I've read in quite a few places that one of the best things a new (or experienced) player can do for their game is to study tsumego/tesuji. So, I've been working my way through some tesuji collections and doing other problems that I come across.

My question to all of you: How long do you give yourself before looking at the answer?

Context:

Last night I was trying to solve a problem that was above my level. I spent over an hour staring at it on my go board and still couldn't figure it out. Finally, in despair I laid out a couple patterns that looked promising, both of which failed. Then I checked the answer and was shocked that I hadn't seen it myself (the answer didn't strike me as particularly difficult in any way.)

I'm not sure whether to see this kind of experience as beneficial (because I pushed my reading skill to the limit for a prolonged period of time) or to see it as inefficient (because I could have solved 10-20 easier problems in the same amount of time.)

Also, do you think it's a good idea to forbid yourself from looking at the answer until you're confident you've read it?

_________________
Learning go in Japan.

My go blog (mostly in Japanese): http://komoku.tumblr.com
My other blog: http://nihonshock.com
Twitter: http://twitter.com/lloydvincent

I run across quite a few problems that I can't solve. I don't spend anywhere near an hour on them though. Not sure if that's the right way to do it though. I'll be watching for other replies.

You might want to do your problem solving with Drago or some other program. It saves all the time it takes to sort the stones. You can also save the problems with the solutions and wrong answers and quickly go back to look at them again.

I do think (for what it's worth) that looking at a problem on a board (actual or screen image) and going through the solution one move at a time better helps me internalize the sequences. It takes longer but I think it's more effective for me.

Tom

Since doing tsumego is obviously important, I'd be wary of spending too much time on hard problems, because that can easily lead to frustration, which can lead to not wanting to do any more tsumego.

If I can't get the answer in 10 minutes, I assume that the problem is too hard for me. I like the idea of doing about 2/3 easy problems (which I can solve in a minute or two, and 1/3 hard ones. I personally look at the answer most of the time and then, if my solution was not satisfactory review the sequences in my head while looking at the problem again.

Sometimes I will also decide to do a chunk of problems without looking at the answer, but this is much easier if I don't have the answers in the first place, for example, these: http://tsumego.tasuki.org/?page=tsumego.

Enjoy!

_________________
Patience, grasshopper.

I did not vote. I would have liked to vote 30sec, but that was not an option.

I'm not much of a tsumego solver, but if I do solve problems I rarely spend more than 1 min to find the answer (most within 20-30 sec)
I only solve problems of which I have the answer. If I don't check the result, I might wrongly assume some shape is alive or dead when it in fact is not.

here is a nice reference. As you can see spending over 1 min is very rare. If it takes more than 5 min the problem realy is too difficult (in my opinion)

_________________
Laurens
Go school: freegame's Teaching School
Author of: The Next Move a book for 15-3kyu players.

In "Graded Go Problems for Dan Players - Volume One" there are times given like "Solve in under 3 minutes - 1 dan" or "Solve in under 5 minutes - 1 dan".
I think 5 minutes is a good maximum, if you spent more than 5 minutes on a problem it is clearly to hard for you. Good reading is not only the ability to visualize many moves, but also the experience on how to prune the game-tree, knowing which moves to read, and which aren't worth reading.
Spending so much time on a tsumego probably means, you are reading the wrong moves (otherwise you would already have solved it), so your shape-intuition isn't on the required level for this tsumego.
I would say, spending to much time on a tsumego isn't wasted time, but an inefficent use of time. Reading mistakes in real games come probably from a laziness (ah, i just try it without properly reading it) or a bad shape-intuition (you don't see the important points in the problem and spend to much time on reading futile attempts). Both can be trained with easier tsumego. Harder tsumego is then used to further advance your reading ability, but with small steps.

If you have never run a great distance and you want to train for a marathon, you don't just try to run 42km at once. Instead you try a smaller distance and increase it step by step, until you reached your goal. I think its the same with solving tsumego to train your reading.

And of course there is the whole enjoyment aspect. Trying to solve one difficult tsumego in 30 minutes and still failing doesn't sound very appealing to me, especially if i could try and solve several easier tsumego where each one solved, gives me a nice feel of satisfaction.

_________________
Magics way up the hill (personal study journal)

I think the question you have to ask is: if this were a game situation, how much time would I spend? You have some advantage in tsumego knowing that an answer definitely exists. On the real board you'd probably look at it for about 30 seconds and decide there isn't a way. For that reason I think the most value in tsumego comes from transforming 1-5 minute problems into 15-30 second problems.

You definitely need to be willing to spend more than 30 seconds in a game, unless the time limits prohibit it, or you and your opponent have a tacit agreement to play 15 minute games.

_________________
Occupy Babel!

I generally don't like to spend more than 5 minutes on a problem, and I will spend a maximum of about 10 minutes on a problem if I am certain that I should be able to solve it. Generally though, I think 1-2 minutes is the best range of difficulty.

In any game situation when confronted by a problem, I don't have any clues about the status of a group or connection, and I have a hard time believing I would be able to spend more than 10 minutes reading a sequence without damaging my clock severely for the rest of the game. Unless it's a very long tournament game.

You should not spend more than 3 to 5 minutes on a problem. If you cannot solve it within this time, it is too hard for you.

If you have not solved the problem, you should never, never look at the answer. Better try some other problems, even in the same book. In the majority of cases there will be a problem some pages further with a similar clue to it's solution.

If you later (next day, next week) return to the problem you were unable to solve, it is likely that the solution will come in a glance.


Yes, it is.

Generally speaking, you should never look at the answer. This is because there is no "answer" given to you in a real game.

This advice may especially be true, when you have a teacher, who could tell you if you were right or not (he will not give you the answer though).
If there is no teacher, but only the book, I think it is tolarable to look at the answer. Especially to prevent you from using "wrong" or "ineffective" sequences in similar situations (or one of the next problems). But you should do this only after going through all the pros and cons of your "solution" found.

_________________
The really most difficult Go problem ever: https://igohatsuyoron120.de/index.htm
Igo Hatsuyōron #120 (really solved by KataGo)

I think it's personal preference. Get to the point where you really enjoy doing tsumego. Then the time you spend doesn't matter to you.

_________________
be immersed

First, I think it is absolutely fine to spend a little more time on a problem than you would in an actual game. My reasoning is this: when you are playing a game, you are watching the shapes and liberities develop. When you're solving problems, each new problem is a fresh start. For example, when a L&D problem develops in a game, I try to pay attention to all the liberties.

On the other hand, you could factor all of the time watching the shape develop into how much time you spend solving the problem, in which case you could say there is little difference in the time spent between book/program problems and in-game problems. But that's a bit pedantic.

All that said, IMO five minutes is a good limit.

If I'm going through a book, I'll look at the answer to confirm my solution (after I've reached a certain level of confidence about it). If I can't solve it, I'll mark it and come back to it later, without looking at the solution.

You can spend as much time as you want. The longer the better. Don't despair even if you cannot solve the problem. The thought process while you are trying to solve the problem will be getting stronger as you spend more time with problem. Main purpose of trying to solve problem is to increase your thinking process not solving particular problem.

My 2 cents.

it depends. when i do it on website, I spend few seconds. When i do a book, i may spend as much time as possible, and if i cannot find the answer i will mark it, do others first and come back later.

_________________
长考出臭棋.

Thank you everyone for the thoughtful and honest replies! Judging by the poll too, it seems that not many people are in favor of crossing the 5 minute mark, and the argument for compounding your tsumego ability bit-by-bit definitely makes sense.

I guess my idea for attacking the difficult problem in the first place came from my experience learning Japanese--where all information you learn is mutually reinforcing (i.e. studying something difficult expands understanding of simpler concepts). In learning Japanese I've also come to believe that half-effort never gets you anywhere.

Anyway, yesterday I tried two more similarly difficult problems and got them both in about 15-20 minutes each. I do feel this was good exercise for my reading and the sense of accomplishment I got from reading the 25+ step solutions correctly was absolutely blissful.

I'm going to change my general tsumego study plan to focus more on problems I can do in 1-5 minutes, but I'll continue to do harder ones (probably with a 30 minute limit) when I have the opportunity.



Everything I have read says that actually going through the pattern on a board is several times better than just reading from a book. They say that after a while your fingers will learn for you.



Worry not. Quitting out of frustration might be factor for some, but I'm a studying powerhouse.



It's interesting that you're the strongest player to respond, and your approach to tsumego seems much less involved than other players'.
Did you study tsumego differently when you were a kyu-rank player?
Do you think that a player's approach to tsumego should change as they get stronger?



My situation is kind of unique because I play more IRL games than online, and they're rarely timed. I guess I would feel bad going over 5 minutes though, especially if I couldn't come up with a decent play in that time.



Though I don't think tsumego always need to be treated as if they were a real-game situation, I agree that answers should (almost always) be reserved for confirmation after you've solved it.



Thanks! I said "despair" but what I really meant was "begrudgingly acknowledged defeat", giving up doesn't feel good but knowing when you're beat is a virtue (one I probably need to work on...).

_________________
Learning go in Japan.

My go blog (mostly in Japanese): http://komoku.tumblr.com
My other blog: http://nihonshock.com
Twitter: http://twitter.com/lloydvincent

It depends on why you're doing problems. ie: what do you hope to gain? If you want to become acquainted with tesuji and recognize shapes, it's not good to spend that long on a problem, because you'll probably be better off looking at the answer and going "oh, I wouldn't have thought of that", vowing to remember the idea, and moving on to get more problems looked at. Tesuji are especially like this.

If you're doing problems to practice your reading (how many moves you can hold in your head specifically), it doesn't really matter. You can spend a lot of time on one problem or lots of problems, as long as you're reading out variations the whole time.

I'm sure many would disagree with me, though, so I don't think there's a dogmatic answer about the best way to approach tsumego. Until someone does a scientific study about go skill gain and problems, it's going to be all conjecture and anecdotes. So pick a way of thinking that makes sense to you and go with it.

As long as you're spending time on it and progressing in go skill at some measurable pace, and enjoying the process, you can't be doing things too wrong, right?

_________________
1k KGS

I'm not a big fan of tsumego, and I mostly like to do tesuji because I find L&D to be too tedious. Regardless of the problem, I never spend more than a few minutes. My feeling after that is if I haven't been able to get it in a few minutes, then my reading is inefficient enough that I've in effect already gotten the problem wrong. It's like doing a math problem which you can't solve through the actual methods that you're supposed to use, but instead stumble upon the answer randomly.

_________________
"There are no limits. There are plateaus, but you must not stay there, you must go beyond them. If it kills you, it kills you. A man must constantly exceed his level." -- Bruce Lee

Just make sure you get the pattern right. I entered one problem then stared at it for a couple minutes and decided it was impossible. I flipped to the solution and saw that I had left a stone out. No wonder I couldn't solve it!

I think this is good advice. Sometimes you can't solve a problem not because it's too difficult to read, but because you somehow didn't notice a lack of liberties or you looked at it in the wrong way. If it looks impossible after 3 min I go to the next problem. I try the problem I couldn't solve again next day and usually solve it pretty fast.

I seldom spend more than five minutes on a problem because I'm not that patient No idea if this is good or bad.

Sometimes I like to solve many easy problems very quickly (usually problems I've solved slowly a few months before).

It is rare that I completly agree with Kirby, but this is the essence. We all have our opinions - some say do easy problems fast, some say do slightly hard problems and don't give up. Listen to people's advice, but do what keeps you doing it.

Because ultimately you MUST do problems, and you WON'T if you DON'T enjoy it.

Time spent not studying tsumego will be regretted, as Rick said to Ilsa - "Maybe not today, or tomorrow, but soon and for the rest of your life"

Why? I didn't ever put time aside to regularly do problems, and still haven't.