, and to punish my opponent when he makes a mistake?

I have downloaded "Go Education collection" torrent, and I'm currently reading a book on Life and Death. I also want to buy a Go book, just as reminder to practice and get better. What is the best Go book for all levels, or a fundemental book on Go strategy?

What is the best Go Engine and GUI? I have downloaded GNU Go and Panda glGo.

Thank you in advance,
Kind Regards,
Nikola

You can get pretty good in two years. Is that fast enough?




May I suggest starting with the Capture Game, where the objective is to capture one or more opposing stones? You can play on the 6x6 or 7x7.



Oh, you could have such things for 9x9 go, but nobody cares that much.

As for 19x19 go, forget it! The closest thing to exhaustive analysis are what are called joseki, which are mostly about play in the corner. The standard patterns of joseki produce roughly equal outcomes independent of the rest of the board. However, the rest of the board is usually not independent; so joseki are of limited usefulness. Generally people are advised against studying joseki until they get strong enough to make good whole board judgements.

There are named whole board openings, and books written about them. There are certainly players who keep up with trends in the opening, new plays, and new candidate joseki. But when I was coming along, most of us realized that after 10 moves or so (5 moves by either side), we were pretty much on our own, and that was fine with us.



Nowhere.



I know of no go book for all levels, or even a go series, except maybe the Maeda tsumego series, which focuses on life and death. I hear good things about Graded Go Problems for Beginners. Check out Sensei's Library: http://senseis.xmp.net/?BeginnerStudySection

Good luck!

_________________
The Adkins Principle:
At some point, doesn't thinking have to go on?
— Winona Adkins

Visualize whirled peas.

Everything with love. Stay safe.

The pat answer is to start by playing a lot of games. 50 is a typical number. You need to play and lose until you start getting basic patterns in your subconscious so you don't have to think about them.

Regarding openings, there are two things: First, a lot of people don't think of 9x9 games as "real" go, and professionals generally play on 19x19 boards, so theory is a lot more developed for the large board size. Second, while there are some openings that are more popular, openings in go are not regimented in the way that chess openings are. There is too much variation and too many viable possibilities, so you can't just sit down and memorize a set of openings to work with. Go places much more emphasis on pattern and shape recognition and whole board thinking than chess does, but this leads to a lot of situations without a great deal of guidance. There are general principles that apply, but I'd recommend learning not to get slaughtered tactically on a 9x9 board first.

As for programs to play, go is again not like chess. The very best programs use monte-carlo methods, so once they get ahead their play deteriorates so long as they will still win, and even then they are only as strong as a strong amateur. Second, bots have a reputation for playing bizarre moves that don't really make sense by human heuristics, so they're not really a good way to learn to play well. So long as you are losing badly, they have things they can teach you, but I, and I suspect most other people here, would strongly recommend playing other people instead.

Regarding book torrents, the english, or really non-east-asian go community is quite small, and there is not a great demand for books. Even more so than in other fields, it's important to purchase the books you intend to read in order to make it financially viable for those books to exist and for more to be written or translated.

Good go books for beginners might include the Graded Go Problems for Beginners series (the 4th book is probably good through 5 kyu or so, I hear). Janice Kim has a series of 5 books under the Learn to Play Go moniker. Peter Shotwell's Go! More Than a Game may be good to start from nothing, and I've also heard good things about Charles Matthews' Teach Yourself Go!. These are far from the only books available, and you can probably find a lot of material to get you down to 15k or 10k on Sensei's Library.

As for learning speed, Bill mentions getting good in 2 years or so. How do you define good? For a lot of players, when starting they aim at 1-dan, which is probably roughly equivalent to a strong club player in chess. One proverb says that it takes 1000 games to get to that level, but these are 19x19 games with longish time settings, and the proverb should be taken with more than a grain of salt. The fastest I've heard of people getting there is in about a year (I can think of one person I know offhand who accomplished this), but this is very rare and you should not fret if you don't make it there that fast.

Many people play for years without making it out of the kyu ranks (i.e. to 1 dan) while still having an enjoyable time and a sense of improvement. In addition, you will improve much more quickly through the weaker ranks than you will the stronger ones, and will likely get stuck at walls at different ranks where you spend a long time without seemingly improving. These are all normal and nothing to worry about.

If you're used to chess, you might also overly value capturing stones. I know I had a hard time accepting that capturing stones is at most a secondary objective.

With chess, if I can capture just one piece without a sacrifice, or a higher value piece with a lesser value sacrifice, I will have a clear advantage. With go, even in professional games, captures of more than 20 stones occur, and it does not guarantee victory.

It's all about spreading out and fencing off areas of the board, that's where the points come from, and that's what you have to prevent your opponent from doing. All you need is slightly more than half of the board under your occupation. Of course you still have to think about keeping your stones on the board, but in due time that won't seem anywhere near as hard as it does at first.

You should really just focus on tsumego, apart from playing a lot of games. The opening is not that important when you're just starting out. For most amateurs, the real win-or-lose moments tend to come in the middle or end game, where tsumego practice and past experience will hopefully have armed you a little bit.

You really need to build experience playing human opponents once you're past your initial confusion about the rules. There's a clear gap between how humans play and how an AI plays. The mistakes are different, so that knowing how to respond to an AI is almost a separate skill which poorly prepares you for going against humans.

A good book for some simple ideas about the opening on 19x19 is Otake Hideo's "Opening Theory Made Easy" which is probably the simplest book on the subject. It's short and touches the subject lightly.

In general, go players deal with vastly more ignorance, uncertainty and guesswork than chess players. There is too much depth for any but the most dedicated players to fully grasp any part of it down to the little details.

It's quite common for a professional player to make multiple moves in a single game which are not fully understood by the vast majority of amateurs. Some of these incomprehensible moves may be during the opening. This isn't because they're making fun of the spectators, it's legitimately too hard to see. The consequences are then simply to deep in the search tree for anyone not of that level.

My advice is to play people. I am still learning a lot about go myself, and it is a long journey. To improve the most I offer one piece of advice: "play moves that you understand". If you have a plan, or an idea, you can look back on it after the game and understand what went wrong (or right). Looking back on your game and trying to understand what your plan was, and maybe even what your opponents plan was will bring you a long way.

If I have to recommend only one book on strategy it is probably attack and defense. You will keep coming back to it over and over.

Good luck.

Don't play go.

Learn how to do tsumego and start doing it instead of playing.

Then after you learn how to do basic stuff start playing and maintain working with tsumego and playing small 9x9 go.

SKIP the 13x13 and move to 19x19 where you can actually study joseki/fuseki (somewhat similar to openings in chess)

Why on earth would you advise against playing? You'd take out the entire purpose and the most entertaining part.

Even if that's somehow more efficient use of time, and I believe it isn't, taking out the fun part is a great way to destroy your motivation.

Hello Nikola,
Some of my thoughts, for what it's worth:

First, I am not a chess player, but I think your's is a typical chess player's question.
There are no "perfect" 9x9 openings. And go is not about "punishing your opponent" (IMHO).
Get rid of the chess thinking. Go is a completely different game.

Second, I do not fully understand your post: you say you want to become good at go, fast. But then you mainly talk about 9x9. Please define what your criteria are for 'good at go'. Most and best games are on 19x19 boards. In my experience, go is a completely different game on a 9x9 board then on a 19x19 board. So, do you want to become good at go fast on 9x9 or 19x19?

I also do not fully understand your question because the books you mention are about 19x19, not 9x9.

Again: define 'good at go'. I am just a simple kyu, but I know it can take a lifetime to become good at go (if ever....). So, I translate your question to: how to compress a lifetime into (say) 2 years? Time, study, knowledge, intellect, sources, mind, motivation, determination, contacts (players and teachers), physical health, etc: if you are serious, you really have to work on everything.

Short advice: if you want to be good at go on a 9x9 board, I suggest you try to beat Gridmaster/Steenvreter (available on Android/Google Play, perhaps also on iOS?). Next, play lots of games against humans, both in person and online. Later: cultivate some modesty if you switch to 19x19 if you strive to become better on a bigger board.

Good luck and success!
Keep us posted on your progress.

Uhm, pardon me?

My suggestions: play as many games as you can (as others have already suggested), and perhaps begin with the smaller boards so as to learn a few basics of Life and Death, of fighting, killing, and surviving a lot faster. The suggestion to play Tsumego, OTOH, is a good one, though I wouldn't use them as a substitute for playing, rather as a complement.

Playing Go with a real person instead of only doing Tsumego is so much more beneficial and motivating, given that you select your opponents somewhat so that you are not simply crushed every time. Use Handicap stones when playing against stronger players (and yes, I assume some here will begin to talk about how bad it is to use Handicap stones, never mind).

Most of all: play in a way that you find enjoyable! I.e. forgo any of the many suggestions people here (including myself) give if they don’t work for you.

Good luck, and lots of fun, it’s a life-long journey!

Tom

_________________
“The only difference between me and a madman is that I’m not mad.” — Salvador Dali ★ Play a slooooow correspondence game with me on OGS?

Thank you all for your responses. Nice to see that there is an English speaking Go community.

I read that before you start with 19x19 you should get good at 9x9 first. I understand Go is not about capturing, but about territory and influence. The problem I have is not that I get tactically destroyed by the Engine, I can beat Cosumi 0 on 9x9 (and also on all lower boards):

http://www.cosumi.net/en/replay/?b=You& ... ttt&bm=mce

My problem is that I simply get outplayed in the opening, or "encircled" by the enige, every time I try to play a "solid" game, here against Cosumi 1:

http://www.cosumi.net/replay/?b=You&w=C ... eddeaatttt

When I try to prevent getting encircled by expanding more, I get cut, like here on move 12:

http://www.cosumi.net/replay/?b=You&w=C ... cbdbddedff

Or to put in other words; When I play on E5 and any other 3-3 point (like G3), the opponent cuts me on the 4-4 point, every single bleeping time. When I play on the 4-4 point myself, the opponent invades me on the 3-3 point. When I try to be aggressive and attach instead of playing for position, then he outplays me positionally. Every time I play I lose outright in the opening, same against Cosumi 1 and against 20k on online-go.com. I hope you see why this frustrates me so much. I simply don't stand a chance until I learn somewhere how to play strong openings.

-------------

As for learning Tsumego, that's what I intended to do anyway. My goal was always to get good as fast as possible, I get no fun out of playing "headlessly".

In chess there is a book called "My System" by Nimzowitsch, it's the #1 chess classic. It deals with important strategic ideas, and is a must have for any level of player. People often come back to it to refresh their knowledge. Is there not a book like this for Go? I have the feeling that many Go books are written to sell instead of trying to create a truly masterful book.

First game:

_________________
The Adkins Principle:
At some point, doesn't thinking have to go on?
— Winona Adkins

Visualize whirled peas.

Everything with love. Stay safe.

Game 2.

_________________
The Adkins Principle:
At some point, doesn't thinking have to go on?
— Winona Adkins

Visualize whirled peas.

Everything with love. Stay safe.

I guess if I ever had students I would be a different teacher than you are.

I am strongly against using handicap stones, and I would refuse to take any even from a pro (not because I find it humiliating).

Also since he was talking about 9x9 doing tsumego only can really give you a good kick in 9x9. Once you can do tsumego and if you think of it as doing something else, as you come and play 9x9 go you will find it really easy. Then you can improve at 19x19.

As I said, later you need to play and do tsumego but at start I would advise this chess lover to not play go with such ambition because I have seen so many with great ambitions that wanted to play go, only to see them fall because they couldn't understand it.

Tsumego means not only life and death but countless tesuji techniques that once one knows improve the game experience so much.

Do tsumego for the first month or two and then start playing 9x9, you will understand it much faster. Then maintain playing and doing tsumego and skip the 13x13.

Once you come to 19x19 simply play and if you are interested start studying openings.
But never forget to do tsumego, it is the only key in improving.

Also Bonbo, what you are suggesting is for someone who plays go casually and just wants to have fun with improvement being something aside.
Nikola the chess players said the he wants to improve fast, and I am simply giving him tips on how I would start learning go if I had to do it once more.

Third game:

_________________
The Adkins Principle:
At some point, doesn't thinking have to go on?
— Winona Adkins

Visualize whirled peas.

Everything with love. Stay safe.

There's nothing wrong with playing on 9x9, and most of what you learn there will carry over to bigger sizes. The best part is that a game completes more quickly since it takes far less moves. That means you don't have to stare at the consequences of your recent bad decisions for very long.

Cutting and being cut is normal. If you guard meticulously against all cuts, you will end up too small anyway. You should cut more aggressively yourself, actually.

For example:

Click Here To Show Diagram Code

[go]$$B
$$ +-------------------+
$$ | . . . . . 3 . . . |
$$ | . . . . O 1 2 . . |
$$ | . . O . . X O . . |
$$ | . . . . . X O . . |
$$ | . . X . X . X O . |
$$ | . . . . . . . . . |
$$ | . . X . . . O . . |
$$ | . . . . . . . . . |
$$ | . . . . . . . . . |
$$ +-------------------+[/go]

Click Here To Show Diagram Code

[go]$$B
$$ +-------------------+
$$ | . . . . . . . . . |
$$ | . . . . . . . . . |
$$ | . . O . . . X . . |
$$ | . . . . . . . . . |
$$ | . 5 3 . X . . . . |
$$ | . 4 O 1 . . . . . |
$$ | . . , 2 O . X . . |
$$ | . . . . . . . . . |
$$ | . . . . . . . . . |
$$ +-------------------+[/go]

As far as playing Cosumi Level 1 goes, first, take two stones for now; second, never resign.

_________________
The Adkins Principle:
At some point, doesn't thinking have to go on?
— Winona Adkins

Visualize whirled peas.

Everything with love. Stay safe.

Many said many good things, and gave good advice.

As for me, I am just curious. Why do you want to get good, and why fast?
And what do you mean by 'good'? Like world-class good? Like top-in-the-west good? Like club-level good? Like what?
To some, 15k is 'good'. To others, even 5p is 'not so good'. What do you have in mind?

And what do you mean by 'fast'? Go is a complex game, and unlike chess, progress is not all about memorizing opening lines. It is very individual, and so it is hard to say what 'fast' means, but basically it depends on dedication. From my experience, the more dedicated you are, the more you enjoy the game, and the less it is important to you to get 'good fast', since you just having so much fun - and this is when you get the goodest the fastest. Your question suggest something like "I have a particular situation in mind, now tell me what and how much do I have to grind to get there." Which is, to me at least, sort-of counter productive.

So, what is it you want exactly, and why? Really?

PS>
On a practical level, to at least try to answer your question, there are 'opening' databases out there (look for 'joseki' and 'fuseki') as well as collection of games. There are collections of Life and Death problems (commonly referred as tsume-go), and so on... Look on Sensei's Library for good links and articles.

In your situation, however, I think you would profit more from playing more human players and asking for concrete advice after games, possibly even making friends.

And, as somebody already said - get rid of your 'chess thinking'. It does not really apply to go very much for most parts.

_________________
- Bantari
______________________________________________
WARNING: This post might contain Opinions!!

A couple of comments on Game 2.



Main focus: Go for the kill.

Playing Black on the 9x9 and giving komi, it is hard to beat an even opponent without playing aggressively. To play aggressively you need to be able to kill. Might as well learn how now.

If you take two stones on the 9x9, you have good chances of killing something. Go for it!

_________________
The Adkins Principle:
At some point, doesn't thinking have to go on?
— Winona Adkins

Visualize whirled peas.

Everything with love. Stay safe.

If you like reading here's a cool link

http://senseis.xmp.net/?HaengmaTutorial ... ntOfStones

There's actually a lot of cool articles to read

http://senseis.xmp.net/?StartingPoints

just get familiar with the words and looking at the diagrams. When I was first learning seeing all these new things and just reading about them was fun for me. I didn't understand anything. But it was fun.

If you like playing keep playing.

Losing is fine. Keep playing until you recognize when you're in atari. Recognize when you're about to be in atari. Recognize when you can put your opponent in atari.

Play bots if it's more convenient. Use undo a lot if you're getting discouraged. One thing I used to do which could be considered scummy is I would play a move. The bot would play something that devastated me so I'd undo and play where he played. But he'd find another move that devastates me. So I'd try undo again. But I'd have to undo three times before I could salvage the situation.

When playing a lot you start to notice when things are going sour 1 move before it happens.
Then eventually 2 moves before it happens
Then 3 moves before it happens.

And even better you start seeing how to sour your opponents position.
Then two moves ahead of your opponents position etc.

Soon you'll get comfortable with the board.
You'll have reasons behind your moves.(They'll be poor reasons)
You'll have a plan. (It'll be a poor plan)
And then you might want to change learning strategies. Because playing bots and pressing undo develops bad habits but it's convenient to get familiar with the game.

I am familiar with My System. I think the nearest equivalent would be "Lessons in the Fundamentals of Go" by Kageyama. It is at a more basic level, but it covers a lot of topics, and is regarded as a classic that people go back to as they keep learning.

_________________
Regards,

Peter