Go problems don't bring any result?

[Shenoute]

I second that. Lots of easy problems, some intermediate, but trying to read them completely, i. e. every possible sequence, even silly ones. As I see it, at some point tsumego is about training reading skills, not finding the right answer, so reading every possible sequence in a problem ensures you're using the problem to its full extent.

Plus, you say that you do not see improvement but without regular tsumego practice you would probably be weaker by now. Reading skills have to be maintained.

[RobertJasiek]

"Complete" tactical reading wastes time because Regular Reading etc. suffice. Not reading per se must be complete but Regular Reading must clarify all relevant tactical choices. In particular, at branches with at least one successful optimal move, it is sufficient to find one.

[swannod]

I second that. Lots of easy problems, some intermediate, but trying to read them completely, i. e. every possible sequence, even silly ones. As I see it, at some point tsumego is about training reading skills, not finding the right answer, so reading every possible sequence in a problem ensures you're using the problem to its full extent.

Plus, you say that you do not see improvement but without regular tsumego practice you would probably be weaker by now. Reading skills have to be maintained.

I disagree about reading silly lines. To me doing problems are about finding the opponent's *best* counter. A move looks like it works but no it doesn't. In my experience when I misread a simple problem in my game it boils down to I choose a move that looks good on the surface but I didn't find the counter. Or a I choose move that works but I didn't read out what happens when the opponent plays the best counter leading to a mistake.

This is already a ton of work and you have to do this under time pressure. Reading out unlikely lines doesn't seem useful to me (unless your reading says none of the seemingly likely lines are working)

[longshanks]

I can't really offer much advice here other than one thing. Make sure you do them properly. Where properly means reading out all combinations and filtering out those which don't work. I used to do them by guessing. A lot of apps just show you the solution rather than allowing you to 'prove' your solution with the computing trying to refute. Or better play out on a board. And when doing them properly it's OK to step down to super easy ones as it will be harder. Then work back up again.

[Bill Spight]

What are reading skills?

There are at least three:

Calculation of variations
Choice of candidate moves
Evaluation of resulting positions

Last I heard, in chess there was research that indicated that there is not much difference between average players and experts in the ability to calculate variations. The main difference is that average players calculate a lot of irrelevant variations.

Is go different? Maybe. Pros seem to be much better at the calculation of variations than amateurs. However, there has been research that indicates that pros have a larger neural workspace for those calculations than amateurs. IMO, because they built their brains that way when they were children. For most of us it is too late to do that.

But I know from my study of hypnosis that it may be possible to develop visualization techniques to improve the calculation of variations, much as, before the advent of printing, people had developed visualization techniques to improve memory. But I have not heard of anybody developing such techniques, and just doing problems is not the way to do so.

OTOH, dithering is bad. E. g., Does this play work? No. Well, how about this play? No, it doesn't work either. But what about the first play? Maybe I missed something. No, it still doesn't work. But maybe I missed something with the second play. Etc., etc. If doing problems improves your discipline and organization of thinking, that is good.

Gotta run. More later.

[daal]

The main difference is that average players calculate a lot of irrelevant variations.

I've gotten a good deal better at tsumego by trying to develop the following habit: start by examining the move that experience tells me is most likely to work. It might seem like a no-brainer, but without this explicit admonition, my natural tendency is marvel at how the problem seems impossible to solve and then to start trying to prove myself right.

[Bill Spight]

Here is a quote from David Ormerod on a topic we agree about. (From viewtopic.php?f=15&p=85964#p85964 )

Even worse is the school of thought that tells everyone that you don't need solutions. That's partially true when you already have a comprehensive understanding of shape and are strong enough to read a problem out completely. When I do easier problems I often don't need the solutions anymore. However, when I do hard ones, I still get some problems wrong, no matter how hard I try. When I get them wrong, I usually learn a new tesuji or something else to watch out for by looking at the solution. It finds the weaknesses in my reading and fixes them. If I didn't look at the solution I'd continue in the naive belief that I'd solved the problem and would keep using my inferior moves.

I'm going on about this because I find it very regrettable to see enthusiastic players who love Go and really want to improve work through a whole lot of problems and not start seeing results. A lot of people lose motivation and give up at that point and usually it's because they've followed the stupid advice about doing problems without solutions.

Advice that may apply to pros and very strong players (and many pros don't even agree and do look at the solutions) doesn't generalise to everyone. Would you honestly tell a 30kyu, "OK here are some practise problems, but just look at them, don't check the solutions. And these books over here, don't even read any of these, just work everything out from first principles and you'll be fine."

It's a pernicious and specious piece of advice that gets parroted from player to player and I think it's a load of rubbish.

Edit: Corrected misspelling of Ormerod. My apologies, David.

[Cassandra]

Here is a quote from David Omerod on a topic we agree about. (From viewtopic.php?f=15&p=85964#p85964 )

Even worse is the school of thought that tells everyone that you don't need solutions. ...

I am afraid that this "school of thought" goes along with a massive misunderstanding with regard to the "cirumstances".

As a matter of course, you do not need a problem's solution, if you have someone at hand who will check your own "solution" sequence.

That someone at your hand is not very likely to be available in the Western world. This implies that there is a need to also have access to a problem's solution. However, you will have to deny your desire of looking at it "too early".

[Gotraskhalana]

but trying to read them completely, i. e. every possible sequence, even silly ones.

The advice "to read out every possibly sequence" is simply a lie which is easy to check by calculation. Noone can do this.

[John Fairbairn]

I don't believe there is a "school of thought that tells everyone that you don't need solutions."

While I don't deny that there may be some people who do say that (but who?), what I think has happened is that advice "don't look at the solutions" has been transmogrified into "you don't need solutions".

Further, articles don't exist in the Oriental languages and so speakers from there are liable to say "don't look at solutions," which maybe makes the transmogrification more likely. But the proper translation is nearly always "don't look at the solutions," which clearly implies solutions are at hand and must be there for the purpose of looking at. Maybe 8 times out of 10 this is explained in a book's preface, with the real, fuller meaning being that you shouldn't look at the solutions prematurely.

That's when things get confusing because the definition of 'premature' varies according to author, and that may depend, for example, on whether he is presenting the problems as entertainment for puzzlers (a quite common genre) or as ways of imbibing useful techniques for actual play.

In the latter case, the goal is not to solve a problem so much as to acquire a technique, and whether you acquire it best by struggling over one problem or by quickly flipping through a dozen or more problems of the same type (e.g. Meijin Inseki's style) is really only something you can decide for yourself. You can't expect an author to know what type of person you are. You have to take responsibility for yourself sometimes, even in the modern, lazy, spoon-feeding internet age.

[daal]

(but who?)

I read this advice on Sensei's Library in an article about getting stronger written a dozen years ago by a strong German amateur: http://senseis.xmp.net/?BenjaminTeuber% ... meStrong#2 I can't recall having read it anywhere on paper.

[Kirby]

(but who?)

I read this advice on Sensei's Library in an article about getting stronger written a dozen years ago by a strong German amateur: http://senseis.xmp.net/?BenjaminTeuber% ... meStrong#2 I can't recall having read it anywhere on paper.

The Tatsuki Tsumego pdfs also have a preface that points out the lack of solutions. He notes:

I think we can learn more by actually solving the problem, trying all possible variations

This is not the same as saying solutions are useless or anything like that, but maybe it could be interpreted by some to support not having or utilizing a solution.

Regarding my own opinion, I think balance is required. It reminds me of my job in the office. When learning something new, on one hand, I don't want to "keep bugging" the expert in a given area. If I need to figure something out that's not in my own domain, I try to figure it out on my own.

At some point, for challenging problems, though, it can be useful to swallow my pride, gather my questions, and go chat with the domain expert. Usually, they have years of experience and context, and can explain how things work in 5 minutes.

After this, I have a fresh context on the problem, and I'm a little bit more of an expert myself.

If I go to the expert every time I have the slightest confusion, I don't understand or appreciate the complexity of the problem. But if I've gotten deep into it and start just spinning my wheels, consulting an expert can give me a fresh perspective that I can truly appreciate since I've already invested myself in the problem.

It's a balancing act.

[Bill Spight]

A bit more on looking at the solution and failure diagrams

You need to know why a mistake fails, right?

The idea of reinforcement is basic to learning, as is the idea of imitation. Take the latter first. Looking at the solution gives you something to imitate. Imitation can take you pretty far, but obviously is not enough for a game as difficult as go. Still, it is important, and not to make use of it by not looking at the solution would be a shame. As for reinforcement, there are two kinds, positive reinforcement and negative reinforcement, related to reward and punishment, respectively. A game of go provides both: positive reinforcement if you win, negative reinforcement if you lose. One problem, OC, is the question of which among the typically more than 100 moves that you made in the game are good, and should be positively reinforced, and which are bad and should be negatively reinforced? Not always obvious. For problems, solution and failure diagrams provide positive reinforcement to good moves and negative reinforcement to bad moves. Those are very important in the learning process. You ought to look, if only for the satisfaction of verifying your reading (positive reinforcement).

Now, for reinforcement to be effective, you need something to reinforce. Just looking at the solution may invoke imitation, but is useless for reinforcement. You have to work on the problem first. If you are pretty sure that you have solved the problem, then might be a good time to look at the solution and failure diagrams. But what if you are stuck? How long do you spend reading out a difficult position during a game? 1 minute? 2 minutes? 5 minutes? 15 minutes? 1 hour? I expect that there is a sweet spot, but I do not know what it is. As a dan player I was willing to spend 15 minutes, but I think that much time would be counter-productive for most SDKs.

Remember dithering? Going back and forth between apparently unacceptable alternatives? One problem with spending too much time on a problem is that you end up essentially dithering, by repeating unsuccessful lines of play while looking for alternative plays. If you are stuck, you are not finding those good alternatives. But by that process you are positively reinforcing bad plays. Reinforcement is not just about reward and punishment, it is about strengthening neural pathways. Every time you repeat a bad variation, you are positively reinforcing it, even if it leaves you feeling unsatisfied at the end (negative reinforcement). Positive reinforcement is more powerful than negative reinforcement. Taking too long on problems can build bad habits of thinking about go.

Chess grandmaster Kotov advised in calculating variations to calculate each branch of the game tree only once. Obviously, that avoids dithering. That advice has generally been rejected for actual play. However, as a discipline in solving problems I think that it is excellent. You choose your candidate moves at each turn and read each branch to whatever depth seems right at the time (or to the limit of your ability), and if you find no solution, you are done, and you have not reinforced any bad moves. Now you are ready to look at the solution and failure diagrams.

Edit: Having considered this some more, I see that I ignored the role of the subconscious in solving problems. If you have spent time and energy trying to solve a problem, it can help to leave it for a while for your subconscious to work on. Overnight is not a bad period of time. I have sometimes woken up with the answer to a problem or question that I had been working on.

[John Fairbairn]

Now, for reinforcement to be effective, you need something to reinforce.

For me this is the key point, and it applies in all phases of the game. It is almost never treated properly in problems, and surprisingly rarely in other phases.

Meijin Inseki thought that the best way to teach tsumego skills was to show a technique and then just give a pile of problems with the same technique. Drills, in other words.

Now it may seem obvious that there should be value in hair-shirtedly tackling a problem entirely on your own, without knowing what technique is needed. Virtuous, maybe, but not necessarily valuable. There are at least two drawbacks there. One is that, even if you discover the technique, in the current literature there are not enough tightly tailored drills to practise your reinforcement.

Another difficulty is that, left to your own devices (cilices, awls or otherwise), you might not discover the technique at all. That's not rare and applies even to top pros. We all have blind spots. I was looking today at Kubomatsu's book on his famous radio game with Go Seigen when he started playing tengen in public. Part of the radio format was that each player had to give their post-mortem thoughts on their own moves. Kubomatsu had already done quite a lot of research on tengen but said he was astounded by what Go said when Go explained how he thought about White 2. Go said he first considered moves next to the tengen stone. The very idea of such moves had escaped Kubomatsu. He had been aware of Go and Kitani playing sanrensei with White 2-4-6 (it was the Shin Fuseki era) and he saw tengen as a counter to that. So moves such as an attachment against tengen were a kind of blind spot for him

One reason that playing over lots of pro games is so valuable is because of the wealth of "Eh, I didn't know you could do that!" moments.

These moments exist in tsumego, too, and it seems like a useful shortcut to be shown the possibilities first. Why reinvent the wheel before you learn to drive?

Another argument against the "read it out to the bitter end even if you can't" school is that even pros can't. Readers of my Gateway to All Marvels (Gengen Gokyo) will have noticed how often standard solutions have been promulgated by pros for decades or even centuries only for someone like Go Seigen to come along and point out a fatal flaw. Reading is important, but it seems much more valuable if you want to practise your reading of English literature to read mostly within your comfort zone (reinforcement) backed up with a dictionary which you consult for new words (and then use: imitation), rather than try work out what they might mean. And it's certainly a lot better than practising English by trying to decipher Linear B (or parts of Igo Hatsuyoron in the case of go).

Unless of course you really like doing puzzles. Even then, puzzles and playing life and death positions in actual go are two separate activities really, and it would be good to stress that more often.

[Kirby]

How long do you spend reading out a difficult position during a game? 1 minute? 2 minutes? 5 minutes? 15 minutes? 1 hour? I expect that there is a sweet spot, but I do not know what it is. As a dan player I was willing to spend 15 minutes, but I think that much time would be counter-productive for most SDKs.

Perhaps the sweet spot is to spend as long as possible until you start to dither.