Life In 19x19 http://www.lifein19x19.com/ |
|
Otake's masterpiece http://www.lifein19x19.com/viewtopic.php?f=15&t=17210 |
Page 2 of 2 |
Author: | Cassandra [ Sat Jan 18, 2020 12:01 am ] |
Post subject: | Re: Otake's masterpiece |
xela wrote: Perhaps Michael Redmond's "learn joseki and then 'forget' them all" left some traces in his brain which put him ahead of people who had never even tried to learn joseki? "Forget everything you know" must not be taken literally. Probably this advice came from martial arts training (or from classic legends about it). Its deeper meaning is to only act unconsciously (after more than sufficient training). As a matter of course, no samurai has "forgotten" his sword fighting techniques, so that he can no longer use them successfully. But he is not hindered in his actions by the need to "remember" these techniques conciously, which might take a bit too long to survive the fight. |
Author: | SoDesuNe [ Sat Jan 18, 2020 2:16 am ] |
Post subject: | Re: Otake's masterpiece |
Cassandra wrote: Its deeper meaning is to only act unconsciously (after more than sufficient training). As a matter of course, no samurai has "forgotten" his sword fighting techniques, so that he can no longer use them successfully. But he is not hindered in his actions by the need to "remember" these techniques conciously, which might take a bit too long to survive the fight. To add to this: Studies repeatedly showed that strength difference between chess players can be explained by their subconscious pattern recognition (chunking). Stronger players "see" the good/best moves faster (less candidate moves). Furthermore in tactical positions stronger players search only 3.6 moves deep. Weaker players search 5.1 moves deep in the same positions. Both on average. So, yeah: Subconsciousness ftw : ) Source: https://www.chess.com/article/view/the- ... y-of-chess |
Author: | kvasir [ Thu Aug 24, 2023 6:43 pm ] |
Post subject: | Re: Otake's masterpiece |
On the assumption that this topic/thread is about the game and not the meta discussion above I want to add something about this game. I'm very interested in Takagawa's predicament. Playing a game and having KataGo tell you that you never really had a chance is very relatable I only browsed through it before I started KataGo and it could be hard for me to separate my initial observations from what I observed later, however, my initial observations were something akin to the following: 1. There isn’t anything wrong with the first 20-30 moves per se. Which I’m calling the “early” opening. The most common style of the time seems to have been to make some rather mysterious moves in the opening, that I doubt anyone though were the only move except that-one-guy 2. Somehow, black quickly became very thick after the early opening. White was making some unusual moves that let black fix his shapes. 3. The game was probably rather close for the most part but there was an imbalance. The imbalance is that if black was behind he could try to shake the game but if white was behind it was not clear what to do about it except hope for a really clumsy mistake. This is the position after the “early” opening. This is the position when I though white has allowed black to get thick or have strong shapes everywhere. While I ran KataGo I read the various comments in the thread provided and realized that they didn’t closely match what I was seeing from KataGo. I also checked some databases and found that the first few moves in the opening have been played often enough that I’d expect it to have been studied by pros. At some point I realized that the game information gives komi of 5.0, which is 1.5 less than is considered fair. That means white should be at a disadvantage from the start. After analyzing with KataGo I had the following observations: 1. The first first diagram above (move 23) is evaluated as B+2.0 (komi 5.0 and 100k po) which is expected. The second diagram (move 51) is b+6.0 points and that is basically a difference of half a handicap stone. 2. White appears to have made a clumsy mistake in the opening on move 48. I wonder if it is something mentioned in period game reviews. Let me try to explain. 3 The game is bit like a model for when you are behind but don't know what to do about it. I don't know if it can be said to be clear that white was behind but it did turn out that way. White didn't play a poor middle game but probably had to try something provocative early and missed his chance. KataGo won't necessarily give a clear indication of how to create chances but you can use it to get ideas. 4 How to avoid making the opponent strong is a more practical lesson. Something to avoid even if it might be recommended by KataGo. It appears hard to get good help with this from KataGo who will make the opponent strong if it gives a small lead and sometimes when needed it will still find some very special moves that demonstrate that there can be some weakness. This was a great game to look at even though it sounded almost like a walkover when I started reading the thread Maybe there are some faults with my observations, don't mind that |
Author: | John Fairbairn [ Fri Aug 25, 2023 2:30 am ] |
Post subject: | Re: Otake's masterpiece |
Quote: White appears to have made a clumsy mistake in the opening on move 48. I wonder if it is something mentioned in period game reviews. One unattributed review said that White J17 was more usual for 14. Black made a lot of profit with 19. White 20 and 22 embarked on a big moyo strategy. Black 27 and 29 were a probe. If Black answers White 34 at H4, White jumps to 35. The cap at Black 34 was kiai. For the sagari of White 44, cutting at 48 is better timing; then if Black 51 there is bad aji. After 49, Black's has a lot of territory. If White 50 at K2, Black H1, White L3, then Black gets to push at 50. White 52 should be at G13 to attack on the upper side. For Black 57, G13 is correct: he cannot be satisfied to let White hit the vital point at 58. White 66 could also have been at 75, then Black E11, White F13. White 72 was a mistake. He had to push in at 75, then Black E11, White H10, to take away Black's eyes. The capture with Black 73 and 75 makes the big group safe. After this, Black had a lead on the balance of territories and so could wrap the game up with boundary plays. Quote: At some point I realized that the game information gives komi of 5.0, which is 1.5 less than is considered fair. Pedantic point: the komi was "5 + White wins jigo," so 5.5 in effect. |
Author: | kvasir [ Sat Aug 26, 2023 12:26 pm ] |
Post subject: | Re: Otake's masterpiece |
John Fairbairn wrote: One unattributed review said that White J17 was more usual for 14. Black made a lot of profit with 19. White 20 and 22 embarked on a big moyo strategy. Black 27 and 29 were a probe. The comments by the reviewer are reasonable enough. To me the first part is like the narration over some documentary movie. I guess the "more usual" move is the extension, the orientation of the coordinates is different. John Fairbairn wrote: If Black answers White 34 at H4, White jumps to 35. The cap at Black 34 was kiai. I think this means that white can play a good enough move on the bottom if black simply defends on the top and it is not the exact move that is important. It is a reasonable explanation, defending looks to me like something for a handicap game. Maybe 'kiai' was the normal level of active play for the time? John Fairbairn wrote: For the sagari of White 44, cutting at 48 is better timing; then if Black 51 there is bad aji. Now I get a feeling that I have seen the comments before, maybe repackaged. It is also interesting because KataGo doesn't like the good timing move. It does set things in motion, it is the right time for that, but KataGo sort of refutes the move. Even though let's white's stones settle black nevertheless appears to take good advantage of white's thinness. Black KataGo will also take control of the flow of the game for a good while in a position that was already advantageous. I guess there is no telling if a human opponent like Otake would play this way and practically, when not playing a computer, someone might just prefer the early cash for white. I wasn't going to rate every observation by the unattributed reviewer but I think it reasonable to call this a good timing, despite KataGo's -1.5 points reaction (the whole sequence starting on move 40 is -3.2 points). I'm not sure what the reviewer means by "bad aji" as black's next move would be very big and white therefore needs to play first in this area anyway. If black had played the other atari in the corner then the good timing move is approved. In this case white's outside is improved. John Fairbairn wrote: After 49, Black's has a lot of territory. If White 50 at K2, Black H1, White L3, then Black gets to push at 50. White 52 should be at G13 to attack on the upper side. I think that is as in the following diagram. John Fairbairn wrote: White 52 should be at G13 to attack on the upper side. Sounds like pro advice. I can think of waiting in a line at EGC for a pro to review my own game and then they come to a situation like this and say "Vital point!", "Attack first!", "Very good!", and "Oh no!" and "Maybe OK" when they see the game move. KataGo seems to think it is not forceful. At first KataGo indicated many acceptable moves in the game position but after checking them one by one something that is certainly similar to the game stands out. The difference is that white takes an important forcing move (blocks or slows down black's access to the center, or just prevents a good black move?) and then goes directly for . Leaving out the game move is the principle of playing C instead of A-B-C. Black's corner is softer when the A-B hasn't been played. The need for might not have been on white's radar and he is otherwise basically handling the situation similarly to what KataGo suggests. The difference is rather small in terms of evaluation. John Fairbairn wrote: White 66 could also have been at 75, then Black E11, White F13. Maybe this is why I bother replying with all the diagrams. The suggestion is to cut but white invests too much in this attack. Normally the good way to attack is to take some profit first. The opposite is to trade something for the attack, sometimes referred to as "bribing your opponent", that way you can struggle to get back what you paid up front. KataGo says black is just fine but I think the point is that white isn't fine. Maybe the reviewer suggests this for the reader to think about. John Fairbairn wrote: White 72 was a mistake. He had to push in at 75, then Black E11, White H10, to take away Black's eyes. The capture with Black 73 and 75 makes the big group safe. After this, Black had a lead on the balance of territories and so could wrap the game up with boundary plays. That is something I noticed too and it is very instructive: Don't just let the opponent's groups become safe. The variation or the timing is not perfect from KataGo's perspective but instructive. A move earlier, which is move 70, it suggests something it considers slightly better. I think it wants to try to hang in there and keep the game close on points, which is usually a good idea. It suggests the following way. White first strengthens the side with and finishes by enclosing one side of the center with . Black is practically alive and the evaluation is B+5.2. The difference to the reviewers variation (B+7.3) isn't much. I'm confused how KataGo will keep this position so close. The reviewer suggestion might be better for myself. I think I understand that KataGo wants to avoid giving something up for keeping black from being too easily alive and it suggests a way that has white take some profit first but black is practically alive anyway (as far as I can tell). Maybe those are two different approaches to playing when slightly behind. Reviewer logically (because that is what you do) wants to prevent black from having a completely safe group since safety for black leads to an endgame-like situation, KataGo wants to keep the game as close as possibly and force the black group to defend later and is very confident of its endgame. |
Page 2 of 2 | All times are UTC - 8 hours [ DST ] |
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group http://www.phpbb.com/ |