Life In 19x19

This is just a side note, but I've been probing learning Chinese, and I've discovered a certain similarity between learning the characters and visualizing go positions.

When one writes Chinese characters, there is a specific order of strokes, and when I learn a new one, I've been trying to visualize it by drawing the strokes in my mind's eye. For relatively simple characters, such as: 人 or 四 it is no problem to "see" the completed figure. With a slightly more complicated character such as 我 or 他, I know what the character looks like - pretty much at least (I still don't know how relevant certain aspects of the lines are ), but I can't really see the whole thing. Maybe that's because I draw the lines of characters that I'm not sure about too slowly.

To me, a Chinese character is utterly abstract - though I have heard that there is sometimes some symbolism involved. But since I don't recognize any symbolism, a character is just a random collection of lines and shapes. Like I said, I can't yet judge what is relevant; is the length, angle or curve of a line important? I don't know, so I just try to copy the model as exactly as I can. In go, I do have a better idea what is relevant, but the visualization still seems quite similar because the arrangement of stones in a problem is also abstract, random and nonetheless unique and meaningful.

In any case, I think that my go-visualizing practice will prove beneficial to learning Chinese.

daal wrote:To me, a Chinese character is utterly abstract - though I have heard that there is sometimes some symbolism involved. But since I don't recognize any symbolism, a character is just a random collection of lines and shapes.

this might help: http://www.zein.se/patrick/chinen9p.html
(there are many other similar links with more or less examples and details)

As a teacher and also as someone that devotes many hours weekly to training, i have to say that 99% of the ideas you can get by looking alternate training methods, are not going to be fruitful.

As a previous poster wrote, i am not clear at what the difference between reading and visualization is for you. What is to read 5 moves ahead without visualizing it?

That said, solving problems without looking at them is a very standard training method in Asia. I found it difficult to do when i was 2-3d, but now when i solve problems i just glance at the position, and then close the book.
I have better performances looking at it, but id say that out of 100 problems i solve by looking at them, 90 i can solve in my head (although it always takes more time).

Back then i remember struggling with 9x9 blind go, but now i can play 19x19 for a 100 moves around (considering opponent doesnt do anything funky) quite ok. Then, performance and the ability to find good moves becomes too hard.

My hunch is that if you are reading without visualization, you are having a poor reading method. When you solve problems out of a book, problems that are not easy but take you a little while, what is your success rate? (how many do you get right/wrong when you finally see the answer to it?).

Something very demistifying about reading, is that intuition is purely the result of previously solved problems. The same way you can play a hoshi joseki today without hesitation and 50 years ago pro players spent hours to play worse sequences.
So you must cram into your head as many patterns as possible: and as solid as possible too. You most solve all your problem books 10 times.

You will notice that by the 4th time, you see the answer before you even see the problem. "i know this one, its there". Then by the 10th time you cans olve the problems with 100% accuracy,and greatest long-term effects.

Thats my 2cents.

Now support kaya.gs with your 2 cents .

xed_over wrote:

daal wrote:To me, a Chinese character is utterly abstract - though I have heard that there is sometimes some symbolism involved. But since I don't recognize any symbolism, a character is just a random collection of lines and shapes.

this might help: http://www.zein.se/patrick/chinen9p.html
(there are many other similar links with more or less examples and details)

That's a very handy site!

It may be my own learning style, but I think that practicing the basic strokes and the stroke order until they both seem obvious and natural will make the radicals, and then the characters, much less abstract and arbitrary, let's see how can I tie this back into go, just like learning tactics for contact fights and shape moves makes visualizing long joseki sequences easier.

Kaya.gs wrote:Back then i remember struggling with 9x9 blind go, but now i can play 19x19 for a 100 moves around (considering opponent doesnt do anything funky) quite ok. Then, performance and the ability to find good moves becomes too hard.

I've always thought that the hardest part about doing this would be figuring out where my opponent moved - using coordinates for Go is very hard. If we were sitting in front of a go board and would tap an intersection to indicate a play there, that would probably be much easier. It would also offer you the advantage of being able to project your visualization onto a real board. Though I'm getting better at it, I still struggle a bit with seeing the exact distance between two corners when I'm not looking at a board, especially as more stones get added.

Kaya.gs wrote:As a teacher and also as someone that devotes many hours weekly to training, i have to say that 99% of the ideas you can get by looking alternate training methods, are not going to be fruitful.

As a previous poster wrote, i am not clear at what the difference between reading and visualization is for you. What is to read 5 moves ahead without visualizing it?

The difference is that without visualizing, you might know the result, but you can't see it in your mind's eye. Imagine for example this:

Click Here To Show Diagram Code

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

you know what happens when white plays next, and it's also fairly easy to see the resulting shape.

Click Here To Show Diagram Code

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

When it gets just a bit more complicated though, you might still know what happens, but you might not be able to see it quite so clearly:

Click Here To Show Diagram Code

[go]$$
$$---------
$$|a X O . .
$$|X X O . .
$$|X O . O .
$$|O O . . .
$$|. . . . .[/go]

I know that that when white plays at a here, there will be a space left the size of four stones, but what does it look like?



While seeing is only one aspect of knowing what happens, I suspect it's an important one, and much of the rest of your post supports this opinion.

That said, solving problems without looking at them is a very standard training method in Asia. I found it difficult to do when i was 2-3d, but now when i solve problems i just glance at the position, and then close the book.
I have better performances looking at it, but id say that out of 100 problems i solve by looking at them, 90 i can solve in my head (although it always takes more time).

I had no idea that this is a standard method. (maybe my idea isn't so "alternate" after all). As I said a few times above, my idea is to visualize a problem accurately. That means to look at a position, and then with closed eyes be able to reconstruct the problem accurately in any orientation. Being able to do this is clearly a prerequisite to the training method you mention above.

Something very demistifying about reading, is that intuition is purely the result of previously solved problems. The same way you can play a hoshi joseki today without hesitation and 50 years ago pro players spent hours to play worse sequences.
So you must cram into your head as many patterns as possible: and as solid as possible too. You must solve all your problem books 10 times.

You will notice that by the 4th time, you see the answer before you even see the problem. "i know this one, its there". Then by the 10th time you cans solve the problems with 100% accuracy,and greatest long-term effects.

Basically, that's what I am doing. By spending such an inordinate amount of time with each problem I visualize, I am effectively cramming it into my head. Some people do a tsumego from a book and remember the solution a week later when they look at it again. I'm not one of those people. But the problems I've been visualizing stick with me quite a bit longer. Overlearning, as Bill Spight might say.

Thats my 2cents.

Now support kaya.gs with your 2 cents .

Thug: "This is a stickup! Your money or your life."
Thug: "Look, bud, I said 'Your money or your life.'"
Jack Benny: "I'm thinking it over!"

daal wrote:
The difference is that without visualizing, you might know the result, but you can't see it in your mind's eye. Imagine for example this:

$$
$$---------
$$. . . . .
$$. . O . .
$$. O X a .
$$. . O . .
$$. . . . .

Click Here To Show Diagram Code

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

you know what happens when white plays next, and it's also fairly easy to see the resulting shape.

$$
$$---------
$$. . . . .
$$. . O . .
$$. O . O .
$$. . O . .
$$. . . . .

Click Here To Show Diagram Code

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

When it gets just a bit more complicated though, you might still know what happens, but you might not be able to see it quite so clearly:

$$
$$---------
$$|a X O . .
$$|X X O . .
$$|X O . O .
$$|O O . . .
$$|. . . . .

Click Here To Show Diagram Code

[go]$$
$$---------
$$|a X O . .
$$|X X O . .
$$|X O . O .
$$|O O . . .
$$|. . . . .[/go]

I know that that when white plays at a here, there will be a space left the size of four stones, but what does it look like?

$$
$$---------
$$|O . O . .
$$|. . O . .
$$|. O . O .
$$|O O . . .
$$|. . . . .

Click Here To Show Diagram Code

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

While seeing is only one aspect of knowing what happens, I suspect it's an important one, and much of the rest of your post supports this opinion.

This confirms my hunch. Reading the result of the capture without knowing the resulting shape of the capture is plain incomplete reading.
Exemplar cases when you do this incomplete approach are with ishi-no-shita problems or big eyes. if you can't see what is the consequence of the move, then you are just not able to evaluate the result.

Sure, most of the time we use abstraction: most captures are best local tactics, and it doesnt matter what the result will be.


Using ur very same problem..

Click Here To Show Diagram Code

[go]$$
$$---------
$$|a X O . .
$$|X X O X .
$$|X O O X .
$$|O O X X .
$$|X X X . .[/go]

This article Perception in chess: Evidence from eye movements (and yes, it is mainly chess research, but don't panic/over react ) reports research findings that indicate it is not any fundamental lack of visualization "talent" but rather a lack of subject-related knowledge that is the problem. Specifically the researchers exposed highly skilled players and less skilled players to a variety of chess positions for five seconds and then tested their ability to correctly recall the positions. They found that the skilled players were far superior at recalling real positions arising in actual games, but were not better when the positions were created by randomly scattering the pieces across the chess board.

"... Chase and Simon’s... finding that a master was only superior in the 5-s recall task when structured positions, rather than randomized positions, were presented, challenged the view that chess masters are superior in terms of their cognitive apparatus or processes (e.g. hardware aspects of perception, attention or memory). Rather, Chase and Simon... postulated that knowledge of patterns specific to the domain of chess supported effective search for good moves. Soon, similar findings were reported for experts in other domains including bridge players (Charness 1979), music students (Beal 1985), electronics technicians (Egan and Schwartz 1979), and basketball players (Allard et al. 1980)..."

In other words it is in support of kaya.gs and others advising that we must pack our brains full of Go images by any means that works, but always fundamentally putting in the necessary time. It is against those of us who want to cry off from the hard work because we are "too old" or fundamentally we are "poor at visualization" , etc. In fact we are merely shown to be too lazy once again!

kaya.gs wrote:This confirms my hunch. Reading the result of the capture without knowing the resulting shape of the capture is plain incomplete reading.
Exemplar cases when you do this incomplete approach are with ishi-no-shita problems or big eyes. if you can't see what is the consequence of the move, then you are just not able to evaluate the result.

Sure, most of the time we use abstraction: most captures are best local tactics, and it doesnt matter what the result will be.


Using ur very same problem..

$$
$$---------
$$|a X O . .
$$|X X O X .
$$|X O O X .
$$|O O X X .
$$|X X X . .

Click Here To Show Diagram Code

[go]$$
$$---------
$$|a X O . .
$$|X X O X .
$$|X O O X .
$$|O O X X .
$$|X X X . .[/go]

$$
$$---------
$$|O . O . .
$$|X . O X .
$$|. O O X .
$$|O O X X .
$$|X X X . .

Click Here To Show Diagram Code

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

The position is now ko. Not seeing what the resulting shape is shows that the reading conclusion was incorrect.

You make an excellent point here, but recall: I am bad at reading, and yes, it is bad practice to stop reading at the point of a capture - one would never see an under the stones tesuji for example. To get to the correct solution of a problem however, one must first be able to get to the position before the correct solution. You have to walk before you can run. If I can't envision this:

Click Here To Show Diagram Code

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

then I won't be able to envision this:

Click Here To Show Diagram Code

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

To me, the first diagram is simply a step in the right direction.

I would like to add that I am very appreciative of the constructive criticism. Particularly the recent comments of snorri and ez4u also suggest that visualizing takes a back seat to perceiving relevant information. Nonetheless, although I did not do it in the above example (it was off the cuff, not something I had spent time visualizing) the practice of visualizing does indeed provoke me to spend a significant time thinking about the relevant information. In the cases where I visualize the solution as well as the problem (something I am doing more and more) I also spend an equal amount of time visualizing and considering the failures. Again, many of you have better study habits than me and always call into question presented solutions until you are thoroughly satisfied. I on the other hand tend to say "oh!" and move on to the next problem. With my current method however, I'm allowing myself the time to think about just the one problem.


To illustrate, a recent example from Making Good Shape (reconstructed from memory) shows the following failure:

Click Here To Show Diagram Code

[go]$$ Black must fight a ko to live
$$---------
$$|. . 1 . 6 . .
$$|5 X . . . O .
$$|3 2 X X X O .
$$|4 . O O O O .
$$|. O . . . . .
$$|. . O . . . .[/go]

was indeed my first idea, so instead of saying to myself "I blew that again" I spent quite a bit of time thinking about why it didn't work, why black answers with and not at , what does white do after retaking the ko. I finally see this:

Click Here To Show Diagram Code

[go]$$ oops
$$---------
$$|. . X 1 O . .
$$|X X . 2 a O .
$$|X O X X X O .
$$|O . O O O O .
$$|. O . . . . .
$$|. . O . . . .[/go]

Black can't play at "a" because of a shortage of liberties. Something I wasn't expecting, so I think about why I should have expected it, etc. I'm not good, but I am pretty sure that I am getting better.

BTW, if you also find it difficult to visualize the original problem with all the solution stones muddying it up, here it is

Happen to come across this post when I was finding any discussions on how people / pro / strong amateurs get good and quick reading skills.

Very much appreciate this!

I've been on some kind of experiment lately, I'm not too sure whether doing this is right or wrong so I'm also seeking some validation. As as my progress with tesuji's and L&D, I've no problems solving easy ones and reading them out, provided I've seen the same pattern.

However, when comes to reading for unfamiliar patterns, I tend to stuck in the same wrong solution for a very long time. Sometimes I get to move on, but sometimes I don't. I tried to imagine stones being put on the problem pattern, be it on a book or on screen, but I find that I quickly forget the lines that didn't work out.

Then I figured if I just continue doing L&D problems like this, I'm only learning from brute force memory and not really "read" them out. Thus I came up with a practice I did when I was practicing for Chinese Chess. I looked at the board, and I visualize the board in my head. It seems I can better grasp on what's happening with the use of "mind's eye". But I need to do it with my eyes closed. It'll be awkward if I always do this is real games.

So this gets me wondering, how do Pros / strong amateurs read the sequences? Do they really look at the board and visualize the moves? Or they are using "mind's eye" as well, despite having their eyes looking at the board but their thoughts already wandered elsewhere?

Just fyi, I can solve problems faster using this "mind's eye" visualization and frankly speaking it was very satisfying because the focusing power that I can feel.

Hi zafuri95,

If you close your eyes, can you visualize an apple^(*) ?
With your eyes open, can you imagine an apple^(*) ?
An apple sits in front of you. With you eyes open, can you imagine a second apple^(*) ?
That's it.