[go]$$Wc Semeai $$ --------------------------------------- $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . , . . . . . , . . . . . , . . . | $$ | . . . . . . . . . O O O O O O . . . . | $$ | . . . . O O O O O O . O . . O . . . . | $$ | . . . . O . . O X O a X X X O . . . . | $$ | . . . . O X 1 X X O X b . . O . . . . | $$ | . . . . O . . O X O X O O O O . . . . | $$ | . . . , O O O O X X X O . . . , . . . | $$ | . . . . . . . X O O O X . . . . . . . | $$ | . . . . . . . X . O 2 X . . . . . . . | $$ | . . . . . . . X . O . X . . . . . . . | $$ | . . . . . . . X . O . X . . . . . . . | $$ | . . . . . . . X X . X X . . . . . . . | $$ | . . . , . . . . X X X . . . . , . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ | . . . . . . . . . . . . . . . . . . . | $$ ---------------------------------------[/go]
If Black plays at a, he has six liberties, but it is White to move. When White plays, a reduces Black to four liberties after the connection at b.
Re: New: Capturing Races 1
Posted: Wed Nov 23, 2011 4:07 pm
by RobertJasiek
Teigo's research paper on CGT for approach games seems very useful (in theory) but also advanced. I will have to read it again later to understand his cooling by 2 and usage of infinitesemals.
Re: New: Capturing Races 1
Posted: Thu Nov 24, 2011 2:02 pm
by Javaness2
From the title, this is one of those books I would expect to not want to buy. Other similar titles might be "Monkey Jump Workshop" or "Mathematical Go: Chilling Gets the Last Point" I had a quick look at a sample page, and felt confident, after a few seconds reading, that my instinct was correct. This is not to say that the work in question is utter tosh, but rather, it's just not for me. All I want to know about capturing races can be written on a single page.
Re: New: Capturing Races 1
Posted: Thu Nov 24, 2011 2:38 pm
by RobertJasiek
There are big differences between Monkey Jump Workshop (I think no maths or is there a bit of endgame calculation in it?), Capturing Races 1 (the most difficult maths is forming a difference and comparing it with 0; every book on the endgame has more difficult maths) and Mathematical Go Endgames (advanced mathematics from the POV of a non-mathematician), except that they all are books on specialised topics, quite like a joseki dictionary or a book about endgame. For Monkey Jump Workshop, you are too strong but getting monkeys right is necessary for progress from 11k to 8k. No player needs the book Mathematical Go Endgames because it is about proving maths of 1 points endgames and rules. Interesting for theorists.
Re: New: Capturing Races 1
Posted: Fri Nov 25, 2011 11:07 am
by Bill Spight
Robert's example is quite interesting. Correct play and evaluation depends upon the ko threat situation.
(;SZ[19]FF[4]GM[1]CA[ISO8859-1]AP[GOWrite:2.2.21]ST[2]GN[ ]FG[259:]PW[ ]AW[df][cd][cc][cb][ee][ca][fe][dg][dd][bf][cf][ch][ce][db][bg][ag]AB[eg][aa][he][fc][fa][ac][ae][bi][di][ef][ed][ff][ab][gf][hf][gd][af][fd][hd][ah][fb][bh][dj][eh][dh][eb][cj][bj][ad]PB[ ]PM[2]C[***The shared liberties count for the defender, Black. As for how many liberties White has, that's an interesting question. ***The value of this position depends upon the ko threat situation. Let us look at four main variations, depending upon who is komaster and who plays first.] ( ;B[ge]C[*** White is komaster and Black plays first. *** Can White afford to answer this atari?] ;W[de]C[*** Yes.] ;B[ci]C[*** In general, you take kos last. So in general you make kos last. But that is not the case here.*** Can White afford to fill this ko?] ( ;W[be]C[*** No. See variation.] ;B[ea]C[*** It does Black no good to throw in.*** Since White is komaster, we may assume that Black took the ko, White made a ko threat and took the ko back, and then Black played here.] ;W[bd] ;B[da]C[*** Ditto.] ;W[bc] ;B[ec]C[*** Ditto.] ;W[bb] ;B[dc]C[*** Ditto.] ;W[ba]C[*** After an even number of plays White has won the semeai for 12 points, and Black has 1/3 point for the ko stone in atari. The net value is -11 2/3.*** Of course, as komaster White will win the ko in one more net play.] ) ( ;W[cg] ;B[ea]C[*** It does Black no good to throw in.] ( ;W[be]C[*** This is not White's best play. See variation.] ;B[ec]C[*** Prevents the throw-in.] ;W[bd] ;B[da] ;W[bc] ;B[dc]C[*** Seki.] ) ( ;W[ec]C[*** Since White is komaster, he can use this throw-in to avoid seki. However, if you compare this with the position after B3 made ko, you see that White has lost liberties. This line of play is therefore inferior.] ;B[da]C[*** Since White is komaster, we may assume that Black took the ko and White played a threat and took back before Black played here.] ;W[be] ;B[tt]C[*** Since White is komaster, we may assume that Black took the ko, White played a ko threat and took it back, and Black has played elsewhere.] ;W[bd] ;B[tt]C[*** Ditto.] ;W[bc] ;B[tt]C[*** Ditto.] ;W[bb] ;B[tt]C[*** Ditto.] ;W[ba] ;B[tt]C[*** Ditto.] ;W[dc]C[*** A long row to hoe.] )
)
) ( ;W[be]C[*** White is komaster, and White plays first.*** This position is terminal, except that it must be played out before scoring. Also, Black has several ko threats. As komaster White should wait and let Black go first. The position is worth 12 points for White. W1 has gained 1/3 point, on average.*** See continuation for the playout and variation.] ;B[ge] ;W[de] ;B[ci] ;W[cg] ;B[ea] ;W[bd] ;B[da] ;W[bc] ( ;B[dc]C[*** Prevents an immediate atari, but not best for Black. See variation.] ;W[ec]C[*** As komaster White can win this ko.] ;B[tt] ;W[bb] ;B[tt] ;W[ba] ;B[tt] ;W[dc]C[*** White gets a total of 13 points, an extra point because of Black's throw-in.] ) ( ;B[ec] ;W[bb] ;B[dc] ;W[ba] )
)
( ;W[be]C[*** Black is komaster, and White plays first.] ;B[ge] ;W[de] ;B[ci] ( ;W[bd]C[*** See variation.] ;B[cg]C[*** After an even number of plays White has 12 points for the corner plus 1/3 point for the ko stone in atari, and Black has 1 point for the captured stone. The position is worth -11 1/3 point. This is 1/3 point better for Black than when White is komaster, a small difference.] ;W[tt] ;B[ch]C[*** As komaster, Black wins the ko. White has won the semeai for the corner, as the sequel shows.] ;W[tt]C[*** The position is quiescent, except for the Black ko threats.] ( ;B[ea]C[*** Because of the double ko, it does Black no good to throw in.*** See variation.] ;W[bc] ;B[da] ;W[bb] ) ( ;B[da] ;W[bc] ;B[dc] ;W[ec]C[*** Black cannot be komaster for two kos at once.] ;B[ea] ;W[bb] ;B[dc] ;W[ba] ;B[ec]C[*** Black has simply recouped her throw-in stone.] )
) ( ;W[cg]C[*** White cannot fill the ko.] ;B[ea] ;W[bd] ;B[da] ;W[bc] ;B[dc]C[*** Since Black is komaster, this works.] ;W[tt] ;B[ec] )
)
( ;B[ge]C[*** Black is komaster, and Black plays first.] ;W[de] ;B[ci] ;W[be] ;B[cg] ;W[bd]C[*** After an even number of plays we have reached the same position as when White plays first. OC, since Black is komaster she should wait for White to play first.] )
)
Edit: Added a variation.
Re: New: Capturing Races 1
Posted: Fri Nov 25, 2011 1:23 pm
by RobertJasiek
Thank you for the detailed reminder of how much work kos in approach defects create! :)
Re: New: Capturing Races 1
Posted: Tue Dec 13, 2011 12:35 pm
by Numsgil
Honestly kos and all are nice but what trips me up in actual games are joints. Bamboo and diagonal, specifically. Because the defender has a choice of whether they will respond to defend the joint or not when attacked. So it makes counting liberties in an actual game rather difficult, because you potentially have several different counts depending on whether someone can pull the salamander defense or not. I'm slowly starting to figure it out but I'd still say they easily confuse me in actual games and problems.
Hunter's book briefly treats bamboo joints, but it feels more like an afterthought than a core part. I didn't see anything in the table of contents for this book which would indicate a more thorough treatment here, either
Re: New: Capturing Races 1
Posted: Tue Dec 13, 2011 11:35 pm
by RobertJasiek
Bamboo joints belong to a special type of shape defects, which (shape defects in general(?) rather than only bamboo) I plan to discuss in a later volume. There is little sense in knowing one particular shape but not knowing every other shape.
Re: New: Capturing Races 1
Posted: Wed Dec 14, 2011 12:25 am
by daal
RobertJasiek wrote:There is little sense in knowing one particular shape but not knowing every other shape.
Re: New: Capturing Races 1
Posted: Wed Dec 14, 2011 12:15 pm
by Javaness2
daal wrote:
RobertJasiek wrote:There is little sense in knowing one particular shape but not knowing every other shape.
Shock indeed. Most people learn commonly appearing shapes/patterns. Unless you want to turn pro, you don't learn everything. how much you learn depends on your intelligence/dilligence
Re: New: Capturing Races 1
Posted: Wed Dec 14, 2011 7:26 pm
by RobertJasiek
Javaness2 wrote:Shock indeed. Most people learn commonly appearing shapes/patterns. Unless you want to turn pro, you don't learn everything. how much you learn depends on your intelligence/dilligence
How much somebody learns depends on a) his intelligence / dilligence, b) the scope of the available study material and c) other factors.
Even pros don't have to learn everything. Probably you want to say that players preferring to stay weak or improve more slowly can afford not to study at all or study only a little.
A book starting and ending with bamboo joints gives the reader little food for improvement related to shape defects. Contrarily a book teaching much or even everything gives the reader as much food as he wants to have: If does not want to improve at all, he does not read the book. If he wants to improve only a tiny bit, then he reads only the bamboo joint section. If he wants to improve a lot, then he will try to understand everything in the book as quickly as possible.
"Most people learn commonly appearing shapes/patterns." Really? I would even say: "All people." But what does this convey? It distracts from other important aspects of learning: principles, decision making, reading, values etc.
Shock? Are you afraid of becoming stronger?:) The real shock: As usual for my books, also professionals can learn from them (provided they read them).
About shapes and patterns once more: Vol. 1 works almost independently from shapes or patterns (except for the possible presence of an eye) because the basics of capturing races are a matter of numbers of liberties. A later study of weak eyes is more suitable for a shape approach but still numbers of liberties will be what matters.
Re: New: Capturing Races 1
Posted: Wed Dec 14, 2011 9:20 pm
by Numsgil
Honestly I'd be happy to figure out even a single thing in go lol. I always come back to things I "know" and find my knowledge was actually superficial and lacking in some specific, but key, regard. Usually I come to this realization in that reflective moment when my opponent is removing my stones from the board.
Re: New: Capturing Races 1
Posted: Thu Dec 15, 2011 12:44 am
by RobertJasiek
Start with this fundamental: Connect your important strings so as to have life, always verify whether they are still connected and never forget to do that verification!
Re: New: Capturing Races 1
Posted: Thu Dec 29, 2011 5:26 am
by tapir
Is a review written by a reader of the book available?
While I know I make mistakes in capturing races I believe improving my reading skills to a point where I can read out 99.4% of (usually not highly branching) capturing races is more realistic than both learning and correctly applying the full liberty and semeai categorization tree.
Re: New: Capturing Races 1
Posted: Thu Dec 29, 2011 5:35 am
by Magicwand
robert: do you have any feedback of your books from professionals from korea? i can not see why any korean in right mind will purchase your book about ko and capturing race.
