Author: Christophe Theron
Date: 09:21:39 10/19/00
Go up one level in this thread
On October 19, 2000 at 07:04:14, Uri Blass wrote:
>On October 19, 2000 at 05:50:54, Ricardo Gibert wrote:
>
>>On October 19, 2000 at 01:06:03, Uri Blass wrote:
>>
>>>On October 19, 2000 at 00:51:19, Christophe Theron wrote:
>>>
>>>>On October 18, 2000 at 16:31:54, Ricardo Gibert wrote:
>>>>
>>>>>On October 18, 2000 at 15:57:10, Uri Blass wrote:
>>>>>
>>>>>>On October 18, 2000 at 15:52:09, Ricardo Gibert wrote:
>>>>>>
>>>>>>>On October 18, 2000 at 14:14:20, Uri Blass wrote:
>>>>>>>
>>>>>>>>On October 18, 2000 at 12:34:42, Ricardo Gibert wrote:
>>>>>>>>
>>>>>>>><snipped>
>>>>>>>>>The algorithm to determine whether this type of position is drawn is quite
>>>>>>>>>simple. With the h-pawn, Black draws if the K can get into the four corner
>>>>>>>>>squares g7,g8,h7,h8. Otherwise, the outcome is not assured. Note, it is not
>>>>>>>>>enough for the King to get in front of the pawn as the following position
>>>>>>>>>illustrates:
>>>>>>>>
>>>>>>>>I can give another rule:
>>>>>>>>Blacks draw if it is black to move and the distance between the black king and
>>>>>>>>the corner in king moves is smaller than the ditance between the corner and the
>>>>>>>>white pieces except the bishop(white king in king moves and white pawn in pawn
>>>>>>>>moves)
>>>>>>>>
>>>>>>>>If it is white to move then use distance-1 instead of distance for white.
>>>>>>>>
>>>>>>>>Uri
>>>>>>>
>>>>>>>I like your rule. It's interesting, but it needs to be amended somewhat:
>>>>>>>
>>>>>>>[D]4B3/8/8/8/8/2k5/7P/K7 b - - 0 0
>>>>>>
>>>>>>This position is not a draw by my rule.
>>>>>>The distance of the white pawn to the corner in pawn moves is equal(and not
>>>>>>smaller) to the distance of the black king to the corner in king moves.
>>>>>
>>>>>Yep, I goofed.
>>>>>
>>>>>>
>>>>>>>
>>>>>>>as you can see, your rule is a little dangerous. There may be other examples,
>>>>>>>but this is what I came up with in a about a minute. Still your rule is
>>>>>>>promising.
>>>>>>>
>>>>>>>There are some positions your rule does not give as drawn, but "my" rule does.
>>>>>>>The reverse is also true of course, so maybe the rules should be combined?
>>>>>>
>>>>>>I agree that the rules can be combined.
>>>>>>
>>>>>>Uri
>>>>
>>>>
>>>>
>>>>This discussion is very interesting. So what's the ultimate rule?
>>>>
>>>>I had to work on this recently in Chess Tiger (and Gambit Tiger of course), and
>>>>I'm not sure my own rule is as good as yours.
>>>>
>>>>Waiting eagerly for the next post...
>>>>
>>>>
>>>>
>>>> Christophe
>>>
>>>The logic behind my rule is simple
>>>
>>>If black king is closer to the corner than the white king and the white pawn
>>>nothing can stop it if it begins a trip to the corner.
>>>
>>>It can do the trip using squares that the bishop does not control and the other
>>>pieces cannot see these squares.
>>>
>>>Uri
>>
>>I've been think about "the Blass ending rule" some more and it appears to be
>>robust enough to handle the more general case. The cases where the superior side
>>has more than one Bishop on the same color squares and/or more than one rook
>>pawn on the same file. I can't think of a counter example. Can you?
>
>I agree.
>If the defender side has only king my rule is correct.
Oops! That's a serious flaw in your rule. You mean I just have to add a black
king and it does not work anymore?
Shit! :)
>I assume for the rest of the discussion that without loss of generalization the
>defender is black and white has pawns only in the h file
>
>I believe that we can generalize also for cases when black has pawns but not on
>the g file but we have to be careful to see that there are no black pawns in
>black squares that are closer than the black king to the corner.
For me it's enough. I don't really care about the case where black has still
pawns.
I'm only interested in KB xP / K, where all the white pawns are either on the A
or the H file.
Of course if you can generalize more when black has pawns and come with a
not-too-complex rule, I'm also interested. But it would only be a bonus for me.
Christophe
>We can evaluate the following positions as draws if we are not carful
>
>[D]8/8/5p2/4kB2/6KP/8/8/8 b - - 0 1
>
>[D]8/4p3/4pp2/4kp2/5p2/5K2/7P/7B w - - 0 1
>
>
>Uri
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.