Life In 19x19
http://www.lifein19x19.com/

asking for a review of a game against a computer opponent
http://www.lifein19x19.com/viewtopic.php?f=12&t=12467		 Page 1 of 1
Author:  MinjaeKim [ Thu Nov 26, 2015 11:54 am ]
Post subject:  asking for a review of a game against a computer opponent
I'm recently stuck with a strong computer opponent called zenith go. I haven't played serious games for a while, but this software makes me think really really hard because if not I can't win. The problem is that unlike most of the games I play with human opponents, it is really difficult for me to find what exactly was the mistake that made me lose. Early thanks for any advice.
Author:  Joaz Banbeck [ Thu Nov 26, 2015 1:21 pm ]
Post subject:  Re: asking for a review of a game against a computer opponen
4: You are taking a side when there is an unoccupied corner. The corner is worth more.

14: This is inconsistent. If you are going to pincer him, you must continue to cramp him. F4 is joseki.

32: This is rather close to a strong position. If you must invade, something around L16 looks better.
But are you sure that you want an invasion rather than a reduction?
You might conside a reduction around H15 or K15 or N14 or M15.

44: R6 is big.
Author:  EdLee [ Thu Nov 26, 2015 1:37 pm ]
Post subject: 
Quote:
14: This is inconsistent. If you are going to pincer him, you must continue to cramp him. F4 is joseki.
Not exactly. A quick search on a particular database shows 87% for F4, but still 9% for F3 (73 pro games).
Go has become much more fluid over the past decades.
The game is not decided here, anyway.
Author:  MinjaeKim [ Thu Nov 26, 2015 1:48 pm ]
Post subject:  Re: asking for a review of a game against a computer opponen
Joaz Banbeck wrote:
4: You are taking a side when there is an unoccupied corner. The corner is worth more.

What is your justification for this statement? I don't think the opening was bad for me after all.

Joaz Banbeck wrote:
14: This is inconsistent. If you are going to pincer him, you must continue to cramp him. F4 is joseki.

Both F4 and F3 are joseki, just a matter of choice.

Joaz Banbeck wrote:
32: This is rather close to a strong position. If you must invade, something around L16 looks better.
But are you sure that you want an invasion rather than a reduction?
You might conside a reduction around H15 or K15 or N14 or M15.

Well, this is the hard part. Maybe it was a mistake, but I played this move after a lot of reading for other alternatives. Can you show me some possible sequences that can occur after other reduction moves? By the way, black did not cap move 32 that you named as 'invasion', and my thought was that this is a reduction move since trying to cap in this stone will be a better result for me.

Joaz Banbeck wrote:
44: R6 is big.

Yes indeed. If black secures the corner after white R6, then that will be better for white than the actual game. But what if black plays a move in the center, and when I play somewhere again in lower right, black settles the lower right corner in sente and play another move in center again? I can get some points here but now my center stones are very thin and under attack. Since black is strong on the left side and possibly around the lower right corner too, he is very likely to profit a lot of points from this attack. This is why I played the one space jump instead of R6.

Also I could squeeze black's upper right corner because my center stones were strong enough after the another one space jump.
Author:  Charles Matthews [ Thu Nov 26, 2015 3:05 pm ]
Post subject:  Re: asking for a review of a game against a computer opponen
MinjaeKim wrote:
I don't think the opening was bad for me after all.


By :w42:, I feel White has clearly less potential for development. When did that happen? In the corners and on the sides.

Lower left: the result of the pincer is not impressive. White should be trying to control the big point on the lower side; so that a one-point jump reply to the approach, at F4, looks best.

Upper side: seems to me a wedge was better than approaching the top left.

These may appear to be nuances, but the knack of having your stones falling into place is just a bit subtler than at first glance. For one thing, you can't simply negate the opponent's ideas.
Author:  Bill Spight [ Thu Nov 26, 2015 3:25 pm ]
Post subject:  Re: asking for a review of a game against a computer opponen
A strategic win for Black.

A few comments. :)



Edit: Added a variation. :)

And Charles's point is good. White did not play with an eye to development.
Author:  gamesorry [ Thu Nov 26, 2015 3:29 pm ]
Post subject:  Re: asking for a review of a game against a computer opponen
Probably slow moves:

44: I'll play somewhere in the lower right corner, e.g., R6 or R4.

MinjaeKim wrote:
But what if black plays a move in the center


You can respond in the center first then. Black needs two moves to attack white. Lower right not only contributes to the territory but also affects the thickness of white outside/black corner.

58: Although white tries to seal the black, but if the result is gote for white, it's too small. R3 could still possibly make a live in the black's corner.

66: S7 is probably bigger (and might be sente).

78: J11/M7 might be better. Don't be afraid of black's cut.

108: J7 (one step further) might be better for white.
Author:  Joaz Banbeck [ Thu Nov 26, 2015 10:22 pm ]
Post subject:  Re: asking for a review of a game against a computer opponen
MinjaeKim wrote:
...
Joaz Banbeck wrote:
14: This is inconsistent. If you are going to pincer him, you must continue to cramp him. F4 is joseki.

Both F4 and F3 are joseki, just a matter of choice...


Every move is a matter of choice. :roll:

But we strive to make meaningful choices: to make moves that have a meaning in the position in which they are played, that work well with the other stones already on the board.

As Ed says, 9% of the time pros will play your move. But they do not play it in a vacuum. Look at http://dailyjoseki.com/browse. ( That is probably where Ed got the stats )
Look at the other stones that appear as shaded points. If you walk through the joseki looking at those points, you will see that the 9% are usually - maybe only - played when there is a strong group to back up the pincer stone.

You didn't have that here. If you had a strong presence in the upper left, maybe F3 would make sense. But you didn't have it. Indeed, the only stone up there belonged to your opponent.

The point of F3 in this joseki is to play the safest move in the corner so that it can be left alone while the more valuable side is emphasized. But it only means something if the side is more valuable.
Author:  Joaz Banbeck [ Thu Nov 26, 2015 10:30 pm ]
Post subject:  Re: asking for a review of a game against a computer opponen
MinjaeKim wrote:
Joaz Banbeck wrote:
4: You are taking a side when there is an unoccupied corner. The corner is worth more.

What is your justification for this statement? ...

http://senseis.xmp.net/?FirstCornersThenSidesThenCenter
Author:  Uberdude [ Thu Nov 26, 2015 11:42 pm ]
Post subject:  Re: asking for a review of a game against a computer opponen
Also note that your result on the right side normally occurs with a black stone at r7, in which case it is a good, strong and efficient shape (though the point about the empty corner being bigger still stands). But in your game black doesn't have a stone at r7 so your shape is inefficient as you are defending against a nonexistent threat, and black is more efficient as he didn't waste a stone at r7 which in the standard shape ends up rather too close to strength. Black now has the choice to develop that corner in the other direction with n3 which he gets later, and even later plays r6 to keep away from your strength and be stronger on tbe corner. You also failed to use the strength of your group by approaching black's corner at o3.
Author:  zac [ Fri Nov 27, 2015 2:11 am ]
Post subject:  Re: asking for a review of a game against a computer opponen


In one of Guo Juan's lectures she is explaining in the simplest terms why we would want to aim for the centre when invading. This is one of the examples she gives, with the explanation that white could not be happy with this result.
I wonder what result you were expecting from playing one line off-centre?
Author:  hyperpape [ Fri Nov 27, 2015 11:31 pm ]
Post subject:  Re: asking for a review of a game against a computer opponen
Third move split: viewtopic.php?f=15&t=10867
Page 1 of 1 All times are UTC - 8 hours [ DST ]
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group
http://www.phpbb.com/