It is currently Mon Nov 20, 2017 2:28 am

All times are UTC - 8 hours [ DST ]




Post new topic Reply to topic  [ 68 posts ]  Go to page Previous  1, 2, 3, 4  Next
Author Message
Offline
 Post subject: Re: Iyama's world ranking
Post #21 Posted: Sat Nov 05, 2016 6:51 pm 
Lives in sente

Posts: 1208
Liked others: 258
Was liked: 322
Rank: 5d
GD Posts: 1000
idontgetit wrote:
gowan wrote:
From my position as a watcher of pro games I think a big factor in the shorter time limits is the watchability, in real time, of those games. This must definitely influence sponsors since their commercials would get greater exposure. Personally I've enjoyed watching real time long time limit games on the internet. I try to find the next move and what the players might be thinking about. As for errors, even with unlimited time games people will make
errors because no one, not even alphago, plays perfect go. My impression is that the short time limits push players to make gambles, playing risky moves, and hoping the opponent will make a mistake. Of course, people make mistakes in slow games too. But I want players to win by outplaying their opponents, not relying on chance. I admire the attitude of Otake Hideo 9p who resigned a game when he was dissatisfied with his play in a game in which he was ahead. Another admirable thing is how a two stone game between Dosaku (w) and Yasui Shunchi in 1683 considered Dosaku's masterpiece even though Dosaku lost by one point. There is so much more in go than winning.


Actually, pros tend to make less "gambles" in faster games. The thing is, pros usually don't like to completely lose after a single, huge battle. They would rather pick a way that would be less risky, and with less time, many more moves become risky (because with more time, they can read it out and realize that's it's actually not risky).


I think you've got it reversed. A visiting Japanese 9p told me that the fast TV tournament games have more complicated fighting because no one can read things out at 30 seconds per move. In the long time limit games they can read out the fight and see whether it is worth initiating the fight. In the fast game they may start the fight based on "intuition" rather than reading. My observation of TV games confirms that in these games there tends to be more risky fighting.

Top
 Profile  
 
Offline
 Post subject: Re: Iyama's world ranking
Post #22 Posted: Sat Nov 05, 2016 8:09 pm 
Judan

Posts: 7585
Liked others: 1377
Was liked: 1161
KGS: Kirby
Tygem: 커비라고해
Couldn't it depend on the pro? Some people are more risk averse than others. Given a situation you can't totally read, your intuition plays a role in your decision, and I doubt every pro reacts the same way under time pressure.

_________________
What have I learned from this game that I'll be able to use in my own game if the opportunity presents itself?

Top
 Profile  
 
Offline
 Post subject: Re: Iyama's world ranking
Post #23 Posted: Sat Nov 05, 2016 10:37 pm 
Dies with sente

Posts: 112
Liked others: 1
Was liked: 11
Sure. Of course it depends on the pro. But let me ask you this. How often do you see a pro game finished by move 150? Move 130? Move 100?

Not very often at all. That's what I mean. They might "fight", but it won't be a battle they absolutely positively MUST win. There has to be some flexibility in an eventual trade that still keeps the game relatively balanced or something.

Given a 60% chance for a kill that will end the game right there, and another way that will net them maybe 4-5 points advantage to endgame, they'll pick the latter. The vast majority will, anyway.

But with longer time, that 60% could become 80% or even 100%.

And, I've also heard pros comment in NHK cup and in weiqitv about how they end up not taking risks probably because of time pressure.

Top
 Profile  
 
Offline
 Post subject: Re: Iyama's world ranking
Post #24 Posted: Sun Nov 06, 2016 4:04 am 
Gosei

Posts: 1901
Liked others: 12
Was liked: 2814
idontgetit: logic is a treacherous mistress because one step seems to lead seductively and inexorably to the next. But if you change the starting point you usually end up, just as seductively and inexorably, up another staircase.

If you change your starting point here, you can also get a different perspective. Start with the premise that pros are super good at making good shape and so rarely leave groups in a state where they can ever be killed, then a cataclysmic fight may never be likely to occur. To follow Thurber, when challenged to a duel, you can choose swords but you can also choose nuclear submarines or dirty words at 30 paces. Slightly different scenarios in each case. But you can also avoid challenges.

In any event, early finishes to games may be more common than you suspect. In the last screen page of the GoGoD database, only 4 out of 43 games went to a count. One third finished in under 200 moves, which I suggest is enough to say they usually ended after the middle game and before the endgame.

More important, you may wish to look up a very long article on the topic of risk in Weiqi Tiandi by Wang Xi. He pinpointed the different approaches in national style between Japan, Korea and China in terms of their approach to risk. He identified the Japanese as relatively much more risk averse (their characteristic souba style). One obvious factor to consider in that was, of course, their much longer time limits.

Top
 Profile  
 
Offline
 Post subject: Re: Iyama's world ranking
Post #25 Posted: Sun Nov 06, 2016 7:13 pm 
Dies with sente

Posts: 112
Liked others: 1
Was liked: 11
we're not having a poetry competition here.

Having good shape doesn't mean pros don't make weak groups. In fact, it's very rare to see pros without weak groups. There are certainly a lot of times when pros could choose a more risky way to attack that might get a better result, but they choose not to.

But all of that is just me trying to rationalize the point. The reason I hold it though, is simply because I heard pros say it in commentaries that I have watched. That's all.

But it is interesting how pros can have wildly differing opinions on the same things. One thing I find funny is how guys like Meng Tailing ('87) always talks about how he is an old pro now, and can't compete with younger pros in close fighting and reading. Even guys like Shi Yue (born '91) are considered "old" in the go world.

Compared to that, the Japanese go world still sees guys like Terayama Rei ('90) and Mukai Chiaki ('87) as "wakatte".

EDIT: I would also like to clarify my other point before.

I am NOT saying that longer games WILL result in a lower quality game.

What I am saying are two things:

1. OCCASIONALLY, you will have found a move with a shorter time, but then because of a longer time, you end up over thinking and choosing another move, when the first move was actually better. OCCASIONALLY.

2. Longer games ARE NOT NECESSARILY better quality than say 4 hour ones. could they be slightly better sometimes? Sure. But they're probably not better all the time, and not by a huge amount. I mean, do you think a 10000 hour time limit would be significantly better than a 9950 hour time limit? Obviously not. There IS a rate of diminishing returns.

The question is just about how much marginal return can you get with an extra 4 hours, when you have 4 already? On one hand, your time just doubled. Which seems like a lot. But on the other hand, people do have a limit with how much their brain can process.

Top
 Profile  
 
Offline
 Post subject: Re: Iyama's world ranking
Post #26 Posted: Sun Nov 06, 2016 7:50 pm 
Judan

Posts: 7585
Liked others: 1377
Was liked: 1161
KGS: Kirby
Tygem: 커비라고해
Sooner or later we'll have computer software that can help in evaluating move quality, so the generalizations about game time to move quality can be more precisely known.

_________________
What have I learned from this game that I'll be able to use in my own game if the opportunity presents itself?

Top
 Profile  
 
Offline
 Post subject: Re: Iyama's world ranking
Post #27 Posted: Sun Nov 06, 2016 8:37 pm 
Lives with ko

Posts: 189
Location: Chicago, IL
Liked others: 159
Was liked: 35
Rank: KGS 3 kyu
Kirby wrote:
Sooner or later we'll have computer software that can help in evaluating move quality, so the generalizations about game time to move quality can be more precisely known.


Hopefully sooner.

_________________
Go Books

Top
 Profile  
 
Offline
 Post subject: Re: Iyama's world ranking
Post #28 Posted: Sun Nov 06, 2016 9:40 pm 
Lives in sente

Posts: 1108
Liked others: 0
Was liked: 436
Kirby wrote:
Sooner or later we'll have computer software that can help in evaluating move quality, so the generalizations about game time to move quality can be more precisely known.


Even if Computer software do this stat, it is not going to make any difference. There will be always someone complaints. There is saying "Garbage in, Garbage out.".
Someone will manipulate input to software to produce results they wants. All stats are just reference not absolute. Just leave it alone.

Top
 Profile  
 
Offline
 Post subject: Re: Iyama's world ranking
Post #29 Posted: Sun Nov 06, 2016 10:06 pm 
Judan

Posts: 7585
Liked others: 1377
Was liked: 1161
KGS: Kirby
Tygem: 커비라고해
trout wrote:
Kirby wrote:
Sooner or later we'll have computer software that can help in evaluating move quality, so the generalizations about game time to move quality can be more precisely known.


Even if Computer software do this stat, it is not going to make any difference. There will be always someone complaints. There is saying "Garbage in, Garbage out.".
Someone will manipulate input to software to produce results they wants. All stats are just reference not absolute. Just leave it alone.


True, I see your point. On the other hand, I think analysis software like CrazyStone (but stronger) can be more objective than a human. The argument of move quality vs. time limits has been discussed here often, and typically the same set of people have the same opinions.

Software could at least add semi-objective analysis to the argument instead of all the hypotheticals. I agree that people can pick and choose their data set, but at least there is real data in that case.

Not that it really matters in any case.

_________________
What have I learned from this game that I'll be able to use in my own game if the opportunity presents itself?

Top
 Profile  
 
Offline
 Post subject: Re: Iyama's world ranking
Post #30 Posted: Mon Nov 07, 2016 8:25 am 
Tengen

Posts: 4044
Location: North Carolina
Liked others: 411
Was liked: 662
Rank: AGA 3k
GD Posts: 65
OGS: Hyperpape 4k
The real difficulty is that changes in theory matter too. So to answer the question, we have to distinguish cases where the pro makes a mistake because they misread from the case where they didn't know about theory that would be developed later.

It's a variant of the old "how would Shusaku do if he could learn modern joseki?" question.

_________________
Occupy Babel!

Top
 Profile  
 
Offline
 Post subject: Re: Iyama's world ranking
Post #31 Posted: Mon Nov 07, 2016 8:27 am 
Judan

Posts: 7585
Liked others: 1377
Was liked: 1161
KGS: Kirby
Tygem: 커비라고해
Theory is a factor in game quality.

_________________
What have I learned from this game that I'll be able to use in my own game if the opportunity presents itself?

Top
 Profile  
 
Offline
 Post subject: Re: Iyama's world ranking
Post #32 Posted: Mon Nov 07, 2016 9:00 am 
Tengen

Posts: 4044
Location: North Carolina
Liked others: 411
Was liked: 662
Rank: AGA 3k
GD Posts: 65
OGS: Hyperpape 4k
Kirby wrote:
Theory is a factor in game quality.
The question wasn't "are old games better than current games in every way?", but whether more time makes for better play, using old games with longer time limits as one source of information. And longer time limits can make players better apply the theory they already have, and read better, and make better positional judgments, but it won't let them learn theory that took decades to develop.

_________________
Occupy Babel!

Top
 Profile  
 
Offline
 Post subject: Re: Iyama's world ranking
Post #33 Posted: Mon Nov 07, 2016 10:07 am 
Judan

Posts: 7585
Liked others: 1377
Was liked: 1161
KGS: Kirby
Tygem: 커비라고해
True, I see your point that we cannot absolutely answer the time limit question, even with an oracle. Your point about developed theory is one example among many.

Nonetheless, an objective measure of game quality is still something I look forward to, because it's much better than the hypothetical discussions we have now.

But anyway, yes, I agree that you still cannot definitively answer the time limit question, even if you can objectively measure game quality.

_________________
What have I learned from this game that I'll be able to use in my own game if the opportunity presents itself?

Top
 Profile  
 
Offline
 Post subject: Re: Iyama's world ranking
Post #34 Posted: Mon Nov 07, 2016 12:32 pm 
Tengen

Posts: 4044
Location: North Carolina
Liked others: 411
Was liked: 662
Rank: AGA 3k
GD Posts: 65
OGS: Hyperpape 4k
I didn't mean to imply it was impossible, though my first post could be read that way. There's a lot of this kind of work going on in chess, and I think it's providing value. It just also provokes interminable arguments about what it all means.

_________________
Occupy Babel!

Top
 Profile  
 
Offline
 Post subject: Re: Iyama's world ranking
Post #35 Posted: Mon Nov 07, 2016 3:14 pm 
Lives with ko

Posts: 251
Liked others: 95
Was liked: 114
Rank: AGA 6k
I'm not that optimistic about computers shedding light on move quality. It's not clear it would be better than modern pros commenting games of past players, it's just a different set of biases.

For any given move in a game, from computer analysis one can typically get:

1. A probability that black will win.
2. An expectation value for the score.
3. Candidate next moves.

Let's say that black is ahead (probability of winning > 0.5) and it's black's move. From one point of view black can play any move that does not result in being behind, and although humans might consider some of those moves too slack because they believe it should be safe to try for more, it's hard to say whether those are mistakes without looking at the position in more detail. I suppose if a move both lowers the probability of a win and worsens the score then it's likely a mistake.

If black is behind, it's even harder to say what a mistake is, for the same reason that it's hard to review the white side of handicap games. What's the best overplay? Or should one play as if the score is even and wait for the opponent to make a mistake? We have some evidence that computers often don't play in a way that's strategically effective against humans when behind. To do that, one might have to at least take into account a strength estimate of the opponent. It's not an easy problem for either humans or computers. :)

So for half the moves the game---i.e., the ones made the player who is currently behind, it may be very hard to come up with a metric for move quality. Any metric may have distortions that punish or reward highly tactical players, for example.


This post by Calvin Clark was liked by: Waylon
Top
 Profile  
 
Offline
 Post subject: Re: Iyama's world ranking
Post #36 Posted: Mon Nov 07, 2016 3:45 pm 
Judan

Posts: 7585
Liked others: 1377
Was liked: 1161
KGS: Kirby
Tygem: 커비라고해
Well, AlphaGo has trained a function for evaluating a position. Presumably it fares better than human judgment. It would seem to me that such a function could be used to evaluate a move.

CrazyStone already has this type of thing, which is useful in reviewing amateur games. Take that and crank it up to pro strength.

Not sure of the details, but maybe computers can help us gain insight in the future.

_________________
What have I learned from this game that I'll be able to use in my own game if the opportunity presents itself?

Top
 Profile  
 
Offline
 Post subject: Re: Iyama's world ranking
Post #37 Posted: Mon Nov 07, 2016 4:30 pm 
Dies with sente

Posts: 112
Liked others: 1
Was liked: 11
I definitely think it could be done, at least to a point that is superior and more precise to human professional judgement.

Top
 Profile  
 
Online
 Post subject: Re: Iyama's world ranking
Post #38 Posted: Sun Jan 29, 2017 1:59 am 
Lives in gote

Posts: 603
Liked others: 41
Was liked: 190
GD Posts: 10
Iyama Yuta just won against Ke Jie.
According to non-goratings ranking, this would be called an upset due to Iyama's low rank.
But not in goratings, as they're both in top 10.

Top
 Profile  
 
Offline
 Post subject: Re: Iyama's world ranking
Post #39 Posted: Sun Jan 29, 2017 2:13 am 
Dies in gote

Posts: 47
Liked others: 3
Was liked: 8
pookpooi wrote:
Iyama Yuta just won against Ke Jie.
According to non-goratings ranking, this would be called an upset due to Iyama's low rank.
But not in goratings, as they're both in top 10.


When did Ke Jie get out of the hospital? I'd been worried for him for a while because he was playing to exhaustion. Sure enough, he was hospitalized earlier this month.

Image

Top
 Profile  
 
Offline
 Post subject: Re: Iyama's world ranking
Post #40 Posted: Sun Jan 29, 2017 9:15 am 
Dies in gote

Posts: 26
Liked others: 2
Was liked: 3
Rank: 4K
pookpooi wrote:
Iyama Yuta just won against Ke Jie.
According to non-goratings ranking, this would be called an upset due to Iyama's low rank.
But not in goratings, as they're both in top 10.


I'm surprised Iyama won. This is quite the achievement for Iyama, even if Ke Jie is in poor health and was just discharged from the hospital. I got the impression that Ke Jie at his full strength is almost a stone stronger than Iyama.

Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 68 posts ]  Go to page Previous  1, 2, 3, 4  Next

All times are UTC - 8 hours [ DST ]


Who is online

Users browsing this forum: No registered users and 7 guests


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