Sorry, I just realized this is about 6x6 go. Please ignore my previous response.


Verzonden: ma 29-9-2008 20:09
Aan: [EMAIL PROTECTED]; computer-go; computer-go
Onderwerp: RE: [computer-go] Analysis of 6x6 Go

I (EGF 4d) am probably not strong enough to give well founded comments on 9x9 
games, but already move 2 at D3 seems strange from a shape point of view 
(whatever that may be worth on 9x9)
The continuation B C3 B4 D5 seems the most natural continuation once D3 is 
played, but on 19x19 this is kind of exchange is usually bad for white (he gets 
a hane on the head of two stones).
Black's last move at D5 would definitely be better than D2 on 19x19 and I would 
be very surpised if D2 would be better on 9x9.
I'm speculating Leela's tendency to respond B C4 at D3 to be the cause of the 
discrepancy between the 2.0 komi from Leela and the 4.0 komi from Erik. 
Might W D3 be 2 points worse then the optimal white move (unknown to me)?
Is there any support for W D3 being good from professional 9x9 games? I've 
never seen it in professional play, but I'm not a specialist on 9x9.


Van: [EMAIL PROTECTED] namens Don Dailey
Verzonden: do 25-9-2008 22:14
Aan: computer-go
Onderwerp: Re: [computer-go] Analysis of 6x6 Go

On Wed, 2008-09-24 at 19:48 +0200, Erik van der Werf wrote:
> On Wed, Sep 24, 2008 at 6:30 PM, Don Dailey <[EMAIL PROTECTED]> wrote:
> > I don't know if even size boards are special, but it seems to me that
> > such small boards should have very high komi's.   4.0 seems pretty low
> > but then I'm really no expert on komi's and I'm a pretty weak player so
> > I'm not in any position to really say.
> The center is the best opening move for all small odd size boards.
> Small even size boards have a lower komi because there is no center
> point.
> I'm quite confident that 4.0 is the correct komi for 6x6.

I am playing games with Leela at 5 minutes per side on a loaded core 2
duo computer. 

>From the evidence I have now, which I admit is not enough to base a
solid conclusion on,  it looks like 2.0 is the correct komi.

When I set komi to 1.5,  black has won 10 out of 10 games.

When I set komi to 2.5, black onl wins 16.667% or 2 out of 12 games. 

When I did the 7x7 study over a year ago (or maybe 2) I noticed that at
reasonably strong levels it tended to be very one sided in one direction
or other based on how you set komi. 

My plan is to run a LOT of games at 2.5 komi and then analyze the
results based on the move sequences looking to see if some common early
black blunder is preventing wins for black at 2.5 komi.  

When I do this I will try to reorient the move sequence to some
canonical representation so that we are not looking at too many
equivalent games with different orientations. 

Superficially, I noticed this:

      1 W C4 D3 C3 D4 D2 E2
      1 B C4 D3 C3 D4 D5 C2

Which means when black played D5 on move 5 he won, but when he played D2
he lost.

   1 B C4 D3 C3 D4 D2 E2 D5 E5 E3
   1 W C4 D3 C3 D4 D2 E2 D5 E5 E4

Same above - Black played 2 different moves and got two different

The other games vary before this but could be transpositions of these
positions - I don't have the time right now to compute all the
transpositions to check this out.

I didn't actually look at those moves so I don't know if they are game
changing or not.   Are there any strong players willing to comment on
these 2 diversions?

The other possibility is that white is supposed to WIN all those games
and is making the occasional error.   The results indicate that is a
more likely possibility.

Here is the complete list of games up to the 9th move.  The first column
is the number of times this exact result/sequence was played.

      1 W D4 C3 C4 D3 B3 B2 D2 C2 E4
      1 W D4 C3 D3 C4 C5 B5 B3 C2 D6
      1 B C4 D3 C3 D4 D2 E2 D5 E5 E3
      1 W C4 D3 C3 D4 D2 E2 D5 E5 E4
      1 B C4 D3 C3 D4 D5 C2 E5 B2 D2
      1 W C4 D3 C3 D4 D5 E5 D2 E2 E4
      1 W C4 D3 D4 C3 B3 B2 E3 E2 D2
      1 W C3 D4 C4 D3 D2 C5 E2 B5 E4
      1 W C3 D4 C4 D3 D2 E2 D5 E5 E4
      2 W C3 D4 C4 D3 D5 E5 D2 E2 E4
      1 W C3 D4 D3 C4 B3 E3 E2 E4 B5
      1 W C3 D4 D3 C4 B4 B5 D5 B3 B2
      2 W C3 D4 D3 C4 B4 B5 E4 E5 D5

- Don

      1 W C4 D3 C3 D4 D2 E2
      1 B C4 D3 C3 D4 D5 C2

But for now, perhaps you stronger go players can look at the following 6
moves sequences that represent the games.   The first column is how many
times this exact result/sequence occurred.   For instance you see that
white won 3 times when the game started "C3 D4 D3 C4 B4 B5"

Does anyone see any obviously bad moves for black?

      1 W D4 C3 C4 D3 B3 B2
      1 W D4 C3 D3 C4 C5 B5
      1 B C4 D3 C3 D4 D2 E2
      1 W C4 D3 C3 D4 D2 E2
      1 B C4 D3 C3 D4 D5 C2
      1 W C4 D3 C3 D4 D5 E5
      1 W C4 D3 D4 C3 B3 B2
      1 W C3 D4 C4 D3 D2 C5
      1 W C3 D4 C4 D3 D2 E2
      1 W C3 D4 C4 D3 D5 E5
      1 W C3 D4 D3 C4 B3 E3
      3 W C3 D4 D3 C4 B4 B5

What I see that is slightly interesting (just from this data, not
looking at the actual position) is that  C4 D3 C3 D4 D2

> Erik
> _______________________________________________
> computer-go mailing list

computer-go mailing list

Reply via email to