BQM 510

    Keywords: EndGame

has this endgame technique a name? can we describe it as amplifying - increasing the temperature of the local plays to make B3 sente "prematurely", while if waiting any longer it may end up as a sente move for white? (i ve seen this in a professional game.)

Herman: I'm fascinated, I've not seen it before

Abydos1: How is it not sente since B3 threatens to jump in at B1? Thus, it looks to me that, B3 has the same sente value as B1 (both threaten to connect B1 out).

tapir: The point here is the size of the threat (actually much bigger in the real game than in my diagram, thanks Bill, not the size of the play itself, though it is influenced by whether you can play it in sente or not), which makes this move sente. The score of this diagram and comparison diagram 2 is the same. But you can't be sure whether comparison 1 or 2 will happen. To put it otherwise, of course B3 would be sente at some time of the game - however around this time W1 (in Comparison 1 diagram) will then probable be sente as well - or even earlier. By making the threat of B3 larger (much larger in the game) Black succeeds in turning B3 sente before White contemplates playing W1 (from comparison 1). I see this principle at work somewhere in professional games, but the whole concept of increasing the threat of a move to turn it sente is lost to most amateurs at least at my level. Maybe the next diagram shows this even better, if less spectacular, playing B5 would not be sente since the threat to take one point in sente later is so small, however W5 would be sente if black waits too long. By throwing in some more stones at B1 and B3 Black can keep sente with B5. What I grasp for is the concept of increasing the value of the followup to turn a move into sente. Complete with a nice name to remember it better.

Abydos1: I get the concept of playing a throw in to make the value larger and thus you're able to play it earlier but my point is that since white responds to B1 with W2 then white would respond to B3 with W2 as well since the threat is the same (if B1-pass-B3 connects, same result as B3-pass-B1). Thus both B1 first and B3 first have the same sente value; yes, the B1-W2 exchange makes B3 a larger threat since it threatens to capture the 4 white stones but since we're assuming white responds to B1 then as I see it black should be able to just simply play B3-W2 in sente.

Edit: After looking at it some more I see where I'm mistaken now. If white ignores B1 black can continue at a before having to play B3 to connect; that's not the case if black starts with B3.

Maybe the same concept at work - common endgame sequence  

Bill: The atari is not sente. It does not threaten the placement.

Normal follow-up after black+circle  

(Whether B1 is sente depends on the particular situation, but it is close, anyway.)


W2 is the response to B1. The placement is either gote instead of sente, or a smaller gote than simply taking.


This ko is normally not so good for Black.

Point value comparison:
tapir: Depends on with which diagram you compare. (Identical in points to diagram 2, two points better than diagram 1) The idea seems to be, that B1 in the second comparison diagram isn't sente now. Yamashiro Hiroshi played it in the 50th Oza, and won the game by 1.5 moku. :)

Comparison 1  
Comparison 2  

Bill: I think it would be good to show the actual game.

Kudo (W) - Yamashiro  

If B4 were at 6, White could indeed keep sente and play elsewhere. :)

What to call B4? B4 is a placement. It is played, as we see, as a sacrifice to take sente. Sacrifices to take or keep sente are not uncommon.

BTW, note the size of the threat that Black creates in the actual game. The smaller threat in the example diagram might not be big enough to take sente. :)

Tapir: Yes, indeed. But has this technique a name in any language? I feel like this sort of play is very common in professional play, "heating up" local temperature to sente-ize a certain move or to create ko threats. But placement can be literally anything. Placements to reduce a group to one eye, placements to kill, placements to gain some territory, here it is a placement to turn an otherwise gote move to sente. Isn't this something special, deserving an own term?

BQM 510 last edited by tapir on January 22, 2011 - 16:10
RecentChanges · StartingPoints · About
Edit page ·Search · Related · Page info · Latest diff
[Welcome to Sensei's Library!]
Search position
Page history
Latest page diff
Partner sites:
Go Teaching Ladder
Login / Prefs
Sensei's Library