Life In 19x19

Try memorizing pro games then replaying them in your head.

I agree with kirkmc, here. Reading practice is a way to help your visualization.

I think that your idea of practicing visualization in isolation is a good one, probably. It has the benefit of focusing on this very important part of reading.

The good part of go problems, though, is that (provided the problem is constructed well) you can have feedback on whether your visualization was correct (i.e. if you solved the problem).

I have recently had some similar thoughts.

One thing that I noticed long ago is how some players will make a play that they learned from some problem, but it does not work because there is a slight difference in the position. It made me think that you do not really understand a problem unless you can set it up (or at least set up an equivalent position) from an empty board. Otherwise you may think that a similar, but essentially different position is equivalent.

Bill Spight wrote:It made me think that you do not really understand a problem unless you can set it up [...] from an empty board.

This seems important to me and probably that is what Cho Chikuns understands about studying L&D. That's why I have started to analyze easy problems to the elementary techniques (reduce/expand eye space, nakade, vital points, throw-ins, damezumari etc.) until I can lay out the problem myself.

I so much understand your point of view, Daniel !
I feel exactly the same: I'm currently going through 501 Tesuji (for the third time) and if I start to get a decent feel for where the first move is to be played, I'm not able to read through the 10 moves some solutions are requiring to go through. No matter how long I try, I can't get the confidence that i'll get ahead from a capturing race, barely a positive feel but it's one thing to try it in a problem and to dare to play it in a game,
I don't what to do to get deeper reading skill...

I have the same problem with visualizing, and my visual memory seems quite weak. After a while (about 3-7 stones depending on the problem) the virtual stones blurr and I have to start over. That's why I often fail at problems where shortage of liberties is involved.

To overcome this I have started to anatomize tsumego into their components, i.e. the basic techniques for L&D and tesuji, such as expand/reduce eyespace, nakade shapes, throw-ins, shortage of liberties, snapbacks etc. I will start with very easy tsumego until I know every variation so well that I can create this tsumego on my own board without looking in the book. Then I will write down this tsumego in my own documentation and every now and then I will go through those which I have already finished. I know from this only verrrrrrrrrry slow progress is possible, but better slow than nothing

When I visualize the game, I don't so much use my imagination as I do simply look at where I want the stone to go. The motion of my eyes helps to establish the position. If I forget a stone, I reestablish the location by looking back at that point.

I think this is a great idea daal. Sometimes challenging yourself by making something more and more difficult breaks down a wall and makes the original activity seem easier when you go back to it. In any case, it can't hurt to try something different. I'd be interested to hear how it goes after several months.

You remind me of a time about five years ago, when I was a poor student... (that's right kids! here comes a boring story from old man David )

I had a life and death book that I used to read on the long walk home from my crap supermarket job. Because the street lights were quite far apart, I could only glance at the problems when I walked under one. I spent the rest of the time solving them in my head. I honestly can't say how much this helped as opposed to doing problems 'normally', but I think the challenge was helpful (and it seemed like a good way to use the time).

And yes, I acknowledge that I was a fair bit younger than you when I started doing that. That's something I'm genuinely grateful for.

David

malweth wrote:Try memorizing pro games then replaying them in your head.

I've done this for a month or so, and I believe even discussed it with Daal at the time. My observation was that memorizing pro games improved my ability to memorize pro games. It did, however, not have a noticeable effect on my ability to visualize variations in tsumego or "see" stones.

I did not have instructions on how to memorize games with a specific method, so the "natural" way it occurred was that once I had "brute force"-memorized the first four moves, I recalled the following moves in form of sequences and pattern, but I didn't actually really have a clear image of the position in my mind. Differently put, I memorized a keima, and then the joseki that followed, or attached a made-up reason why the pro played a move that didn't make obvious sense to me. It went more like: "then he played here, and here, then there, and the response was over there".

I could replay the games on the board and in my mind in the correct order, but I wasn't able to put together the board position in a random order or have a clear image of it in my mind. If I missed one move when replaying it physically or mentally, it all fell apart.

I also noticed that I apparently can't "hold" more than four games at the same time. Four worked flawlessly, I could recite them at any given time. But as soon as I memorized a fifth, another whole game would get "pushed out". Actually, it impacted all the other memorized games, as if the mental bin had been emptied to make room for new stuff. (I know this isn't how it works, but that is what I experienced. :p)

It did have a positive effect on memorizing other information, though. Short memory stuff, unrelated to Go.

dfan wrote:Yeah, this is really important. In games like Go and chess, there are three equally important skills involved in calculation:

1) "Visualizing" a future position accurately
2) Choosing plausible moves (both now and while exploring the tree) that are likely to create good results (pruning the tree)
3) Evaluating the position at the end of a line strategically

#2, as you say, is key for solving problems in any reasonable amount of time.

Watching Malkovich games is a good way to observe #3; a lot of times you'll see players write out a variation they're excited about, followed by observers five stones stronger commenting "Why does he want to do this? The result is terrible."

I think people often take #3 for granted, I see so many game reviews where (especially in the opening/early middlegame) the reviewing player shows an improved line of play, and the person getting the review says something to the effect of "I saw that, but is that really good for B?" only to get the response "B should be very happy here!" Of course it also helps that #3 is vital in properly implementing #2 (How can you know when to prune a variation when you can't give the right evaluation?).


As far as visualization goes, I too am one who isn't the best at holding a mental picture....but there is one thing that I feel really helped me that was actually something that had nothing to do with go. Back when I was in school I made part of my morning routine working the daily crossword in the paper. Now personally I much, much prefer pen to pencil, but for working a crossword this may not always be the best approach. In order to avoid filling in a word I would regret later I would always try to solve entire blocks of the crossword at a time (3-4 words down and 3-4 words intersecting across). This way once I finally filled in the section I could be fairly confident I wasn't making an error. Eventually I found I could complete most or sometimes even all of the puzzle without even filling in a letter (Note: these weren't particularly large, or particularly difficult crossword puzzles). The thing here is, instead of having to hold a mental picture of the whole puzzle (something I often find difficult), I only had to retain a list of words (clue answers) then whenever I was looking at the appropriate section section of the puzzle mentally reconstruct the area. Basically it converts a memorization/visualization exercise into a derivation.

I think I now end up doing the same thing in go....Instead of remembering what the whole board will look like in a sequence I'm reading, I keep a mental description of exchanges (push-block-cut....press-extend-jump...etc) and when I'm looking at an area of the board apply those sequences in my brain. I think it also helps in the sense that often order of sequences matter, it's easy to reorder them if you are continuously mentally "plopping them down" on the board (as opposed to retaining a static picture of your current sequence).

Bill Spight wrote:I have recently had some similar thoughts.

One thing that I noticed long ago is how some players will make a play that they learned from some problem, but it does not work because there is a slight difference in the position. It made me think that you do not really understand a problem unless you can set it up (or at least set up an equivalent position) from an empty board. Otherwise you may think that a similar, but essentially different position is equivalent.

If Mark Twain played go -- The difference between an equivalent position and a nearly equivalent position is the difference between lightning and a lightning bug (=

Just a little update.

It's been about three weeks, and still this is pretty much the only thing I've been doing as far as studying. Basically I've been going through my books and when I find a tsumego or tesuji that doesn't have too many stones and seems like it might be representative or otherwise worth remembering I practice visualizing it in various corners and orientations. Also, following Kirk's suggestion, I've been including josekis.

At first glance, this visualization as I call it might seem like memorization, and although this is still not my emphasis, indeed I am in fact memorizing. This is particularly the case with josekis, where this study form is much more like the (bad) memorizing and less like the (good) understanding the meaning of each move. However, one advantage of this is that I am learning to visualize the outcome of a joseki at it's onset. Again, for many of you this goes without saying, but for me it is almost revolutionary. Despite the difficulty of assessing the result (see dfan's #3 and Mef's comment above), it makes a huge difference to be able to see what I am trying assess.

Progress tends to be slow. I'm basically down to one or two positions a day. Although the problems are generally quite easy - even for me - I spend significantly more time with each problem than I ever did with other go problems, and I am starting to see improvement in my ability to visualize stones. This is hard to describe, but here is an example:

Click Here To Show Diagram Code

[go]$$
$$---------
$$. . . . .
$$. O O . .
$$. O . . .
$$. O . . .[/go]

Instead of visualizing this as an (upside-down) "L," I am starting to see four white round stones when I imagine it. In other words, my reliance on abstractions is diminishing while my ability to see what is actually going on is improving. I hope.

Also there are a few side effect that I've been noticing. First, I've been playing slower games, and I find myself more willing to take the time to try to find a good move. I think that this is related to my studies, because I have gotten used to not rushing. Also, I haven't had any particularly dull or uninteresting games in a while. Again, I attribute this to spending more time looking instead of following my basic instinct (play the most obvious move possible). Lastly, losing is no longer getting under my skin. I'm not really sure what brought about this change, but I think that it might have to do with shifting to a long-term perspective in which being currently stuck at a certain rank is less of an issue.

daal wrote:Lastly, losing is no longer getting under my skin. I'm not really sure what brought about this change, but I think that it might have to do with shifting to a long-term perspective in which being currently stuck at a certain rank is less of an issue.

This happens to me as well (especially in Go). I'm a very competitive person in most avenues of my life and I find it difficult to do something just for the joy of it. Go has been an exception to that though. Playing Go to me is much more like when I'm playing music. There's something about the flow of the stones and pace of the game that I find incredibly beautiful.

When an opponent gets the better of me, I find myself enjoying his moves.

This could also be just because I'm a beginner so I know I need to lose a lot and knowing that Go is not so much about winning and losing.

It's a very interesting thread.

Some research has been done on this in chess, but it's mostly subjective and based on interviewing grandmasters on what they see in the mind's eye when they are playing. What was interesting to me when I first read about it that is that most players don't report high-fidelity images. What they are seeing---if they see anything at all---is not vivid images of castle turrets and horse's manes, etc. They are not even keeping track of individual pieces in a clear way! It's much more abstract than that, but keeps track of the essential strengths and weaknesses in a position. If pressed they can do such exercises such as a knight's tour around the board, but this is not the typical skill being used. Most of the strong chess players rarely conjure an image of the board.

What is probably essential is to get the correct answer for a particular goal: "a few moves in, in this variation, am I short of liberties?" Or, "if it goes like this, can white make an eye here?", etc. If you can get to that point in any way, a lucid representation may not be required. Intuitively, it might seem like you would need such an image to answer these questions, but maybe the brain doesn't work that way and it may be less important than one might think.

When I play competitive strategy games (doesn't really matter what, can be Magic, can be Yomi, and be LOTR:Confrontation, can be just about any RTS under the sun), I usually just end up feeling a pressure of sorts. Do I need to be defensive here, or am I allowed to attack? If so, in what way? Precise visualizations don't enter it that often, unless I really need to count some tricky sequence out to see if some greedy thing works. Otherwise, it's all intuition and feeling the game flow. (Which my games are admittedly more heavy on compared to Go's determinism and perfect information)