How to do Tsumego

[paK0]

So, I've been wondering about that lately and I'm interested in what you guys think.

Usually when I do Tsumego I try to read everything out and only then look at the solution.

But I've heard that some people think its more productive to do a lot of them fast and I was wondering how that works. Look at it for 10 seconds and then take your best guess? I can see that since you see more problems in a given timeframe your shape recognition might get better, but don't you miss out on reading practice that way?

[nacroxnicke]

no, you should always read the variations. When people say do tsumego fast, they don't mean that one shouldn't read them, they mean that one should do easy ones for your level and try to read faster than usually.

[Kirby]

Yes, nacroxnicke is right.

Doing many fast tsumego means doing problems that are easy enough that you can read them out completely in a short amount of time.

Some people advocate easy problems, but I think both easy and hard problems are useful (as long as they are not so hard that you give up).

[Bill Spight]

So, I've been wondering about that lately and I'm interested in what you guys think.

Usually when I do Tsumego I try to read everything out and only then look at the solution.

But I've heard that some people think its more productive to do a lot of them fast and I was wondering how that works. Look at it for 10 seconds and then take your best guess? I can see that since you see more problems in a given timeframe your shape recognition might get better, but don't you miss out on reading practice that way?

IMO, to answer that question depends on a number of factors. First, what is your level of tsumego? I don't mean the level of the problems you work on so much as the level at which you get about half of them right. Second, how old are you? Third, how long have you been playing go?

[DrStraw]

http://senseis.xmp.net/?SteveFawthrop%2 ... feAndDeath

[Kirby]

So, I've been wondering about that lately and I'm interested in what you guys think.

Usually when I do Tsumego I try to read everything out and only then look at the solution.

But I've heard that some people think its more productive to do a lot of them fast and I was wondering how that works. Look at it for 10 seconds and then take your best guess? I can see that since you see more problems in a given timeframe your shape recognition might get better, but don't you miss out on reading practice that way?

IMO, to answer that question depends on a number of factors. First, what is your level of tsumego? I don't mean the level of the problems you work on so much as the level at which you get about half of them right. Second, how old are you? Third, how long have you been playing go?

Interesting... Could you share your opinion for each of those conditions (eg. if you are young, X is good, etc.)?

[paK0]

So, I've been wondering about that lately and I'm interested in what you guys think.

Usually when I do Tsumego I try to read everything out and only then look at the solution.

But I've heard that some people think its more productive to do a lot of them fast and I was wondering how that works. Look at it for 10 seconds and then take your best guess? I can see that since you see more problems in a given timeframe your shape recognition might get better, but don't you miss out on reading practice that way?

IMO, to answer that question depends on a number of factors. First, what is your level of tsumego? I don't mean the level of the problems you work on so much as the level at which you get about half of them right. Second, how old are you? Third, how long have you been playing go?

Yeah, I'd also love some elaboration on that, after that I can piece together what would fit for me the best.


@DrStraw: Interesting read, seems like repetition is the key.

[Boidhre]

no, you should always read the variations. When people say do tsumego fast, they don't mean that one shouldn't read them, they mean that one should do easy ones for your level and try to read faster than usually.

I've read a slightly different take on this, fast tsumego aren't necessarily about reading, they're about reinforcing shape recognition we already have. Say a 2k going through 1001 Life & Death Problems quickly won't be that concerned with reading out the tsumego fast but checking to see if they recognised the shapes involved correctly. I've never seen this recommended as a sole method of study, but I can see the logic of it for like the above someone sitting down with a problem book that is very easy for them and they're just drilling shapes more than reading (mainly because this shape knowledge helps with the tougher problems later on etc).

*shrugs*

[Boidhre]

Actually if your level is around where I think it is (15-16k KGS level?) then I think this is a good book: http://www.ymimports.com/p-1413-essenti ... rpYafRdUQo

This is not a traditional problem book. The problems may be easy for you (at least in the earlier parts). But each page drills you on a particular shape and/or life and death technique. 6 problems, starting with a trivial one and making each one slightly more complicated. I say it's not a problem book in the traditional sense because the technique is named at the top of the page. The idea is drilling rather than problem solving as it is normally presented.

If you're looking for a pretty thorough grounding in basic L&D techniques then this is a pretty good book I think. (Oh, btw, there are no included answers, you'll have to hunt around online for the pdf of them, legit just the book is meant to be solved without them and the answers only used if a particular problem is confusing/unclear).

[paK0]

Thanks for the tip, I'll take a look at it, its a little below my level(12k kgs) but if I can't handle 1001 life and death problems in a reasonable way(it should arrive soon) I might order that one.

[Boidhre]

Thanks for the tip, I'll take a look at it, its a little below my level(12k kgs) but if I can't handle 1001 life and death problems in a reasonable way(it should arrive soon) I might order that one.

Nah, not below at all! For drilling stuff you don't mind if the problems are a bit or quite easy, you're trying to reinforce what you know not necessarily just learn new techniques (which you also do with it). There is a series up to 4 books, I've not seen the 4th one but the 3rd one is sdk level or thereabouts.

[emerus]

no, you should always read the variations. When people say do tsumego fast, they don't mean that one shouldn't read them, they mean that one should do easy ones for your level and try to read faster than usually.

In short, I believe nacroxnicke is correct.

This is what I understand from a teacher who is a professor of Go from Korea(taught professionals). He says don't do tsumego that you can't read out completely in 20 seconds. He also says to study the solutions to a few very difficult problems. Just the solutions, don't attempt the problem.

This is the way for the children in Korea but I'm 24 and he doesn't suggest an alternative method. Just to do as many problems as I can stomach in this way.

[paK0]

So, do lots of easy things seems to be the consensus.

how do you guys feel about all the explanations?

I'm working through Graded go problems vol 2 right now, which seems to fit, I can get almost everything right in less than 20 seconds, but I noticed that some of the hints are awfully specific, like: "Sacrifice three stones to capture the white group".

Does reading them detract from the learning experience?

[Solomon]

So, do lots of easy things seems to be the consensus.

how do you guys feel about all the explanations?

I'm working through Graded go problems vol 2 right now, which seems to fit, I can get almost everything right in less than 20 seconds, but I noticed that some of the hints are awfully specific, like: "Sacrifice three stones to capture the white group".

Does reading them detract from the learning experience?

I wouldn't say they detract from the learning experience, but if you get the gist of the solution and the relevant variations, then they are, at worst, mildly redundant.