World Draughts Forum

It is currently Sun Sep 24, 2017 02:32

All times are UTC+01:00




Post new topic  Reply to topic  [ 39 posts ]  Go to page 1 2 3 Next
Author Message
PostPosted: Fri Jun 09, 2017 17:25 
Offline

Joined: Thu Apr 26, 2007 17:51
Posts: 792
Location: FRANCE
Hi,

We all know the following position reached between Scan and Mobi Dam during 2015 Olympiad.

Image
Black to play

We agreed to say that instead of 12-18 black could reach a draw by 15-20 or 4-10 but my question (and the challenge for you) is the following : is the position really a losing one after the played moves 12-18 31-26 ?

Image
Black to play

My point is the following:
After 15-20 40-35 23-28 36-31 20-25 49-43 14-20 42-38 4-9 38x27 3-8
I do not see any win for white!

_________________
Gérard


Top
   
PostPosted: Sat Jun 10, 2017 18:32 
Offline

Joined: Tue Jul 07, 2015 06:48
Posts: 235
Real name: Fabien Letouzey
Hi Gérard,

TAILLE wrote:
My point is the following:
After 15-20 40-35 23-28 36-31 20-25 49-43 14-20 42-38 4-9 38x27 3-8
I do not see any win for white!

You seem right! Scan would play the white moves in your line, and sees the draw after that.

Fabien.


Top
   
PostPosted: Sat Jun 10, 2017 19:06 
Offline

Joined: Sat Apr 28, 2007 13:53
Posts: 725
Real name: Ed Gilbert
Location: Morristown, NJ USA
Gerard, your subject reminds me of the old Cher movie, "Come Back to the Five and Dime, Jimmy Dean, Jimmy Dean". Maybe that was intentional. :-)

-- Ed


Top
   
PostPosted: Sat Jun 10, 2017 20:23 
Offline

Joined: Thu Apr 26, 2007 17:51
Posts: 792
Location: FRANCE
Fabien Letouzey wrote:
Hi Gérard,

TAILLE wrote:
My point is the following:
After 15-20 40-35 23-28 36-31 20-25 49-43 14-20 42-38 4-9 38x27 3-8
I do not see any win for white!

You seem right! Scan would play the white moves in your line, and sees the draw after that.

Fabien.


Hi Fabien,

Suppose Scan would have been the black side. Could you tell us if Scan would have found the black right moves?
The current Damy version is not able to find it in a reasonable time ... I have to continue to work!

Same question to other programmers.

_________________
Gérard


Top
   
PostPosted: Sat Jun 10, 2017 21:07 
Offline

Joined: Tue Jul 07, 2015 06:48
Posts: 235
Real name: Fabien Letouzey
TAILLE wrote:
Suppose Scan would have been the black side. Could you tell us if Scan would have found the black right moves?

It probably wouldn't have. Are you suggesting that the drawing line is (mostly) unique?

If I number your line from 1 (after 12-18 31-26), Scan would play:
1... 7-12 or 23-28 (instead of 15-20)
2... 23-28
3... 18-23 (instead of 20-25)
4... 3-8 (instead of 14-20)
5... 3-8 6. 38x27 4-9 (it looks the same)

All the scores are around -1.00, so Scan is not aware of the draw at this stage.


Top
   
PostPosted: Sun Jun 11, 2017 06:46 
Offline

Joined: Tue Aug 14, 2012 21:24
Posts: 127
Real name: Catherine Bourneuf
Hi Gérard,

If i understood, this game is draw !?? This mean that instead of 12-18, Mobydam must play 15-20 to draw the game??

Curiously, in this position, Mobydam always plays 12-18.

Catherine.


Top
   
PostPosted: Sun Jun 11, 2017 09:25 
Offline

Joined: Thu Apr 26, 2007 17:51
Posts: 792
Location: FRANCE
Catherine wrote:
Hi Gérard,

If i understood, this game is draw !?? This mean that instead of 12-18, Mobydam must play 15-20 to draw the game??

Curiously, in this position, Mobydam always plays 12-18.

Catherine.


Hi Catherine,

Of course you perfectly know that this game was won by white. My goal is only to analyse thoroughly this kind of game in order to find where was the losing move and in order to find some hints to improve Damy in this difficult phase in which each side has 8-10 men.

My feeling is that the move 12-18 played by Mobydam was not the losing move but our programms seems in great difficulties to decide between draw or win.

After 12-18 31-26 I "think" that the move 2-8 played by MobyDam is a losing move but currently Damy is not able to prove this point.

_________________
Gérard


Top
   
PostPosted: Sun Jun 11, 2017 11:03 
Offline

Joined: Thu Apr 26, 2007 17:51
Posts: 792
Location: FRANCE
Fabien Letouzey wrote:
TAILLE wrote:
Suppose Scan would have been the black side. Could you tell us if Scan would have found the black right moves?

It probably wouldn't have. Are you suggesting that the drawing line is (mostly) unique?

If I number your line from 1 (after 12-18 31-26), Scan would play:
1... 7-12 or 23-28 (instead of 15-20)
2... 23-28
3... 18-23 (instead of 20-25)
4... 3-8 (instead of 14-20)
5... 3-8 6. 38x27 4-9 (it looks the same)

All the scores are around -1.00, so Scan is not aware of the draw at this stage.


Hi Fabien,

I begin with your first proposal 12-18 31-26 7-12
The issue is that a move like 7-12 or 2-8 (instead of 15-20) weakens the right black side.
Damy is not able to prove the win after 7-12 but I suspect it exists.

After 7-12 49-43 the position is the following

Image
Black to move

How do you continue?

You have to avoid the losing sequence of the game and in addition the sequence
4-9 35-30 23-29 34x23 18x29 30-25 14-20 25x14 9x20 42-38 20-24 38x27 is also a loosing one.

_________________
Gérard


Top
   
PostPosted: Sun Jun 11, 2017 11:47 
Offline

Joined: Tue Jul 07, 2015 06:48
Posts: 235
Real name: Fabien Letouzey
Hi Gérard,

TAILLE wrote:
Image
Black to move

How do you continue?

2-8, but apparently there is trouble after (2-8) 36-31 23-28 30-24 14-20 34-29 18-22.

Fabien.


Top
   
PostPosted: Sun Jun 11, 2017 12:24 
Offline

Joined: Tue Jul 07, 2015 06:48
Posts: 235
Real name: Fabien Letouzey
TAILLE wrote:
The issue is that a move like 7-12 or 2-8 (instead of 15-20) weakens the right black side.
Damy is not able to prove the win after 7-12 but I suspect it exists.

It seems that you are right, and 7-12 is dubious. With more time, Scan prefers 15-20 after all, with a reduced white advantage.

Your question was whether Scan could find the black moves. In a couple of minutes (single core), the answer is no.


Top
   
PostPosted: Sun Jun 11, 2017 14:12 
Offline

Joined: Thu Apr 26, 2007 17:51
Posts: 792
Location: FRANCE
Fabien Letouzey wrote:
TAILLE wrote:
The issue is that a move like 7-12 or 2-8 (instead of 15-20) weakens the right black side.
Damy is not able to prove the win after 7-12 but I suspect it exists.

It seems that you are right, and 7-12 is dubious. With more time, Scan prefers 15-20 after all, with a reduced white advantage.

Your question was whether Scan could find the black moves. In a couple of minutes (single core), the answer is no.


Hi Fabien,

If you agree on 15-20 (instead of 7-12) we can now switch to your second proposal : 15-20 40-35 23-28 36-31

Image
black to play

and now 18-23 (instead of 20-25)
On this move white can switch to 39-33 28x39 34x43

Image
black to play

and white theat is to continue with 49-44 followed by 42-38
How do you try to draw with black?

_________________
Gérard


Top
   
PostPosted: Sun Jun 11, 2017 14:58 
Offline

Joined: Tue Jul 07, 2015 06:48
Posts: 235
Real name: Fabien Letouzey
TAILLE wrote:
Image
black to play

and white theat is to continue with 49-44 followed by 42-38
How do you try to draw with black?

3-8 30-25 20-24 42-38 8-12 38x27 7-11 but, again, the score is slowly increasing for white.


Top
   
PostPosted: Sun Jun 11, 2017 17:05 
Offline

Joined: Thu Apr 26, 2007 17:51
Posts: 792
Location: FRANCE
Fabien Letouzey wrote:
TAILLE wrote:
Image
black to play

and white theat is to continue with 49-44 followed by 42-38
How do you try to draw with black?

3-8 30-25 20-24 42-38 8-12 38x27 7-11 but, again, the score is slowly increasing for white.


Oops very strange sequence indeed.

In my previous post I suggest for white to play 49-44 followed by 42-38 in order to avoid the move 30-25 which is very often positionnally very bad if you want to win.

In order to show you this point let's take the hypothetical sequence after your proposal
3-8 30-25 20-24 42-38 8-12 38x27 7-11 and now 49-44 4-9 41-37 9-13 37-32 11-17 27-21 13-18 31-27 2-7 (don't look at the sequence itself but on the resulting position):

Image
white to play

Look now at the following position

Image
white to play

The above position is a winning one : all the white moves are winning except 27-22 (obvious), 21-16 and 30-25.
You can see in particular that after 30-25 19-23 we reach the first diagram and white cannot win!

I often said that in draughts game one of the most difficult move decision is to play 30-25 because this move may be very good or very bad depending on a lot of criterias.
As a consequence, unless I am sure 30-25 is a good move I prefer to keep the man 30 on its square.

_________________
Gérard


Top
   
PostPosted: Sun Jun 11, 2017 17:18 
Offline

Joined: Tue Jul 07, 2015 06:48
Posts: 235
Real name: Fabien Letouzey
TAILLE wrote:
Oops very strange sequence indeed.

2-8 might be better, as the white score is not increasing. The PV is 2-8 43-39 7-12 49-43 23-28 41-36 20-25.


Top
   
PostPosted: Sun Jun 11, 2017 20:55 
Offline

Joined: Thu Apr 26, 2007 17:51
Posts: 792
Location: FRANCE
Fabien Letouzey wrote:
TAILLE wrote:
Oops very strange sequence indeed.

2-8 might be better, as the white score is not increasing. The PV is 2-8 43-39 7-12 49-43 23-28 41-36 20-25.


Yes Fabien with 2-8 the black position seems far better to reach the draw.
BTW I have now identified clearly several basic improvments in order to help Damy to solve such position. I will now take several days to program and run some new tests.

_________________
Gérard


Top
   
Display posts from previous:  Sort by  
Post new topic  Reply to topic  [ 39 posts ]  Go to page 1 2 3 Next

All times are UTC+01:00


Who is online

Users browsing this forum: No registered users and 3 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
cron
Powered by phpBB® Forum Software © phpBB Limited