Life In 19x19 http://www.lifein19x19.com/ |
|
Early attacking moves http://www.lifein19x19.com/viewtopic.php?f=11&t=13205 |
Page 1 of 1 |
Author: | Tapani [ Fri May 20, 2016 4:57 am ] |
Post subject: | Early attacking moves |
Hello all, started playing go very recently, and have pretty much just played against GoDroid Android app. My question is about early contact moves. Everywhere it is said how it is so bad to attack early on, but the app does it all the time (on the easy levels). And I don't understand the explanation about early attacks in SL. To recap, according to SL the following play by black is considered bad, because white can play at a, and strengthening his group. But, in my very newbie mind, the follow up could be black playing at a or b (or even c), starting to stake out the shaded territory. The explanation that white gains "strength" is just abstract and hence irrelevant. Why is this thinking wrong? It is all about making territory, after all? Note: I am not saying SL is wrong, just that explanation is on a level that might not be understood by the target audience ![]() Thank you in advance. |
Author: | hyperpape [ Fri May 20, 2016 6:03 am ] |
Post subject: | Re: Early attacking moves |
Let's compare the alternatives. The actual potential for territory in the upper left is not much different here. In Diagram 1, White can clamp at b later to reduce black after the contact move. So one point is that the contact move adds very little. But the bigger point is that in diagram 2, black can aim at pincering the white stone near a, and either reducing a white territory in the lower left, or making territory there. After the contact move, white is stronger, and this pincer is unlikely to work. If black pincers, white has forcing moves at c or d in diagram 2 to strengthen himself. Moreover, black's pincer would be weaker facing the two stones. Hopefully this is concrete enough, but it does take some experience. Strength can be an abstract concept. Side note: In Diagram 2 I kept the move on the fourth line. But black may locally want to play at c or d on the third line, which gives white less room to make a base. |
Author: | Tapani [ Fri May 20, 2016 6:30 am ] |
Post subject: | Re: Early attacking moves |
Thank you for the response. Somehow "clamping" (is that a common go term?) at b in your diagram 1 never occured to me! Will try that! Also, I think my confused post tried to ask two questions at once. (a) How to play as black instead? Both SL and the response above suggest better alternatives. the second question is rather (b) How to try to exploit an opponents contact move? Clamping is an option, but what if black plays like: Maybe an invasion some where between/behind blacks stones at some point.. but against the computers tactical tricks, such invasion in this position would probably be doomed for me. |
Author: | Uberdude [ Fri May 20, 2016 6:38 am ] |
Post subject: | Re: Early attacking moves |
Unless black has more stones around to make me legitimately scared of the cut, I would hane in response to the contact, then if black pulls back connect white's cut. |
Author: | hyperpape [ Fri May 20, 2016 7:03 am ] |
Post subject: | Re: Early attacking moves |
I actually like Uberdude's move more than the Iron Pillar, but if you do play that as White, you can hane at the head of two stones, and make thickness towards the center. Do you know about the 33 invasion under the stone in the corner? Since that's there, Black's potential territory may be smaller than it looks. |
Author: | Bill Spight [ Fri May 20, 2016 7:07 pm ] |
Post subject: | Re: Early attacking moves |
Tapani wrote: My question is about early contact moves. Everywhere it is said how it is so bad to attack early on, but the app does it all the time (on the easy levels). Well, I don't say that it is bad to attack early on. In fact, you need to learn how to attack, so why not learn early on. (That's what I did as a beginner, BTW. ![]() Quote: To recap, according to SL the following play by black is considered bad, because white can play at a, and strengthening his group. But, in my very newbie mind, the follow up could be black playing at a or b (or even c), starting to stake out the shaded territory. . . . First, the attachment, ![]() Consider the play in this diagram. ![]() ![]() ![]() ![]() ![]() ![]() ![]() Now let's compare that with a sequence without the ![]() ![]() ![]() ![]() ![]() ![]() ![]() Quote: The explanation that white gains "strength" is just abstract and hence irrelevant. Why is this thinking wrong? It is all about making territory, after all? Well, if you compare these two results in terms of potential territory, it certainly looks like the one where ![]() ![]() In these diagrams Black continues the attack against the White stones. You may get a sense that the attack is stronger in the first diagram, simply from the fact that in that case White has only two stones, versus three stones in the second diagram. Also, the larger potential White territory in the second diagram means that even if Black takes away some of it, White will typically retain more, which not only give territory, but also room to make eyes (life). Lastly, as Uberdude points out, the hane may be better for White than the stand. ![]() ---- And no, it is not all about making territory. One of the greats of the twentieth century, Takagawa Shukaku, said that go is about making territory, but it is very difficult to make territory. Territory arises mainly through skirmishes which result in strong enough walls to prevent invasions. My observation is that if you take care of the strength and weakness of stones, the territory tends to take care of itself. (Most people are more territory oriented than I am. ![]() In any event, strengthening the stones that you are attacking is usually a bad idea. Attaching to them tends to do that. Quote: Note: I am not saying SL is wrong, just that explanation is on a level that might not be understood by the target audience ![]() Well, as you see, I had to go to some length to provide a more complete explanation, and even so it may not be all that clear to a beginner, especially as some things need to be taken on faith. And the SL example is sketchy, leaving most of the board undefined. It is certainly possible to imagine surroundings in which the attachment is as good as a more distant approach, or even better. IMO, a more tactical example, for instance, one where the approach kills while the attachment does not, would have been better. Here is a slightly different example, which might be clearer. Which attack is better, the first one or the second one? |
Author: | EdLee [ Fri May 20, 2016 10:12 pm ] |
Post subject: | |
Hi Bill, |
Author: | Bill Spight [ Fri May 20, 2016 11:58 pm ] |
Post subject: | Re: |
EdLee wrote: Hi Bill, Yes, my last suggestion is questionable, as well. Especially as the invasion looks good. ![]() Now, it is true that needlessly strengthening the opponent's stones is a common mistake, and not just at the beginner level. But maybe the lesson of this discussion is that strategic examples are too abstract to introduce the topic, as Tapani points out, and that tactical examples are better. ---- In my case, since I started out as an attacking player, I soon came to appreciate the importance of the strength of stones. And I remember that time well enough to know that one of the options in my last example would have been anathema to me fairly early on. ![]() ![]() |
Author: | Bill Spight [ Sat May 21, 2016 1:27 am ] |
Post subject: | Re: Early attacking moves |
Tapani wrote: As you will soon discover, go has quite a lot of proverbs, and nearly all of them have exceptions. ![]() There is a proverb that applies here. Hane at the head of two stones. That is, play ![]() This is one of the most robust proverbs, among those that admit exceptions. ![]() ![]() Now it is true that the cut, ![]() ![]() But then ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() Black may not play ![]() ![]() ![]() |
Author: | Tapani [ Sat May 21, 2016 1:37 am ] |
Post subject: | Re: Early attacking moves |
Wow, I am delighted to see so many responses to my ponderings. @Uberdude, thank you, playing hane was my gut feeling too. But the result is black gaining the territory outlined. Maybe white has compensation, but I am not good enough to use/understand that. @hyperpaper, yes I am aware of the concept of corner invasion. But I feel that is a little beyound the point, the possibility for one is by accident. The position is the example SL uses to illustrate bad contact moves. @Bill, Quote: the potential territory related to strong stones are more certain than those related to weak stones This. I think this makes sense to me, and your example with the two alternatives in the corner illustrates it well. I prefer the second with stones on 24 and 44. With the contact move, and stones at 44 and 35 I am worried about white being able to invade the corner. There probably are bookloads of theory with both these positions, but I have not read any of them ![]() Thank you! While my original question is not completely answered, I am still gaining go understanding here. Which is of course more important. In general, I am amazed by how many has taken the time to help a clueless beginner here! |
Author: | Tapani [ Sat May 21, 2016 1:42 am ] |
Post subject: | Re: Early attacking moves |
@Bill, seems you got another post while I was responding. Quote: Territory arises through skirmishes, remember? Hmm. My experience is that unless I explicitely build territory the computer will kill my groups or somehow nestle his stones inside "my" area, and I end up with very little. Since I lose most of my games vs GoDroid (even with undo), my measure of progress is how badly I lose. If I build territory explicitely, I tend to lose with less margin. Maybe this is a drawback of playing the computer as a beginner? |
Author: | Bill Spight [ Sat May 21, 2016 2:51 am ] |
Post subject: | Re: Early attacking moves |
Tapani wrote: @Bill, seems you got another post while I was responding. Quote: Territory arises through skirmishes, remember? Hmm. My experience is that unless I explicitely build territory the computer will kill my groups or somehow nestle his stones inside "my" area, and I end up with very little. Since I lose most of my games vs GoDroid (even with undo), my measure of progress is how badly I lose. If I build territory explicitely, I tend to lose with less margin. Maybe this is a drawback of playing the computer as a beginner? Well, when I started out there were no computer programs to play against. On the 19x19 it was not unusual for me to lose three groups, despite receiving handicaps. ![]() ![]() ![]() I assume that GoDroid can give handicaps. (Free placement of the handicap stones is best, IMO.) Instead of worrying about how much you lose by, may I recommend adjusting the handicap? Try changing it after each game, taking one stone less if you win, one stone more if you lose. Or try adjusting the komi, adding 10 points if you lose, subtracting 10 points if you win. That way you will win around 50% of the time and will be facing appropriate challenges when you play. Good luck! ![]() |
Author: | EdLee [ Sat May 21, 2016 11:53 am ] |
Post subject: | |
Tapani wrote: My experience is that unless I explicitely build territory... Hi Tapani,...my measure of progress is how badly I lose. If I build territory explicitely, I tend to lose with less margin. The win/loss margin does not necessarily reflect what you think it means. Continue to play. After some time -- maybe after you have finished about 100 games -- have your serious games reviewed. |
Author: | Tapani [ Sun May 22, 2016 9:48 pm ] |
Post subject: | Re: Early attacking moves |
Thank you all for the repsonses. Did not expect any reply to my original post, possibly just some grumpy veteran dissing me as a newbie. The responses have been nice and helpful. Good to get happily surprised sometimes. Will play some more, and experiment a little with "influence" instead of making territory ![]() |
Author: | Charles Matthews [ Mon May 23, 2016 5:46 am ] |
Post subject: | Re: Early attacking moves |
Tapani wrote: Will play some more, and experiment a little with "influence" instead of making territory ![]() It is good to remember this: how do we gain advantages in go? Two kinds of ways are: outnumbering the opponent locally; and by better local arrangement of stones. In the first case, you have to try to get a local advantage, by attacking, because you are outnumbered elsewhere. The second case is in a sense more versatile. Definitely, though, starting a local fight where you are outnumbered is not a vanilla attack. |
Author: | Bill Spight [ Mon May 23, 2016 7:03 am ] |
Post subject: | Re: Early attacking moves |
Charles Matthews wrote: Definitely, though, starting a local fight where you are outnumbered is not a vanilla attack. Well worth repeating. ![]() |
Page 1 of 1 | All times are UTC - 8 hours [ DST ] |
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group http://www.phpbb.com/ |