Sunday, February 28, 2010

Chronos GX Touch

This is basically another tournament followup post.

In addition to recording the games on my netbook, I also used my Chronos GX Touch clock for a timer. It's pricey, but if you want the best chess/go clock, this seems to be it. Also, it's solid enough that, if in a pinch, you needed to kill someone with it, you could do so. And once you pick the brain, blood, skull, and hair out of the creases, it'd be good for another tournament.

For the first game, I used neither netbook or clock -- just the NOVA club's wind up chess clock (no byoyomi, you have to do it by hand -- I believe this is why they used Canadian byoyomi).

I must say, I love the Chronos. First time using it in a tournament setting. I had forgotten how to set it, and honestly the configuration of settings is pretty... confusing... unless you do it a lot.

I never hit byoyomi in the tournament (45 minutes per side is a lot... I got close a couple of times though). If I had, I'd have really loved the clock. Otherwise, we'd have to reset the analog clocks, and keep track of number of moves played (5 minutes per 20 stones (15 secs per stone average)). That sounds like a pain in the ass, and highly error prone. I couldn't imagine trying to determine a game like that on just time....

Anyway, even without the tremendous byoyomi pains-saving, it's still pretty nice. The analog clocks have squishy, unreliable metal pegs to press. The Chronos has nice touch sensors -- all it takes is a light brush of any finger.

I find it to be... a much more peaceful way of ending your move.

This concept (peaceful playing) will be followed up in the next post or two...

Saturday, February 27, 2010

IRL Tournament: NOVA's 4th Chinese New Year Tournament


In-Real-Life no less (I'm glad I'm getting a chance to use my IRL subscription, it's expensive).

Like the NOVA Winter Warmer, it was fun but grueling. Left the house at 7am, got back at 9pm.

Also like the Winter Warmer, I won only one of four, although this time I bumped my entry rank to 12kyu from 13kyu. Also, every competitor was close rank, no more than one rank difference.

But this time I recorded the games! (well, at least 3 of them).

I used my netbook and SmartGo (and a bluetooth mouse) to record the games. I also had printed kifu (from the earlier post) to use, just in case.

Table space was a problem, but the netbook's footprint is smaller than the 8.5x11" kifu, and a hell of a lot easier to use. Besides, SmartGo is the destination for the data anyway.

The cool part is, at the end of the game, instead of counting on the board, I can just press F11!

This (and the record) was crucial in the 3rd match, where I lost by one half a point.

Actually, in all three losses I had a solid lead at one point in the middle, made a stupid mistake (or 2... or 3...) and lost it all.

It's pretty consistent in live games with long times. At about an hour I get mentally exhausted and make moves without reading them fully. It effectively puts me 2-3 stones below how I play the first part of the game. Not sure how to counter it...

This tournament was different in that there were some sort of sponsors. This meant there was loot to be had. One of the people guilty of getting me into/back into go (dhogarty on DGS) won 4/4 and his rank band, and made off with all sorts of loot. He also entered as 9kyu, which quite officially makes him a solid SDK (single digit kyu).

Oh, and in the 4th round, I used the great wall opening (?!). This was foolish, as I am quite weaker at this opening than my normal, but I was so distracted by the fact that I could (got black and an even game in a handicap tournament) that I couldn't resist.

How long until I regretted it? After laying the first stone down (3-10) I looked at it, and regretted it.

To be fair, I was possibly leading in the middle...maybe. I killed no groups, which, with the great wall, is usually a sign of failure.

Sensei's Library: Killing Shapes

In case you haven't figured it out, the blog is named after the shapes that are used to kill groups that have only one eye. 

As usual, the best reference is Sensei's Library.

In fact, I ripped these images from there:
If you can almost fill the opponent group's single eye, with a killing shape, then it is dead.

The key here is almost fill. This means with one empty point still inside the eye, not two or more. This means that once the shape's outer liberties are filled, it will be forced to capture your killing shape. After it does this, it's your turn, and you can play on the vital point shown in the pictures.

In practice, the sequence of events is not played through, because both players can see this coming.

Additionally, there are other circumstances -- for instance, a bent four in the corner is also a killing shape, but only in the corner (I lost a game at tournament today because of this!)

So, know your killing shapes.

Friday, February 26, 2010

Printable Kifu

I had a difficult time finding these for some reason, but I finally found them the night before the NOVA Chinese New Year Go Tournament.

But, I found them at this location, and have them here for your convenience: printable kifu!

    I will attempt to record my games at the tournament, either on paper or using my netbook and SmartGo (if opponent allows).

    Maybe this time I'll actually get my AGA rating...

    I will submit as an AGA 12 kyu (I did 13 kyu last time and was 1/4 game).

    Get Into Shape: Ponnuki

    You have to love the ponnuki, it's the epitome of 'good shape' in go, almost (but not quite) to the degree that the empty triangle is bad shape.

    Why is it good shape?

    Eye space.

    Eye space and efficiency.

    Eye Space
    First, let's look at the marvelous example of the ponnuki in the corner, the coolest living shape in go (IMHO).

    Barring a double play (from ko threat or sneaky-playing-while-you're-in-the-bathroom), this corner shape is alive. Obviously if white drops approach extensions down to the edge you must play to prevent being killed, but basically you're fine.

    White stones are here just to show how pinned in you can be and still be a-ok.

    Next, one of the most common and powerful locations for a ponnuki, the edge. Although not alive by itself, it is a complete eye. The lower gaps are safe because of the edge (opponent must approach). The upper edges aren't, but they are effectively miai -- you can lose one as long as you have the other.

    I was going to do a separate eye post (as a Beginner Tactics), but haven't done so yet. It helps to remember this basic definition though:

    The chain of stones surrounding eyespace must have no more than one enemy stone occupying it.

    Bearing that in mind, you can see that the ponnuki on the edge is ok as long as once one of the upper corners is taken by an opposing stone, you fill in the other (or verify it is cut-safe). Obviously we assume you will block peeps at the edge.

    Even ponnukis in the middle of the board are strong, but far from guaranteed eye space.

    So, in theory, these are not yet solid eyes because maybe the opponent could cut into more than one corner. Despite this it's still a rockin start at getting an eye, and will almost certainly end up being one.

    These are completed eyes based (again, barring double plays) on their being only two points left which are miai. But, at this point, it should all be fairly obvious, yes?

    The best thing about the ponnuki is that it is not just powerful, but cheap
    At most it's 4 stones, but really, since a ponnuki often forms after capturing a stone, it's often effectively only 3 stones in cost.

    Don't Send Ponnuki-shaped Valentines Just Yet

    Of course, ponnukis have a seamier side as well. They often form in and around ko fights. Often a hanging connection is formed in the corner above the edge just to pick a fight. Can you believe that?

    Tuesday, February 23, 2010

    KillingShapes now on Twitter!

    Now the blog can be ignored in TWO different media outlets!


    Monday, February 22, 2010

    NOVA Chinese New Year Tournament coming up, Feb 27th

    Pre-Registration Required!!

    If you are in the DC metro area, I'd recommend going to this, but you'll have to send an email to the contact at the link in order to do so.

    Sunday, February 21, 2010

    DGS and Fischer Time

    Just a quick post, thought I'd share a post on using Fischer time exclusively for DGS.

    Makes sense to me.

    Thanks to DGS player W. Spencer Clark I (DoubleU) for pointing this out!

    Saturday, February 20, 2010

    February KGS Tournament: ** VINDICATION **

    September 2009

    I'd just started playing go seriously again for a month or less.

    My first ever go tournament. Handicap tournament, thankfully.

    Second game, I play an 11kyu (I was a 16kyu, just barely), so  I get a 5 stone handicap.

    37 minutes and 281 moves later, I have lost by 237.5 points.

    As I pointed out in the blog post (my third post here), I'd have actually lost by less by playing something like the landslide fuseki. As a sidenote, sadly, the original location of the landslide fuseki (Temple of Go) does not appear to exist any longer...(?)

    The opponent who destroyed me so thoroughly, Verda, went on to win that tournament.

    Needless to say, I left remembering the name Verda and fearing it.

    February 2010

    I'm playing my second game of the Feb KGS tournament, and who am I paired against?

    Verda, first time since the tournament in September.

    He's even got the crown logo on his name that KGS applies when you've won a tournament.

    Except now he's not 11kyu, but 8kyu. Of course, I'm not barely a 16kyu, but a solid 11kyu. So instead of 5-6 stones difference, there is now 2-3.

    But... this is an even game.

                                No handicaps.

                                                         No mercy.

    All I remember going into the game (15 minutes per side, aggressive byoyomi), is  

    I really want to win this...

    You can view the SGF here.

    It was a nail-biter.

    In the middle, I'd surely lost.

    But, amazing comeback (and no little luck).

                    Failed ladder.

                                      Enormous snapback.

                                                               Very exciting.

                                                                                Highlight reel in the next post.
    In the end?

    I win.

     Hell yeah.

    Sunday, February 14, 2010

    Ladders, Cutting, Keimas and One Point Jumps

    One of the first two topics in go introductory books are ladders and nets.

    So this is fundamental stuff, right? At 12-10kyu, or whatever I am now, it should be old news, right?

    For a long time, I thought ladders and nets were relatively rare -- either would only enter my thoughts 0-2 times per game.

    This is before I really analyzed cutting, and started to look at every connection and cut point carefully, reading each one out several moves.

    Once I did this, two things happened:
    1) I suddenly jumped 1-3 stones in effective skill level.
    2) I started seeing ladders 50-100 times during a game (they just rarely showed up on the board)

    For example, every time I see a keima (my own or opponents, played or potentially playable), I consider the two possible ladders involved in it's cut, and whether or not they will work. This alone makes ladders suddenly an integral part of my games now.

    This directly stemmed from the first chapter in Lessons in the Fundamentals of Go, an interesting and hilarious (possibly unintentional) book that I hope to get around to reviewing in more detail. The ladder portion was basically telling you to practice to learn to read ladders, and that you are a ass-clown if you do not. Basically. This book is strange.

    What in the hell is going on in the picture on the cover of this book?

    Do I want to know?

    Cutting the Keima

    The ubiquitous keima connection always involves two ladders.
    First, let's look at how to cut one.

    You have to start at either a or b in the diagram. These are miai, meaning if the opponent want to maintain the connection (and he will), he must play the other point.

    After the opponent blocks, you must play either c or d depending on whether you played a or b -- you would not have started the cut (one would hope) without intention to follow through.

    The first move forces the next two moves, and you have the choice of two sequences (attacker chooses):

    Which you choose depends on the cuttability of each pair of points.
    The cuttability of the first point requires, at the minimum, that the ladder works.

    Here are the two ladders possible, one for each starting point:

    The ladder failing is necessary for the cut to succeed, but far from sufficient -- there are many ways for a cutting stone to be captured besides a direct ladder.

    The basic summary of cutting a keima is: any of the two pairs of cutting points must work

    In creating and maintaining your own keimas, ensure that the above does not become true.

    Cutting the One Point Jump

    The one point jump is rarely cut. The reason is that there are four potential cutting points, and the defender gets to choose which one it is! This means that all four cutting points must be cuttable! This stems directly from the flexibility of the connection -- the defender gets to choose the cut point from four available, so if just one of the four is not cuttable, the entire connection is safe. On top of that, at the end of the 5 move sequence, the defender has grown a lot of thickness (qb).

    OPJ Cutting Points -- defender chooses, so all four must be cuttable to allow connection to be cut.

    That's a pretty high requirement, which is why OPJs are so rarely cut.

    Let's look at it in detail.
    1. Cut begins on center point.
    2. Defender gets to choose one of two sides to block from.
    3. Attacker's extension is forced.
    4. Defender can patch one of the two cutting points
    5. Attacker can now cut that one point.

    Here are the four cut sequences, and remember: defender chooses!

    Nets and Cuts

    Oddly enough, I seem to have recently had a blind spot for nets. I'm not sure why. I'll have to focus on that next.

    Let's look at this example:

    Without the ladder breaker, this would be non-cuttable.
    With the ladder breaker, it IS cuttable.

    In case you forgot how ladders work.

    In case you forgot how ladder breakers work (now white can choose from several double-ataris)

    With just a slightly longer 'backboard', you can now utilize a net even though the ladder still fails. This is the trap I've fallen into recently because I'm so excited that the ladder works, that I jump in and cut, only to be caught in a net.

    In case you forget how nets work...

    Cutting Other Connections

    As far as cutting two-point jumps and ogeimas, that's for another time. I think ogeima cutting is similar to keima cutting, but I haven't looked at it in detail.


    A big thing being glossed over here is cuttability. I'm not sure there's a magic answer here, just do some reading.

    Reading Ladders

    After reading the first chapter in the book mentioned, I took the author's advice and started trying to read large ladders better. It's not really that difficult, and it's saved me a few times in games since.

    I tend to try to find the nearest star point to the start of the ladder, then find the nearest one to a side or potential ladder breaker, and start the zig-zag pattern at that point in the same direction. Works pretty well for me.

    Because ladders now affect the game so much more for me, my whole board game has become much more intriguing. Now, whenever I see a cut point that involves a ladder, now the other side of the board affects the cuttability of that cut point. Ideally, if I can place a ladder breaker across the board (preferably in sente), I've turned a uncuttable point into a cuttable one, and all the consequences that entails.

    Drawn Out Ladders

    One of the high drama situations in a game. It is the go equivalent to going 'all in' in poker. It doesn't happen too much in high skill games though (unlike the other high-drama situation in go -- high stakes ko fights, which also draws parallels to poker).

    If a ladder is being drawn out, it means either:
    1) Each player is reading the ladder differently, and betting (often the entire game) that the other player is wrong. Since one must be wrong, that means they cannot read (some) ladders, which is why you don't see this (rare exceptions) in high level games.
    2) One player is a beginner who can't read ladders.

    I know I've drawn out a ladder one move by accident before I've realized it was a ladder, even recently. But when you see it drawn out like 10 moves, it's usually the game determiner.

    And there's nothing quite as fun as winning due to a long drawn out ladder because you read it correctly and your opponent did not. I'm also amused by the occasional 'am I crazy?' pauses when a ladder is being drawn out -- it doesn't take much thinking to figure out the next move in a long ladder, so it's normally "snap-snap-snap-snap", but occasionally one player or the other pauses to re-read it to ensure they aren't nuts.

    At least, I know I do.

    Saturday, February 13, 2010


    No matter what, EVERY time I play go, whether in person or not, I sweat. A lot.

    As in, I have to change shirts afterward.



    But, I feel it says something for the game.

    After all, I'm sitting on my ass, barely moving.

    Only occasionally reaching over to click a mouse or place a small stone.

    And I'm enjoying myself.

    Yet I sweat bullets, and am often physically exhausted after an hour's game.

    I think it speaks to the intense concentration the game demands. I think it's a good thing.

    But it's still nasty.

    Crazy game

    The only 1 of 4 games I won tonight on KGS was against an alleged 10 kyu player, but I seriously doubt this rank.

    Game went like this: black and white play 4-4 points, each on his own side.
    Then black does a 3-3 invasion (?!) as his third move. WTF?

    Ok, I build a wall towards my other stone. But he goes beyond joseki, and keeps crawling on the 2nd line. Bewildered, I happily push him down the 2nd line until he stops. Then hane under, I block, but instead of patching his side he invades the OTHER 3-3 corner!

    Confused by happy, I build this wall facing the other. Again, he crawls excessively along the 2nd line, then does not finish securing his corner.

    What's next move? Invasion on my side, near one of the walls! So I pincer, he jumps out. Awesome I think, I jump out too, making a nice fat moyo on one side. Jumps out more! I follow.

    Here's the weirdest part: cuts on of the walls where he'd be laddered to death. Ok, so I atari to start the ladder, and he plays it out! Um, ok, so I continue, FIFTY more moves. I had to pause a few times to make sure *I* wasn't the insane one, but nope, it was true -- he drew a failed ladder across the entire board up to the 2nd line, then resigned.


    Here's the game, if you want to see it. And here's the players rank graph (I played him when he was 10kyu, the night previous to this posting... if he keeps playing, the graph should have tanked by now).

    Weird. I'll update this post with a picture of the game soon.

    Coming Up

    I actually mean to post more, here's what I have coming up:
    • Sensei's Library postings on
      • connections
      • cutting
      • ladder vs nets
    • A weird 9x9 against brand new player chicagoben that I'm not quite sure how to score (variation previous to blunder at end). Seki with ko related to eye killing shape and no place to play extra stones... should be interesting.
    • ?