Page 5 of 6
Re: Memorizing Tsumego via SRS
Posted: Sat Sep 15, 2012 4:55 pm
by ez4u
Bill Spight wrote:speedchase wrote:professionals all the time brag about how they can read 100 moves ahead. Sure they are exaggerating, but...
...
Go to move 147.

"*** At this point Tamura (later Shusai) thought for eight hours and read the game out to the final ko, 130 plays."
And if you are lucky enough that your opponent reads out the same 130 plays, you are good to go.

BTW, did Tamura himself make that claim or was this another nice story by a commentator?
Re: Memorizing Tsumego via SRS
Posted: Sat Sep 15, 2012 5:11 pm
by speedchase
Redundant wrote:The point still stands.
How? obviously you need to be able to prune. We aren't discussing that at all.
Edit: Bill, that is awesome
Re: Memorizing Tsumego via SRS
Posted: Sat Sep 15, 2012 5:41 pm
by Bill Spight
ez4u wrote:Bill Spight wrote:"*** At this point Tamura (later Shusai) thought for eight hours and read the game out to the final ko, 130 plays."
And if you are lucky enough that your opponent reads out the same 130 plays, you are good to go.

BTW, did Tamura himself make that claim or was this another nice story by a commentator?
I have seen the claim in a couple of different sources, but neither one written by Shusai himself.

Re: Memorizing Tsumego via SRS
Posted: Mon Sep 17, 2012 6:04 am
by Redundant
speedchase wrote:Redundant wrote:The point still stands.
How? obviously you need to be able to prune. We aren't discussing that at all.
Edit: Bill, that is awesome
Reaching a known position is a very efficient way to prune. Knowing common positions allows you to read deeper by reducing the branching factor.
Re: Memorizing Tsumego via SRS
Posted: Mon Sep 17, 2012 10:20 am
by speedchase
Redundant wrote:Reaching a known position is a very efficient way to prune. Knowing common positions allows you to read deeper by reducing the branching factor.
Obviously knowing the lines would be useful, but it isn't necessary, which you seem to be implying.
Re: Memorizing Tsumego via SRS
Posted: Mon Sep 17, 2012 3:12 pm
by Redundant
speedchase wrote:Redundant wrote:Reaching a known position is a very efficient way to prune. Knowing common positions allows you to read deeper by reducing the branching factor.
Obviously knowing the lines would be useful, but it isn't necessary, which you seem to be implying.
It's necessary to play optimally in timed games.
Re: Memorizing Tsumego via SRS
Posted: Mon Sep 17, 2012 4:32 pm
by speedchase
Redundant wrote:It's necessary to play optimally in timed games.
Noone can play optimally in timed games, regardless of whether or not they have memorized corner positions.
Re: Memorizing Tsumego via SRS
Posted: Tue Sep 18, 2012 7:22 am
by karaklis
speedchase wrote:Noone can play optimally in timed games, regardless of whether or not they have memorized corner positions.
Right, but the difference in strength could be some twenty stones or more.
Re: Memorizing Tsumego via SRS
Posted: Tue Sep 18, 2012 7:31 am
by Redundant
speedchase wrote:Redundant wrote:It's necessary to play optimally in timed games.
Noone can play optimally in timed games, regardless of whether or not they have memorized corner positions.
Of course I didn't mean optimal in an absolute sense,

Re: Memorizing Tsumego via SRS
Posted: Tue Sep 18, 2012 8:19 am
by speedchase
Redundant wrote:Of course I didn't mean optimal in an absolute sense,

How did you mean it then? It can't be subjective, because learning new things changes your strength.
Re: Memorizing Tsumego via SRS
Posted: Tue Sep 18, 2012 9:16 am
by hyperpape
Maybe we could communicate better by using more than one sentence.
Re: Memorizing Tsumego via SRS
Posted: Tue Sep 18, 2012 9:31 am
by speedchase
hyperpape wrote:Maybe we could communicate better by using more than one sentence.
What do you mean?

Re: Memorizing Tsumego via SRS
Posted: Tue Sep 18, 2012 3:01 pm
by Redundant
Everything of worth in this thread has already been said. Right now we're just poking at random words.
Playing optimally here was a bad turn of phrase, but having sequences/statuses memorized dramatically improves the ability to read, by letting you use your time on things that are actually important. I also don't see any restriction on limiting myself to dealing with timed games, because the time alotted in a game is already a bounded quantity, with the bound less than the time needed to read perfectly.
Re: Memorizing Tsumego via SRS
Posted: Fri Oct 26, 2012 7:59 pm
by NoSkill
I used lee Chang Jo books to open to a random problem then test first instinct against answer. At 6k it helped a lot after just one 1 hr session in a car. But I wonder about basic shapes like j and l... I need to study them. For spotting first move memorizing is good, but seeing possible eyes and realizing switching order of moves can work is learned diffferentlly
Re: Memorizing Tsumego via SRS
Posted: Fri Oct 26, 2012 8:11 pm
by speedchase
NoSkill wrote:I used lee Chang Jo books to open to a random problem then test first instinct against answer. At 6k it helped a lot after just one 1 hr session in a car. But I wonder about basic shapes like j and l... I need to study them. For spotting first move memorizing is good, but seeing possible eyes and realizing switching order of moves can work is learned diffferentlly
Yeah, that seems like a good way to train intuition. It really isn't the same thing as memorizing