nimzovich wrote:
Words certainly have sharpened. I have always suspected that KQ vs KR was quite an emotive topic. Good thing the posting was not on BOOC.
I'm sure there'll be time for that too. :)
KQ vs KR Ending...
Sort:
normajeanyates wrote:
So try this. Play your strongest program, WITHOUT tablebases; against your WEAKEST tablebase-accepting-program, Give the latter the following tablebases: KQKR, KQK. That will do. Give your strongest the position: 8/8/8/8/2r5/8/2k5/K6Q w - - 0 1
not to re-ignite anything, but I just downloaded 4-piece nalimov, and gave this position a try. my first try took 51 moves, second try 52, then, once I realized how I should get my king out from the corner without wasting so much time, it took 43 moves and about 5 minutes. there's nothing particularly hard in this position, it simply starts farthest away from the standard 3rd rank defence, so you need to cross the board first. other than that, it's just business as usual. at move 25 the standard 3rd rank defence arises, and I blitzed the rest in roughly 20 seconds.
a nice addition to training positions for anyone who wants to learn the KQkr mate though.
The ChessBase webiste carried an article aboutthis endingapprox 18 months ago, together with a database of positionsand analysis. The author had discovered several "rose" positions (kind of a circle with each piece separated by a knight's move) ... from which there was only a single move which produced the shortest path to victory. The author was also very impressed by a Gelfand example of the ending and suggested he must have donw quite a lot of work on it in advance.
I also recall that David Howell won it quite smoothly at 30s a move (increment only by that stage) in the World Junior a couple fo years back.
Paul
The difference between mating a computer and a person lies in the fact that the computer probably looks at things in terms of, "Which move is mate int the least number of moves?", creating predictibility. Try mating against a human of similar strength, as they may more random (though not blundering) moves, which may or may not be harder. Anyway, the reason I'm posting is that I'm wondering whether or not there is a set of rules and/or patterns that can be used to mate in such a position?
Dec 12, 2015
0
#26
Had this come up seven or eight times only twice did I have the Queen. There are numerous drawing positions for the player with the rook and I've used two of them and drawn by 50-move rule at least four times as well. Obviously, the key for the player with the Queen and the theoretically won game is to understand the nature of the stalemates and the forced rook sacrifices that bring them about; and understand the ideas of the "perpetual-checks" and then avoid them and avoid the 50-move draw as well. Much easier said then done!!!! In Pandolfini's Endgame Course pages 49-55 you get a good introduction especially to the weaker side's traps. Chernev's Practical Chess Endings diagram #251 shows the key winning position which the stronger side achieves with himself 'on the move.'. He must then triangulate and re-create the position with the weaker side on the move . . . that part is easy! Chernev's then analyses the winning moves - a bit awkward the first couple times through as you study it, but then you see the pattern and it becomes pretty easy also. The bottom line is that the weaker side loses his rook or gets mated within four moves despite all his options . . . easy once the stronger side has triangulated his opponent into zugzwang. Good luck!!
do you know the moves of chess? Are you sure? I wouldnt be if i were you ...