Life In 19x19

A lot of Tesuji-problems might help you to spot a couple of ("good") moves to start with.

SoDesuNe wrote:A lot of Tesuji-problems might help you to spot a couple of ("good") moves to start with.

Yeah, I bought Get Strong at Tesuji precisely for this purpose.

Bah, internet go is so unsatisfying after the weekend.

Working through Attacking and Defending Moyos as a bit of fun (above my level but I love whole board problems) I think I've decided I much prefer books for problems than using the iPad so I don't think I'll be making sgf versions of the books I got. Oh and GoGoD arrived, which should be a lot of fun to play around with. I love messing with databases. Again probably above my level but eh, I'd be buying it at some point anyway so not a big deal.

I'm wondering if there's any improvement in my play both from the past month and on a real board after the weekend. I played a game against IgoWinHD on the iPad on my board giving its 12k engine 2 stones. Now I sincerely doubt it's actually 12k. I don't expect a proper review as this is a game versus an AI and a win aswell. But general thoughts on whether I'm improving would be most welcome. I haven't played IgoWinHD much at all. I beat it in an even game when user was set to 12k yesterday so I set it to 11k today for a laugh and played a very fast game against it trying to apply some of the concepts I'd been reading about over the past two days about reducing and building moyos. I'd like to know if I was reducing effectively in this game. Tactically I made a few blunders as always.



Edit: The ratings on this program are way off it's just funny. I just made the 9k engine resign after giving it 2 stones. There's no way in hell that I'm anywhere close to 7k. Double that number and you've something approaching my rank if still probably too strong.

It will gone away after u know ur mistakes.
u need to see the benefit from ur moves. If u read too much that can confuse urself. so, see the most advantage moves.

You're certainly improving. The question is, what is the numeraire by which to judge the rate of improvement? At your level I think it makes the most sense to judge your improvement by reviewing your own games and asking the following sorts of questions:
If you really care about having a definite quantitative rank, then don't ask us! Log on to KGS, turn on automatch, and ranked games until you collapse from exhaustion.

What is the difference between 14 kyu and 10 kyu? Not much.

Thanks jts. I'm not looking for an exact rank, just to know if I've been improving at all. I feel like I have but I don't know enough about the game to be sure about it. I'm more annoyed by IgowinHD being out by so much. When I set it to 11k I wanted to be beaten and learn something, not win by resignation.

Bill Spight wrote:What is the difference between 14 kyu and 10 kyu? Not much.

When I was looking over the game I decided I needed a much stronger setting on the engine to punish moves like G1 or else I'll keep making them when playing instinctively. I tried it set to 8k, still the same problem of the AI being overrun, so 5k next which will hopefully thrash me and give me something decent to practice against when I can't commit to KGS.

Current reading:

Theory: Tesuji by Davies

Problems:
Speed Baduk 7 (like a more structured Go Problems for Beginners Series)
Graded Go Problems for Beginners 3 Level 1 Problems
Get Strong at Tesuji 1 star problems
Attacking and Defending Moyos (Fascinating)


I keep 4+ Problem sets going at any time to give me variety and have something to suit my mood at any given time.

wah.. U're really lucky. I only play in ogs. but in this 1 month, I already raise 2 stone handicap. because, last month I play with my fren in some place, I still need 4 handy. but,yesterday I can win even with 3 handy. and in 2 handy I win a game and lost a game too.

Boidhre wrote: When I was looking over the game I decided I needed a much stronger setting on the engine to punish moves like G1 or else I'll keep making them when playing instinctively.

I suspect that the program replied to G1 with H1 because it has a heuristic that rewards plays near the opponent's last play, or makes it a preferred search region. That might make it play better against better opponents.

Bill Spight wrote:

Boidhre wrote: When I was looking over the game I decided I needed a much stronger setting on the engine to punish moves like G1 or else I'll keep making them when playing instinctively.

I suspect that the program replied to G1 with H1 because it has a heuristic that rewards plays near the opponent's last play, or makes it a preferred search region. That might make it play better against better opponents.

Hehe. It tenukis a lot at 9k. I think it's just tuned to be more "local" at 12k or something.

Example:

Myself versus hailthorn. Comments for both sides would be appreciated.

First ranked game in a month etc, I think I might be getting over my fear of KGS.



I took Bill's advice and played as if I was 12k.

Finally found a setting that gives me a decent game.



I need to do some serious work on my ability to disrupt territory I think. The 5k Engine seems about right though. It fought a lot better against me and punished me a few times by tenukiing when I made a bad move. I'll try it on 3k next and see if I can get myself roughed up so I can learn something.