Author: Uri Blass
Date: 07:31:13 09/26/00
Go up one level in this thread
On September 26, 2000 at 08:11:19, Mark Young wrote: >On September 26, 2000 at 05:31:47, Uri Blass wrote: > >>On September 26, 2000 at 05:00:19, Mark Young wrote: >> >>>On September 25, 2000 at 22:28:03, Robert Hyatt wrote: >>> >>>>On September 25, 2000 at 15:40:20, Mark Young wrote: >>>> >>>>>On September 25, 2000 at 13:48:30, Robert Hyatt wrote: >>>>> >>>>>>On September 25, 2000 at 13:21:34, Mark Young wrote: >>>>>> >>>>>>>On September 25, 2000 at 09:01:19, Antonio Dieguez wrote: >>>>>>> >>>>>>>>hello! >>>>>>>> >>>>>>>>I come to ask two things, please... >>>>>>>> >>>>>>>>What are the results of the basics endgames KRB vs KR and KRN vs KR, both draws? >>>>>>>> if it is, there is some exceptions except the obvius? >>>>>>>> >>>>>>>>And can someone post some of these mate in n positions with n very very very >>>>>>>>high? >>>>>>>> >>>>>>>>Thanks! >>>>>>> >>>>>>>Easy position to win!! >>>>>> >>>>>> >>>>>>This is wrong. _most_ KRB vs KR and KRN vs KR are _drawn_. There are some >>>>>>wins for the stronger side, and even a couple of wins for the side without >>>>>>the B/N, but in general these are dead draws. >>>>> >>>>>You must not have had your coffee yet today, but the above is known as sarcasm. >>>>> >>>>>Thank you for the info on KRB vs KR endings I did not know that. >>>> >>>> >>>>Sorry... I apparently missed the sarcasm. :) >>>> >>>>It is amazing that an extra piece can't force a win. And it is more amazing >>>>that because of this, many programs will evaluate a KRB vs KRPP as winning for >>>>the KRB, when in reality, the RPP wins or it is a draw, the RB has _no_ winning >>>>chances. >>> >>>:) I thought you would have understood, the point being that even though the >>>position is a win for the stronger side in this case, what human could win this, >>>as one minor slip brings the position back to a draw again. The win is so >>>complex if you look at the mating line, a player would have to calculate all the >>>move to mate, as positional judgement just fails in positions like this. >> >>The distance to conversion is clearly smaller than 100 plies so some small >>mistakes are not going to prevent white to do the mate. >> >>It is clearly that white have to calculate but positional judgement can also >>help(if you learned winning positions before it can help because you know target >>positions to go for). >> >>It is possible that GM's who learned KRB vs KR can mate at tournament time >>control against programs with tablebases from this position and I do not know. >> >>Uri > >You are mistaken, KRB Vs KB is to complex for any positional judgement. Take a >look at the below position. Tell me if this almost identical position is a win >or a draw without tablebase, you can only use human calculation or positional >judgement. If the below position is a draw, tell us what positional factors you >used to come to that conclusion. I am not an expert of KRB vs KR It is clear that good GM's who learned KRB cs KR know 3 kind of the positions: 1)positions when they know that KRB wins and they know to win. 2)positions when they know that it is a draw and they know to draw. 3)positions when they need to search because they do not know. I understand that you say that part of the positions when they need to search are so complicated that they cannot see by search if it is a win or a draw. You may be right and I do not know but it is clear that positional judgement can help them in the positions of type 1,2 and at least in part of the positions of type 3. 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.