It is currently Fri Mar 01, 2024 12:38 am

All times are UTC - 8 hours [ DST ]




Post new topic Reply to topic  [ 25 posts ]  Go to page 1, 2  Next
Author Message
Offline
 Post subject: Thoughts on Honinbo Shusaku
Post #1 Posted: Sun Feb 10, 2019 12:16 am 
Lives in gote

Posts: 646
Liked others: 62
Was liked: 116
I'm not sure this is the optimal forum for this. If it isn't, I hope an administrator will move it to where it belongs.
Honinbo Shusaku
I'm reading and playing through 'Invincible: The Games of Honinbo Shusaku' right now, and it's probably the most enjoyable Go book I've ever laid hands on. Of course, that's a personal opinion. For me it's so because it's a mixture of Go history and amazing game records.

Going through Invincible (I'm about 35 games in, now), I've had some thoughts on the man (, the legend) that I've wanted to share. Feel free to disagree and bring additional information into the mix.

Need to know, in case you didn't: in Shusaku's time there was no komi. Difference in ranking determined the handicap. Weaker players took black. There's more to it, if you're interested, look it up on Sensei. It's an elegant handicap system.

After 35 games, I've a few thoughts:

- Part of his legacy, his fame, of being 'invincible' probably stems from his slow promotion. While it was definitely fast for the time, you still feel his promotion to 7-dan should've been quicker and he spend a while on the wrong handicap, taking black against inferior opponents.
Take his first two Castle Games as an example. He really coasted through these with black. I think he could've won both with white, but he was still a 6-dan, so he was to take black.
Of course Shusaku was amazingly strong, but at the time he was probably (one of) the strongest players alive, his rank didn't reflect that.

- Shusaku feels very modern at times. He was already amazing at 12-13 years old and arguably at the top of his game in his early 20s. This resembles modern players and was not very common in the old days, where Go players of stature were often way older.

- As playing goes, he also feels modern, some times. Of course he plays Go from that time. But at times, he has a modern feel to it, sometimes even an AlphaGo feel. But perhaps I'm not strong enough to comment on that.
I am strong enough to recognize a 3-4 high enclosure with a low extension just below the star point and that was hardly ever seen in those days, has a very modern feel to it of fast development.

- Sekiyama Sendayu should've been a professional player and played a very aggressive and interesting style. He was dubbed as the strongest amateur, but seeing him play a nijubango against Shusaku, he feels stronger than many other professionals.

- Shusaku mastered the concept of non-attachment in Go. His positional judgement is so sharp, especially seen in long ko's, that he sacrifices stones so easily when he's calculated he doesn't gain by saving them. For an amateur player such as myself, it is often surprising to see him so eager to give up stones he feels will be a burden and cost him points. Controlled sacrifice is seen in many of his games.
He often plays tenuki when you think: what? are you kidding me?
Funny thing is: often his opponent will not take the stones (immediately) probably because it's gote and not worth enough, and sometimes the stones do end up living. Lots to learn from that kind of playing.

I'll do an update if I have any future ideas on the master :)

Top
 Profile  
 
Offline
 Post subject: Re: Thoughts on Honinbo Shusaku
Post #2 Posted: Sun Feb 10, 2019 9:04 pm 
Dies with sente

Posts: 102
Liked others: 9
Was liked: 8
Rank: 4K
Also from John Fairbairn's excellent book The Life of Honinbo Shuei:
Quote:
In the 1862 event, Shuwa had slated Sanei 3-dan to play Shusaku 7-dan, who up till then had won all nineteen of his Castle Games. But Shusaku opted out. He is supposed to have said, “Up to now I have not lost a single game in the Castle Games. I would like in future to remain undefeated. I will have to give Sanei two stones, but a two-stone game will be very close and if Black makes no bad moves, there is no way White can win. If Sanei were to be made 5-dan then I would happily play him.” If true, this story brings suspicions that Shusaku wasn’t so invincible after all and had maybe carefully arranged his opponents even in earlier years, and that Shuwa, who would naturally want to boost Honinbo prestige, was complicit in it. There is at least confirmation of the 1862 episode in that Sanei himself smugly referred to it later: “I know that Shusaku, who was certainly of Meijin class, absolutely refused to play me on two stones. His withdrawal was consistent with that.”

It seems that there is quite some substance behind the Shusaku 'hype', but 'hype' it was nonetheless. Or am I being too harsh? Before the title of 'latter Go saint' was transferred to Shusaku, it was bestowed upon the scheming Jowa who, to this day, is regarded as the stronger of the two by many Japanese professionals, and that Shuei surpassed both of them in strength.

Top
 Profile  
 
Offline
 Post subject: Re: Thoughts on Honinbo Shusaku
Post #3 Posted: Sun Feb 10, 2019 11:53 pm 
Lives in gote

Posts: 646
Liked others: 62
Was liked: 116
Since none of us were there, it's hard to say. Probably the truth is somewhere in between.

A cynic might take Shusaku's wish to never take white against his sensei as being afraid to lose. Another sees it as a sign of great respect.
From what I've read so far, Shusaku seems to have been a very respectable man, muuuuch more so than scheming Jowa.
His title of Kisei is probably deserved, but perhaps more players have deserved it.

I'm not strong enough myself to judge his comperative strength. Was he a strong player. Absolutely. Was he the strongest at his time? Highly likely. Was he the strongest player ever? Unlikely, but he won't be too far off.
He certainly was a prodigy, a player of immense strength and a real Go Saint.

However, he himself has had much experience with "wrong" handicap games. Is it so wrong of him to want to keep his perfect record and not see it ruined by a wrongful handicap?
Sure, some of his castle games were easy because he was black and was definitely the better player, but he never took 2 stones in a handicap game, either. So it was never unfair. Neither did he achieve his 19 wins easily. He had to beat some damn fine opponents to get it.

All that to say: we don't know. I'd like to believe he was a stand-up guy. Sure he cared about winning and losing. But I think he was respectable nonetheless.
His name is most likely a bit overhyped, but sometimes that happens. While others wallow in obscurity, equally unjustified. Is it right that someone like Shuei is a lot less famous? Probably not.

But that doesn't mean Shusaku doesn't deserve the fame neither. He was one fine player, someone with an impressive record and with amazing skills, already at a young age. His (once again honorable and respectable) death probably have a hand in his fame, too.

Top
 Profile  
 
Offline
 Post subject: Re: Thoughts on Honinbo Shusaku
Post #4 Posted: Mon Feb 11, 2019 4:49 am 
Oza
User avatar

Posts: 2401
Location: Tokyo, Japan
Liked others: 2336
Was liked: 1332
Rank: Jp 6 dan
KGS: ez4u
There have been many discussions here over the years about measuring/comparing historical figures. I was playing with GoGoD (my slightly out of date version) and produced the following table. This is based on the listed players taking White in no-komi games. The players are ranked by their winning percentage
Code:
No-Komi Games         Total  Games    Winning
as White              Games    Won Percentage
Honinbo Shuho           254    144      56.7%
Honinbo Shuei           115     65      56.5%
Honinbo Shusai          266    150      56.4%
Go Seigen               288    155      53.8%
Sakata Eio               85     45      52.9%
Suzuki Tamejiro          60     31      51.7%
Karigane Junichi         38     19      50.0%
Kitani Minoru           330    153      46.4%
Kubomatsu Katsukiyo     150     69      46.0%
Honinbo Shusaku          72     33      45.8%
Honinbo Shuwa           124     56      45.2%
Hashimoto Utaro         364    153      42.0%
Segoe Kensaku           164     68      41.5%
Takagawa Shukaku        206     85      41.3%
Iwamoto Kaoru           179     69      38.5%
Honinbo Jowa            122     47      38.5%

_________________
Dave Sigaty
"Short-lived are both the praiser and the praised, and rememberer and the remembered..."
- Marcus Aurelius; Meditations, VIII 21


This post by ez4u was liked by 5 people: Bill Spight, Elom, gazzawhite, TheCannyOnion, Waylon
Top
 Profile  
 
Offline
 Post subject: Re: Thoughts on Honinbo Shusaku
Post #5 Posted: Mon Feb 11, 2019 9:43 am 
Lives in gote

Posts: 646
Liked others: 62
Was liked: 116
Interesting figures, ez4u, thanks!
Of course the numbers only tell a small part of the story, but they're nice to have and it's a piece of the puzzle.
Someone like Doteki never had a lot of chance to make the statistics and loses out completely here :)

Shusaku doesn't have the greatest record with white. That could have multiple reasons. But part of it is probably that - no - Shusaku was not the best player to ever live. He is most likely the most celebrated, but that has much to do with his integrity as a person (especially compared to Jowa at that time) and his 19-0-0 in the Castle Games.

Nor is it possible to have a 'best player' of all time, I think. (considering humans). Each of the greatest had incredible strength but different styles, different opponents, different struggles and also sometimes different circumstances of playing.
Huang Longshi never started with an empty board. Who knows how much 'worse' he could've turned out if his fuseki was a disaster, for example? Though I'm not saying he'd have been bad at it, just a manner of speaking :)

Probably pros today would beat guys like Shusaku (the top pros) in a modern game, especially since they've been trained with tigher time limits. I don't know if they'd necessarily do better in a game with limited or no time limits. Maybe they would, maybe they wouldn't.

However, I do think the games of the modern pros are not necessarily of a higher quality than those of, say, Shusaku. On the contrary. Because of the time limits (and sometimes overfull schedules), I think game records of Shusaku, Shuei... are more likely of higher quality (on general), because it's way harder to actually blunder with so lenient time limits. Today, even Ke Jie can make a big blunder like that :lol:

Top
 Profile  
 
Offline
 Post subject: Re: Thoughts on Honinbo Shusaku
Post #6 Posted: Tue Apr 12, 2022 2:36 pm 
Lives in sente
User avatar

Posts: 714
Liked others: 109
Was liked: 138
Rank: Shokyu
Universal go server handle: CDavis7M
I've been going through Shuwa's Castle Games and I changed my thoughts on Shusaku, and maybe this is the best place for it. I used to think that Shusaku's reputation was inflated because he was mostly playing Black in the Castle Games (and I still do a bit) -- same opinion you see elsewhere. You know, a win with Black is not even a win if you are not yet to tagaisen -- it's just a step on your way to tagaisen. But here's what I've realized, Shusaku also won as White in the Castle Games, and he won against top players who also did well against Shuwa.

Comparing Shuwa and Shusaku's Castle Games against to the same opponent shows that Shusaku really did hold his own: Shusaku beat Sakaguchi Sentoku on White while Shuwa lost 2 of 6 against Sakaguchi as White. Shusaku beat Hayashi Yubi on White twice while Shuwa lost to Yubi on White. And Shusaku won against Ito Showa on White while Shuwa won once and lost once against Ito. Sakaguchi 7d and Ito were two of the "Tenpo Top 4." Hayashi Yubi was 6d when Shusaku (7d) beat him. So, these wins were not against push-overs. Especially not while they were taking Black. Another point to consider is that more than half of Shusaku's Castle Games were played as 7-dan so it's not like he was relying on receiving a handicap.

There is also the complaint about Shusaku not wanting to play against Yasui Sanei mentioned above. The quote given states that Shusaku didn't want to give Sanei 2 stones. However, Hayashi Hakuei (7d) and Ito Showa (7d) did play Sanei (2d/3d) and they both had to give Sanei 3 stones. Of course they both lost. I don't see why Shusaku would not also have to give 3 stones, not 2. And the whole situation with Yasui Sanei was weird anyway. I believe only heads, heirs, and 7-dans could compete. But Sanei was an heir young because his father died young. So I don't know if his promotions were in line or not.

Anyway, I now feel like the common complaints (always took black, under-ranked, never played even against Shuwa, didn't want to play Yasui Sanei) are overblown. I don't know if Shusaku is so great as to be a "Go Saint" over the other Shu-players, but he did well in the Castle Games he at least deserves that bit of fame.

Notes from the GoGoD encyclopedia. ** indicates a non-scheduled/requested game.
Honinbo Shuwa (W) v. Sakaguchi Sentoku II (B). W+1. 1840-12-10a
Honinbo Shuwa (W) v. Sakaguchi Sentoku II (B). W+3. 1847-12-24a
Sakaguchi Sentoku II (W) v. Honinbo Shuwa (B). B+1. 1844-01-06b
Honinbo Shuwa (W) v. Sakaguchi Sentoku II (B). B+R. 1845-12-15b
Sakaguchi Sentoku II (W) v. Honinbo Shuwa (B). B+2. 1847-01-03d
Honinbo Shuwa (W) v. Sakaguchi Sentoku II (B). W+3. 1847-12-24a
Honinbo Shuwa (W) v. Sakaguchi Sentoku II (B). W+2. 1849-12-31e
Honinbo Shuwa (W) v. Hayashi Hakuei (B). W+R. 1849-12-31f**
Honinbo Shuwa (W) v. Inoue Matsumoto Inseki. W+2. 1850-12-20e
Honinbo Shuwa (W) v. Ito Showa (B). B+2. 1851-12-09f **
Honinbo Shusaku (W) v. Inoue Matsumoto Inseki (B). W+2. 1852-12-27a
Hayashi Genbi (W) v. Honinbo Shuwa (B). B+7. 1852-12-27h
Honinbo Shuwa (W) v. Ito Showa (B). W+5. 1853-12-17f **
Honinbo Shusaku (W) v. Inoue Matsumoto Inseki (B). W+R. 1855-01-05a
Honinbo Shuwa (W) v. Sakaguchi Sentoku II (B). B+R. 1855-01-05c
Honinbo Shusaku (W) v. Ito Showa (B). W+R. 1856-12-14a
Honinbo Shuwa (W) v. Hayashi Yubi (B). B+5. 1858-01-01c
Honinbo Shusaku (W) v. Sakaguchi Sentoku II (B). W+3. 1858-12-21a
Honinbo Shuwa (W) v. Inoue Matsumoto Inseki (B). W+6. 1858-12-21b
Honinbo Shusaku (W) v. Hayashi Yubi (B). W+4. 1861-01-18a
Honinbo Shusaku (W) v. Hayashi Yubi (B). W+R. 1861-12-18b **
Honinbo Shuwa (W) v. Inoue Matsumoto Inseki (B). B+1. 1861-12-18c
Hayashi Hakuei (W) v. Yasui Sanei. B+R. 3-stones. 1861-01-18c
Ito Showa (W) v. Yasui Sanei. B+R. 3-stones. 1861-12-18d


By the way, there is a 7-stone (requested) castle game.

Click Here To Show Diagram Code
[go]$$W Yasui Sentetsu (W) v. Nakabo Kinzo (B). B+5
$$ ---------------------------------------
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . 6 . . . . . 2 . . . . . . |
$$ | . . . X . . . . . , . . . . . X . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . 5 . . . . . . . . . . . . . 1 . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . X . . . . . X . . . . . X . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . 9 . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . 7 . 3 . . |
$$ | . . 0 . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . 4 . . |
$$ | . . . X . . . . . , . . . . . X . . . |
$$ | . . . . . . . . . . . . 8 . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ ---------------------------------------[/go]


Last edited by CDavis7M on Tue Apr 12, 2022 3:04 pm, edited 2 times in total.
Top
 Profile  
 
Offline
 Post subject: Re: Thoughts on Honinbo Shusaku
Post #7 Posted: Tue Apr 12, 2022 3:01 pm 
Gosei

Posts: 1619
Liked others: 542
Was liked: 449
Rank: senior player
GD Posts: 1000
A piece of modern support for Shusaku comes from the strong AI players. One of Shusaku's most quoted remarks besides "I had Black" was his comment that the kosumi response to a 3-4 approach to a komoku corner stone (known as the Shusaku kosumi) would never be a bad move. The Shusaku kosumi fell out of favor in the 20th century but AI has resurrected it.


This post by gowan was liked by: Elom0
Top
 Profile  
 
Offline
 Post subject: Re: Thoughts on Honinbo Shusaku
Post #8 Posted: Thu Nov 23, 2023 2:04 am 
Oza
User avatar

Posts: 2399
Location: Ghent, Belgium
Liked others: 358
Was liked: 1014
Rank: KGS 2d OGS 1d Fox 4d
KGS: Artevelde
OGS: Knotwilg
Online playing schedule: UTC 18:00 - 22:00
I'm reviewing the castle games with AI. It's once more a sobering experience, not so much about Shusaku's playing strength, rather the professional commentary. Not saying it's wrong, but it bears almost no relationship to the AI evaluation. I have to compile more data but in the first three games there seems to be a pattern of Shusaku increasing the gap in the endgame, while he tended to fall back in the opening and middle game. This runs counter the commentaries, which have Black (Shusaku) obtaining victory early.

Of course, it might be that AI underestimates Black's intrinsic advantage and only realizes it as the endgame unfolds. But the estimations are not the only difference between AI and the pros. Many of the bad/losing moves by White (the opponent) are KataGo's "blue move". What KataGo finds more serious mistakes by either player, is glossed over by the commentary.

See
https://senseis.xmp.net/?InvincibleTheG ... WithKataGo
https://senseis.xmp.net/?InvincibleTheG ... WithKataGo

Top
 Profile  
 
Offline
 Post subject: Re: Thoughts on Honinbo Shusaku
Post #9 Posted: Thu Nov 23, 2023 3:33 am 
Oza

Posts: 3640
Liked others: 20
Was liked: 4618
Not meaning to discourage you, but rather to re-direct you, I think you may be starting off-kilter here.

First, you clearly haven't read any of my books such as Kamakura which are based on a wealth of pro commentaries (up to nearly 30 in one case) and it easy to find multiple cases where pros differ markedly from each other. You even see the extreme: "terrible move" versus "brilliant move" and other commentators ignore the move altogether. So AI saying something different from ONE book on Shusaku is nothing unusual.

Second, there have been books and magazine articles in Japan comparing the old greats to AI. There have been similar things in China, but I haven't seen the actual publications - just the results. Their conclusions have been rather different from yours. They have been pleasantly surprised at how good the old players were.

Third - and this, I think, this explains the point just above, the researchers there are looking at it with a somewhat different mindset, and one I think we need to follow. Indeed, the mindset in question is one that would be of great value to amateurs irrespective of whether AI is in the equation.

I'll have to add a bit of personal interpretation here, and so I may be the one off-kilter, but I am working from what I've read.

The key word, I think, is "consistency", which can often be transmuted into "style" of play. Unlike AI bots, a human player has to set out on the vast ocean of the go board with absolutely no sense of what terrors might face him. Some, Viking-like, are happy to sail adventurously across the seas. Some prefer to stay close to land. Some, laden with liferafts, prefer the slow boat to China. Some, like the Titanic, ignore the lifeboat joseki drills and tussle with icebergs in a mad rush to break some spurious but glorious record in go history.

Each player chooses his own version of consistency, or style. And what the oriental researchers have found, using AI, is that certain players in history have been better at achieving consistency than others and, on the whole, the level of consistency marries up with present stature. The one big surprise has been a downgrading of the stature of Dosaku, which never felt like a surprise to me as I (as a mere amateur) could never detect a Dosaku style. This pleased the Chinese because the stature of his contemporary Huang Longshi (the true discoverer of tewari, incidentally) was confirmed. Shusaku was regarded as highly consistent, taking on extra fuel and lifeboats at the start of the game, which slowed him down at the beginning of his journey, but ensured he got to his destination.

However, I never got any sense of how this consistency was measured. Katago was not the bot used, and technical details were not given except that there was a strong hint (at least in the Chinese case) that the AI was also being used to detect a player's style. This was connected with research to allow historians to detect which games were forgeries (a big problem in the old Chinese world) and/or to date games (hence the current belief that the games once attributed to the 2nd century are now clearly of the 9th century or later, etc).

As to the point of human commentaries on human players being capable of being different, just as there are strong players and not-so-strong players there are good and not-so-good commentators. It has long been recognised that the best commentator on old no-komi games was Kato Shin. He was famous for his anti-komi remark "komi go is not go". He seemed to understand the styles of no-komi players better than anyone (Iwamoto was also highly rated in that regard, but I think the Shusaku commentaries you are reading may mostly be by a modern "komi player", Miyamoto), and his commentaries are much less about this move or that move being objectively good or bad and more about each move being good or bad at enabling the player to achieve his destination in his preferred mode of travel.

I think this consistency is what amateurs need to study, not Katago's decimal points. It is best studied through handicap games, which many amateurs foolishly avoid as if handicap stones carry the plague. I have mentioned before that, in contrast, my words-oriented brain avoids numbers like the plague, especially decimal ones - I was of the generation that was most comfortable with the 12-times table - but I do mentally somehow keep track of word frequencies. The one word I remember from the early days when all we had in English was Go Review from Japan, and it was packed full of large-handicap games rather than title matches, was "consistent" (or its cognates, of course). I later noticed this in Iwamoto's Japanese commentaries, too, as it happens.

I'll repeat that this is not meant to discourage you, but rather to re-direct you. Anyone who posts to start a discussion on L19 deserves a medal. Those who hide in the trenches, however, ....


This post by John Fairbairn was liked by: Thumb
Top
 Profile  
 
Offline
 Post subject: Re: Thoughts on Honinbo Shusaku
Post #10 Posted: Thu Nov 23, 2023 11:21 am 
Oza
User avatar

Posts: 2399
Location: Ghent, Belgium
Liked others: 358
Was liked: 1014
Rank: KGS 2d OGS 1d Fox 4d
KGS: Artevelde
OGS: Knotwilg
Online playing schedule: UTC 18:00 - 22:00
John Fairbairn wrote:
Not meaning to discourage you, ....


Not so easily discouraged, John, and over time I've grown to appreciate your comments even when they could sound discouraging. :)

(...)

Quote:
The key word, I think, is "consistency", which can often be transmuted into "style" of play. Unlike AI bots, a human player has to set out on the vast ocean of the go board with absolutely no sense of what terrors might face him. Some, Viking-like, are happy to sail adventurously across the seas. Some prefer to stay close to land. Some, laden with liferafts, prefer the slow boat to China. Some, like the Titanic, ignore the lifeboat joseki drills and tussle with icebergs in a mad rush to break some spurious but glorious record in go history.

Each player chooses his own version of consistency, or style. And what the oriental researchers have found, using AI, is that certain players in history have been better at achieving consistency than others and, on the whole, the level of consistency marries up with present stature. The one big surprise has been a downgrading of the stature of Dosaku, which never felt like a surprise to me as I (as a mere amateur) could never detect a Dosaku style. This pleased the Chinese because the stature of his contemporary Huang Longshi (the true discoverer of tewari, incidentally) was confirmed. Shusaku was regarded as highly consistent, taking on extra fuel and lifeboats at the start of the game, which slowed him down at the beginning of his journey, but ensured he got to his destination.


That's interesting. Maybe I should clarify that I'm not taking my AI analysis as a criticism, rather a "critique" in Kant's sense, of Shusaku. In the book, Shusaku is not blowing his own trumpet, it's the professional commentators. Some of Shusaku's choices are "consistently" met with negative evaluations by KataGo, like "not playing in the open corner" or "pincering without backup of influence on the opposite side", things I already knew, but also technical moves like the "bump to reinforce in sente", which KataGo seems to think reinforces the opponent more and needlessly so.

(...)

Quote:
As to the point of human commentaries on human players being capable of being different, just as there are strong players and not-so-strong players there are good and not-so-good commentators. It has long been recognised that the best commentator on old no-komi games was Kato Shin. He was famous for his anti-komi remark "komi go is not go". He seemed to understand the styles of no-komi players better than anyone (Iwamoto was also highly rated in that regard, but I think the Shusaku commentaries you are reading may mostly be by a modern "komi player", Miyamoto), and his commentaries are much less about this move or that move being objectively good or bad and more about each move being good or bad at enabling the player to achieve his destination in his preferred mode of travel.


The commentators in those first 3 castle games are Sanno Hirotaka and Ishia Yoshio. It's a small sample still. I found myself wondering if they weren't blinded by hindsight of Shusaku's great achievement and/or biased towards the idea that great victories must be due to great openings and/or simply enjoyed commenting on the opening more.

Quote:
I think this consistency is what amateurs need to study, not Katago's decimal points.


I've long been cured from the decimal desease. Only when a devaluation gets around the -2 mark I will consider the move as a potential mistake by AI standards and possibly interesting for amateurs. When reviewing my own games I put the threshold at -4. People have argued before that even this is a wrong approach and we should look for patterns of mistakes, even if small. I find that even harder to do, especially if the precision of AI with moderate playouts is >1.

Quote:
It is best studied through handicap games, which many amateurs foolishly avoid as if handicap stones carry the plague.


Another L19 brother has once advised me to play KataGo with handicap, jubango style. I did and might repeat the exercise. I'd agree that it's comparatively good for learning. Even games are comparatively esthetic and more entertaining, also to review. For me at least.

Quote:
Anyone who posts to start a discussion on L19 deserves a medal. Those who hide in the trenches, however, ....


Mohicans we are :)


This post by Knotwilg was liked by: Thumb
Top
 Profile  
 
Offline
 Post subject: Re: Thoughts on Honinbo Shusaku
Post #11 Posted: Mon Nov 27, 2023 3:49 pm 
Oza
User avatar

Posts: 2399
Location: Ghent, Belgium
Liked others: 358
Was liked: 1014
Rank: KGS 2d OGS 1d Fox 4d
KGS: Artevelde
OGS: Knotwilg
Online playing schedule: UTC 18:00 - 22:00
Just finished the review of castle game 4. It's interesting, though not necessarily relevant, that this "masterpiece" by both Showa (in the opening) and Shusaku (in the fighting) gets a rather coherent critique by the pros in "Invincible" and (my) KataGo. Although the losing move 140 as per KataGo's evaluation is not marked as critically by the pros, they do highlight it as a confused response by Showa to Shusaku's fighting spirit. Moreover, Ishida's speculation on what would have followed, should Showa have responded "correctly" (connecting the ko), is exactly how KataGo would proceed (a huge furikawari).

The exercise has also managed to convince me a little more of what's being said about Shusaku, i.e. that he would maintain a steady lead but when provoked could resolve to very sharp fighting. Indeed, after Showa's mistake in the game brought the advantage back to 10 points or so, Shusaku carries it into a 3 point victory.


This post by Knotwilg was liked by 3 people: gowan, Harleqin, swannod
Top
 Profile  
 
Offline
 Post subject: Re: Thoughts on Honinbo Shusaku
Post #12 Posted: Fri Dec 01, 2023 6:30 am 
Oza
User avatar

Posts: 2399
Location: Ghent, Belgium
Liked others: 358
Was liked: 1014
Rank: KGS 2d OGS 1d Fox 4d
KGS: Artevelde
OGS: Knotwilg
Online playing schedule: UTC 18:00 - 22:00
Castle game 5 review: https://senseis.xmp.net/?InvincibleTheG ... WithKataGo

Like many of the previous games, the initial advantage of the no komi game gradually vanishes, up to a point where it's an even game. Then the flow reverses and in the end Shusaku gets a comfortable win by 7 points.

As user KMR comments: "this is a nice proof that humans usually overestimate the value of a moyo, or maybe that moyos are comparatively hard to deal with for humans (...) In Invincible, some games were commented by GoSeigen, i wonder if there is a significant difference in the level of agreement of the commentary and KataGo, when comparing Go Seigen with other commentators." That's a nice topic for research.

Indeed, the pro commentators didn't pick up on what KataGo considered the losing move of 108 and instead see the course of the game as a confirmation of Shusaku's correct judgment of sacrificing the corner for central influence.

Top
 Profile  
 
Offline
 Post subject: Re: Thoughts on Honinbo Shusaku
Post #13 Posted: Fri Dec 01, 2023 8:37 pm 
Lives with ko

Posts: 198
Liked others: 4
Was liked: 16
Very nice topic. Thanks to you I am watching the games one more tume and I try to have a different eye on it.


This post by lichigo was liked by: Knotwilg
Top
 Profile  
 
Offline
 Post subject: Re: Thoughts on Honinbo Shusaku
Post #14 Posted: Mon Dec 04, 2023 8:49 am 
Oza
User avatar

Posts: 2399
Location: Ghent, Belgium
Liked others: 358
Was liked: 1014
Rank: KGS 2d OGS 1d Fox 4d
KGS: Artevelde
OGS: Knotwilg
Online playing schedule: UTC 18:00 - 22:00
https://senseis.xmp.net/?InvincibleTheG ... WithKataGo

The 6th castle game (or the analysis with KataGo) follows a similar pattern of previous ones, where the no komi advantage wears down towards the end of the middle game but that's where Shusaku shifts gears or the opponent gets frustrated by a lack of significant progress. Commentator Sanno describes the game as over or hopeless precisely when KataGo thinks it is very close.

Top
 Profile  
 
Offline
 Post subject: Re: Thoughts on Honinbo Shusaku
Post #15 Posted: Mon Dec 04, 2023 9:47 am 
Oza

Posts: 3640
Liked others: 20
Was liked: 4618
Quote:
Commentator Sanno describes the game as over or hopeless precisely when KataGo thinks it is very close.


Here again I think we need to defend the human commentators. A very common type of comment is along the lines of "The game is very close but now Black cannot lose." This is because, in their judgement, there are no further areas where mistakes or upsets are likely to occur, and they feel safe in assuming that normal, good play by fellow professionals will guarantee the result, even if it's a win by a mere 1 or 2 points. It's a bit like a soccer game where one side goes 1-0 up and then allow the other side to mount wave after wave of attacks. This may be because of an assessment of their own strengths and weaknesses - if they themselves attack, they may be more likely to fall prey to a quick break. They may see grim defence as one of their strengths. And, of course, individual players like Shusaku may have made grim defence is likewise an attribute they shine with.

As to the repeated assessments that Shusaku gradually loses his initial lead by the middle game, without denying that thats' what the numbers may show, I'd be more inclined to trust the common view that Shusaku opted to invest in thick moves (thick in the endgame sense) with no apparent return in the middle of the game, because he was confident in his ability to turn that kind of investment into cash against the players he usually encountered.


This post by John Fairbairn was liked by 2 people: Akura, Knotwilg
Top
 Profile  
 
Offline
 Post subject: Re: Thoughts on Honinbo Shusaku
Post #16 Posted: Mon Dec 04, 2023 10:10 am 
Oza
User avatar

Posts: 2399
Location: Ghent, Belgium
Liked others: 358
Was liked: 1014
Rank: KGS 2d OGS 1d Fox 4d
KGS: Artevelde
OGS: Knotwilg
Online playing schedule: UTC 18:00 - 22:00
John Fairbairn wrote:
Quote:
Commentator Sanno describes the game as over or hopeless precisely when KataGo thinks it is very close.


Here again I think we need to defend the human commentators. A very common type of comment is along the lines of "The game is very close but now Black cannot lose." This is because, in their judgement, there are no further areas where mistakes or upsets are likely to occur, and they feel safe in assuming that normal, good play by fellow professionals will guarantee the result, even if it's a win by a mere 1 or 2 points.


If you look at the one but last diagram in the link, KataGo suggests attaching to the lower left in order to attack the top. I can't judge professional capability but that looks complex enough for me for Black to still make a mistake, if the "numbers" say he has a one point lead.

In contrast, in the last diagram there are no more such opportunities and I can agree any pro carries victory home with that position for Black.

So, if I can rely on KataGo, I would say Sanno was a little early announcing victory. Of course KG can be wrong and the humans correctly evaluated the position by 94-96 as a comfortable lead for Black.

Top
 Profile  
 
Offline
 Post subject: Re: Thoughts on Honinbo Shusaku
Post #17 Posted: Mon Dec 04, 2023 10:15 am 
Oza
User avatar

Posts: 2399
Location: Ghent, Belgium
Liked others: 358
Was liked: 1014
Rank: KGS 2d OGS 1d Fox 4d
KGS: Artevelde
OGS: Knotwilg
Online playing schedule: UTC 18:00 - 22:00
John Fairbairn wrote:
Quote:
Commentator Sanno describes the game as over or hopeless precisely when KataGo thinks it is very close.

As to the repeated assessments that Shusaku gradually loses his initial lead by the middle game, without denying that thats' what the numbers may show, I'd be more inclined to trust the common view that Shusaku opted to invest in thick moves (thick in the endgame sense) with no apparent return in the middle of the game, because he was confident in his ability to turn that kind of investment into cash against the players he usually encountered.


I would be convinced of that explanation if next we saw a gradual point increase with a steady flow of the game. However, in many cases the opponent makes some kind of mistake and it goes downhill from there. I'm more inclined to make psychological inference. Shusaku played thickly and didn't allow for any major upset, however allowed for the game to become potentially close, exhausting his opponents and leading them into risk taking by frustration.

The sample has grown to 6 games. I'll continue to at least 10 and then make a first stab at a bigger scale interpretation.

Top
 Profile  
 
Offline
 Post subject: Re: Thoughts on Honinbo Shusaku
Post #18 Posted: Mon Dec 04, 2023 2:08 pm 
Oza
User avatar

Posts: 2399
Location: Ghent, Belgium
Liked others: 358
Was liked: 1014
Rank: KGS 2d OGS 1d Fox 4d
KGS: Artevelde
OGS: Knotwilg
Online playing schedule: UTC 18:00 - 22:00
https://senseis.xmp.net/?InvincibleTheG ... WithKataGo

In this 6th castle game, Shusaku played White against Inoue Matsumoto Inseki and grinded his way to a 2 point victory. Miyamoto Sensei concludes that Matsumoto Inseki has made no mistakes but Shusaku played brilliantly, moving into the lead imperceptibly, the key moment being the skillful settling of his group in the top right.

KataGo indeed doesn't reveal any big point losing mistakes by Black but there are some points in the game where he could have counterattacked to maintain or regain the lead, most notably the cut at B53, which the book dismissed, and the clever play at B115, sacrificing the cutting stones in a flexible manner. Both these moves were rather inventive and required a superior positional judgment. One can't blame Matsumoto for having missed those few occasions of superhuman strength.

The steady graph, with oscillations of at most 2-3 points, confirms Miyamoto's assessment of a "masterpiece".


This post by Knotwilg was liked by: bayu
Top
 Profile  
 
Offline
 Post subject: Re: Thoughts on Honinbo Shusaku
Post #19 Posted: Mon Dec 04, 2023 4:41 pm 
Oza

Posts: 3640
Liked others: 20
Was liked: 4618
Quote:
I would be convinced of that explanation if next we saw a gradual point increase with a steady flow of the game. However, in many cases the opponent makes some kind of mistake and it goes downhill from there.


Of course, but I think the mistakes are often induced by the style of play, especially a thick style. I say this with the benefit of having just read a commentary on a Go Seigen game where he made a thick shape à la Shusaku (who he studied as a child). His opponent knew he had to choose between a move that counteracted the thickness and a simple tsume (checking move) on the side. But his problem was that he couldn't find a way to evaluate the thickness related move (even though he had lots of time - a two-day game). Basically it was a case of trying to count thickness-associated territory and the collateral effect of the various effect on diverse groups. The sort of thing that is bread and butter for AI.

In the end he chose the thickness-related move but realised within a few more moves that he had made the wrong choice and he resigned after not much more than 100 moves. It seems to me that this is the same trajectory you are describing, with opponents becoming confused or frustrated when facing massive thickness (again, it's the boundary-play type I and the original commentary I'm reading that is being talked about).

I think this trajectory may be the key point that emerges - it's valid strategy if you can exploit just how difficult it is for even the very best pros to evaluate thickness late in the game (i.e. where it is about making territory and eliminating aji rather than attacking on a large scale). Or at least if, like Shusaku, you are better at it than they are.

Another point that I think is worth emphasising is that mistakes are inevitable in ALL human play. It therefore seems legitimate and sensible to factor that into how you play, even if it means a lack of "purity". In his letters and also anecdotes about him, Shusaku appears to be the kind of person who would think deeply about psychological effects. One example is when he was accosted by some brigands but ended up talking to them in such an affable way that they were almost then asking him to rob them!


This post by John Fairbairn was liked by 2 people: bayu, Knotwilg
Top
 Profile  
 
Offline
 Post subject: Re: Thoughts on Honinbo Shusaku
Post #20 Posted: Tue Dec 05, 2023 3:28 pm 
Oza
User avatar

Posts: 2399
Location: Ghent, Belgium
Liked others: 358
Was liked: 1014
Rank: KGS 2d OGS 1d Fox 4d
KGS: Artevelde
OGS: Knotwilg
Online playing schedule: UTC 18:00 - 22:00
Castle game 8 is a nightmare. The AI review was very inconsistent and very hard to turn into some kind of reason. Inconsistency is when a sequence of "blue moves" changes its evaluation with several points, hence requires major playouts per move to come to an overall equilibrium. Also, the suggested moves go from one area to another, indicating that some kind of global thickness is key in the game.

Again, the evaluation drops from 6.5 to 2.5 then goes up again towards 6. Commentator Shuko calls the game a success for Shusaku's thickness strategy. At least that is consistent!

So I'm going to change gears here and study the commentary first.

Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 25 posts ]  Go to page 1, 2  Next

All times are UTC - 8 hours [ DST ]


Who is online

Users browsing this forum: No registered users and 1 guest


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group