19x19 game with most moves

[Pippen]

What u got? Is there an offical record holder?

[Uberdude]

I saw some guy play a game on KGS with around 30 thousand moves. But maybe you mean serious game in a professional tournament? And official by the authority of what office?

[Pippen]

I saw some guy play a game on KGS with around 30 thousand moves.

What the ...

I remember my longest game had about 400 moves.

[Bill Spight]

Yasui Sanchi vs. Ito Matsujiro has a game record from 1839 with 405 plays. (Black 405 takes a ko, which Black will win.)

Yamabe Toshiro vs. Hoshino Toshi has a game record from 1950 with 411 plays. (Black 411 also takes a ko which Black will win.)

I think that there has been a longer pro game with territory scoring since then, but I am not sure. OC, nowadays everybody fills the dame, so games are longer, eh?

[John Fairbairn]

The 2nd longest in Japan is 406 moves in 1996.

The longest in Korea is 384 in 1984.

The longest in China appears to be 394 in 2013.

The longest international game is 392 moves in 2013.

[EdLee]

Related: what's the max. number of stones used in a pro game ?
Assuming you cannot recycle stones.

Related: what are the odds you'll run out of stones with a standard set ( 181 + 180 ), even allowing for recycling of stones ?



Warning: The next post contains 2 giant SGF's which may cause Safari on an iPhone 4S to be unresponsive...
for a while, so it seems iOS or Safari has crashed.
If you wait long enough, Safari will eventually come back to life.

[vier]

The 2nd longest in Japan is 406 moves in 1996.
The longest in Korea is 384 in 1984.
The longest in China appears to be 394 in 2013.
The longest international game is 392 moves in 2013.

Yes, I find the same games when one is restricted to pro games in well-known tournaments. Here a 417-move record of a game between Ohashi Hirofumi, 4p and Mizuma Toshifumi, 7p, found on a blog.



Of course, constructed problems may contain very long move sequences. Here an example with a 3603-move sequence.

[macelee]

The Ohashi Hirofumi vs Mizuma Toshifumi game was training game by the two professional players. It also reminds me the discussions wer had a while ago on how many ko captures a game can have:

http://lifein19x19.com/forum/viewtopic.php?f=13&t=11243

[John Fairbairn]

Yes, I find the same games when one is restricted to pro games in well-known tournaments. Here a 417-move record of a game between Ohashi Hirofumi, 4p and Mizuma Toshifumi, 7p, found on a blog.

I'm aware of it, as it's on Ohashi's blog. But there he says (a) it was a study group game, and (b) he was recording it on his site "while I can remember it", which implies a possibility of mistakes in remembering (the same remark may be made about the 411 move game, BTW), and (3) in that connection there is a 308 move version floating around, which maybe is as far as they got with reliable moves. Also, as there's always an element of larking about in study groups, were they going for the record?

My reluctance to include it in the database is because the Nihon Ki-in didn't seem even to mention it, though I suppose I might now add it with the above provisos.

[gowan]

I saw some guy play a game on KGS with around 30 thousand moves. But maybe you mean serious game in a professional tournament? And official by the authority of what office?

30000 moves!? That is like 150 usual games on end. I assume they were using a rules set that allowed them to have a repetitive position.

[Uberdude]

I should clarify it was a game against himself with the sole purpose of trying to play as many moves as possible, rather than to win. Normal ruleset. His approach was basically to make a snake of 180 stones in atari, capture them, then 179, capture them, and so on. Sum of 1 to 180 is 180*181*0.5 is about 16200, double that gets you about 30000.

[snorri]

I should clarify it was a game against himself with the sole purpose of trying to play as many moves as possible, rather than to win. Normal ruleset. His approach was basically to make a snake of 180 stones in atari, capture them, then 179, capture them, and so on. Sum of 1 to 180 is 180*181*0.5 is about 16200, double that gets you about 30000.

I hope that was done with a program, otherwise I feel sorry for this person.

[snorri]

Yasui Sanchi vs. Ito Matsujiro has a game record from 1839 with 405 plays. (Black 405 takes a ko, which Black will win.)

Yamabe Toshiro vs. Hoshino Toshi has a game record from 1950 with 411 plays. (Black 411 also takes a ko which Black will win.)

I think that there has been a longer pro game with territory scoring since then, but I am not sure. OC, nowadays everybody fills the dame, so games are longer, eh?

They fill in the teire, too, which in classic games seems to have been often been left as an exercise to the reader. I guess we're lucky to have old records at all and shouldn't complain if the game recorder glanced over the board at move 180, mumbled, "looks like B+2" and went off for some sake. (That's only how it happens in my imagination, BTW and is not the result of extensive research. )