Progress in Black and White

[Uberdude]

However, after 22 - when I tenuki to Q10 - the result of him cutting isn't good enough compensation no? At least CS didn't mind the result after Q10 - O2 - Q2 so the prior exchange worked out because he misplayed with N4 I guess. I feel like he then compounds his problems by his joseki choice in the upper right.

o2 q2 is still a nice exchange for black because it makes m4 sente which is very nice for a big thick wall (and some l2 area yose benefit). The l3 block was really big for the lower left moyo, so if you can't get m4 cut black is happy, I think even if you get sente for q10.

I blame my ever so slight headache but what is the big problem with S4? is it Q2 R2 S5 S3 Q5 (after which I probably need to go back to take N2, or exchange S6 then take?? Or just Q2 R2 O2 Q1 N2? For the latter, isn't me connecting out much larger than him taking the three stones?

The first, like you say connecting on the side is bigger. So the cut just creates (in sente) the option to take the 3 stones later (so you can count it as half of the points for if black does capture them) which black wouldn;t have if he simply blocked s4 at s5 (his tenuki was kinda crazy!).

[Majordomo]

http://lifein19x19.com/forum/download/file.php?id=8236

Woho! My first win in an even game against a 1d. I've added comments to the SGF - I think my opponent misplayed in the opening and then never complicated the game when he was behind - leaving me with enough points despite bad endgame.

Comments are appreciated!

[PlaySlow]

Congratz on your 1st win against 1d! Also I liked black's position in the early game very much. Your tenuki'ing in some critical moment gained you great momentum imo.

[Majordomo]

Ugh - so I'm bad at updating this thing - I'm also quite bad at playing Go recently. I still browse the forums / reddit / GoKibitz, but playing has been on a low ever since reaching 1d. Getting games became harder (not a lot, but just enough that I skim the list and then log out because I didn't get anything immediately), and playing on Tygem or FoxGo is a hassle on a mac - I'm tempted to buy a Windows Surface or somesuch just to have access to those servers because I feel that they would be the best places to play to continue improving.

That said, I'm continuing my lessons and I do a fair bit of Go problems so I hope I'm not deteriorating.

A month ago I had the pleasure to play on a real board against my 1~2 dan friend whom I played as my first ever game on a 19x19 board, and just like that game we played even. I lost (kind of big) - after a messy game where I think we both messed up a fair bit, but me much moreso than him. I've gone over the game quite a bit and was at least happy that I found a way to turn it around at one point and we entered the endgame somewhat even (his advantage) - but, the endgame is his favourite part of the game and my worst part, so the small lead became a large one. Anyway, I thought I'd post the game here - input is welcomed (I know I screwed up the upper left joseki, and for my knights move on the bottom I think he should make the empty triangle and cut no?):

Note - I'm black



Ideally I'd remember the endgame properly (the SGF is cut short) but it's a bit vague for me - but he won by 14 points on board + komi

[skydyr]

Here are a few things that occurred to me looking through the game. Grains of salt required of course.

[Majordomo]

Thanks a lot for your comments - they mirror a lot of what we thought when going after the game.

I was completely unfamiliar with the variation in the top left - but yeah, that hane is crucial, and like you pointed out I needed one more move to be honest when I chose the bad variation that I did.

For the bottom left, I'll be honest, I didn't read as far as the Ko until I'd already commited to it - I had a vague hope that it would just be dead, but when I started looking more closely at it I only found the ko, and like you said, I don't really have the threats for it yet.

103 was a chicken move - I need to keep white split here and at least keep us both running. With 107 I was playing desperate to catch up, but yeah, I don't think I have much of a shot here except that white messed up on top (I think he should hane on the inside and just take the points).

A major flaw playing on a real board I found was playing without a clock - so the "accuracy" of our play varied wildly because we'd be spending however much time felt polite just to keep the game progressing decently. Still it's something else to be able to go over the game afterwards on a board and just talk it through. Now I just need to break out of my lull and either get that tablet to play on Tygem etc or be willing to spend the time to find opponents on KGS / rank up on IGS.

[Majordomo]

Well that took a turn - some 8 months without Go, and what I think is a rather severe regression of what minor skills I had. Turns out kid nr 2 was not conductive to improving or even playing Go.

Anyway, I think I'm going to be able to keep playing with some regularity - my old account on KGS is gone due to inactivity so I've been playing around of a variety of accounts on IGS, KGS, Tygem and OGS - trying to keep some lingering rank / game anxiety to creep up on me again. I just have to accept that some games will be wrecked by some distraction here at home and try not to care too much.

Here's a pair of games against the new bot on the block LeelaZ, it's unranked but has beaten a couple of 1d and low kyu so I was kinda happy about beating it (almost twice! If I just didn't miss him playing atari I'd be way ahead in the lost game as well):



Will try to use this place to post my thoughts on the more interesting games I play, there's been a lot of 10s blitz (trying to get a hang of things again) but I think that's good - Focusing too much on rank and only playing if I could play "seriously" (as in feeling well rested and having no chance of distractions) kinda ruined it for me so yeah.

[EdLee]

Congrats on the baby.

Wonder if it's slow
( curious about Leela(8d) and Zen7 for all the moves... )

Wonder about K16 direction instead.

Wonder about M8/N8 direction instead.

Wonder about K12, or M4.

hmm... ( wonder about C17 ).

J10..

[Knotwilg]

Game 2

Summary

You have a good sense of shape, making important groups strong:

24, 26, 28, 32, 46, 48, 50

You also win a few small fights and calculate well:

14, 88, 140

But sometimes you play vague moves that have no clear intention or follow-up

34, 52

At 60 you uncharacteristically do NOT reinforce at R12 but go into Black's corner. This is the prelude to the main issue in this game.

(btw 72 confirms that you are afraid of too much black territory at the bottom and you start playing premature endgame)

Main issue:

74 is a slow move and in response 75 is a big move. Now your three stones have lost their power and value and the whole top is prone to attack.

So, when Black attacks small scale at 81, you should think about sacrificing the 3 junk stones.

The major (wrong) decision is made at 96: instead of throwing away the junk stones, you allow Black to capture the bigger part of your structure and make a big territory at the top.

Your defence skills come to the rescue: the combination of 124, 126 and 128 shows great skill

In the endgame you run away from black with great moves like 166 or 186

Full review

[Majordomo]

Ed & Knotwilg - thanks a lot for your comments & review! Despite having gotten around to play quite a few games I still feel a lot more uncertain as to where to play than I ever recall doing before - hence a lot of vague moves I think.


Also I've taken to using Crazy Sensei to help me analyse my more interesting games - so I decided I'd run it through there, only on the 4d level mind and I'm not quite sure if it's actually 4d since my CrazyStone app most certainly is not 4d on the 4d setting (KGS level anyway).

Anyway, here are the links to its evaluation of the two games:
https://www.crazy-sensei.com/?lang=en&l ... 32&move=31
https://www.crazy-sensei.com/?lang=en&l ... 33&move=75

[Majordomo]

So, I guess the bot sort of had a meltdown and I don't really feel like it counts, but beat a 1d KGS bot (0 komi though)! I killed it on the left side which happened after it played a very strange corner move (endgame?) instead of reinforcing itself and after that I played kind of vague and overly cautious but I felt the game was sort of over so I just had to not die anywhere. Often I find that the bots are good at sacrifice plays where they get sufficient outside strength to compensate for me focusing overly on killing a group but here it just sort of died and acted as if it was still alive so it never came back in the game after that.

I played F3 after seeing it recommended by AlphaGo Teaching tool in this position - but the bot replied a bit strange I think and made some mistakes in the subsequent sequence.

CS review:
https://www.crazy-sensei.com/?lang=en&l ... 843&move=2