Only Chess
05 Mar 15
05 Mar 15
Apparently apart from 6 pieces against a lone king they have been calculated and put on a website [1]. The problem is that one has to buy a product to get a years subscription which I don't want to do just for the sake of finding out the theoretical result in a game I've just finished. If someone has subscribed, or otherwise have access to an oracle, do they mind looking this one up. Alternatively what's your opinion, I won, but should I have from the following position? Could my opponent have held a draw?
The position when we entered the seven piece endgame was:
And the full game, with a few comments is:
[1] http://chessok.com/?page_id=27966
05 Mar 15
Originally posted by DeepThoughtWas this a live game or turn based?
Apparently apart from 6 pieces against a lone king they have been calculated and put on a website [1]. The problem is that one has to buy a product to get a years subscription which I don't want to do just for the sake of finding out the theoretical result in a game I've just finished. If someone has subscribed, or otherwise have access to an oracle, d ...[text shortened]... k to stop his pawn promoting.} Ne5 81. Rxf6+ 1-0 [/pgn]
[1] http://chessok.com/?page_id=27966
06 Mar 15
Originally posted by iChopWoodForFreeThankyou, endgames have always been a part of the game I've liked.
Ok. I was going to say the way you played the endgame was admirable but especially so if it was OTB or live chess.
When I finish games I aways add them to the database I keep of my own games and let Crafty annotate them, to look for things I missed in the game. With this one, once I'd done that, I just set the engine running analysing the position in the above diagram and switched on access to 5 piece table bases. It's been going for the last 10 hours and it's not found a win yet. But it wouldn't necessarily, its positional evaluation algorithm is probably flaky for endings and, as it's giving white +1.6, it will tend to choose a Principle Variation which avoids EGTB draws (as they drop the evaluation to zero, compared with objectively drawn positions it doesn't know are draws which it will assess as +1.6) and EGTB wins (as even if the position is objectively lost the positional evaluation will still prefer +1.6 or whatever to +327). So it's hard to know how to interpret it not being able to find a win. It's currently at depth 29 (ply I assume). The branching factor is of the order of 18 and the tree pruning algorithm becomes less effective because of the large number of positions which aren't all that different from each other, which is why engines become weaker in the ending. Which is why I wondered if anyone had access to seven piece tablebases as they are available, but one has to spend money.
06 Mar 15
1 edit
Originally posted by StansiMany thanks. Although I suspect I may have drifted into a draw once or twice after the position I gave in the FEN. I have Linux, which isn't a complete bar on using Windows programs, but means it would be running on a virtual machine and could be very slow. I'll try it out anyway and report back.
You can generate a tablebase for this position using FinalGen (which is free). It's a win for White.
07 Mar 15
Originally posted by DeepThoughtThe position was a draw after 79.Rb7 (but Black needed to reply Nd3+). It was still a win after 70...Nf8. Between these moves it's unclear, as FinalGen isn't able to solve these positions fully (just says White wins or Draw). Maybe someone with access to the Lomonosov tablebases can tell you the status of these positions.
Although I suspect I may have drifted into a draw once or twice after the position I gave in the FEN.
Here's an optimal line from the diagrammed position according to FinalGen (which doesn't mean optimal in the depth-to-mate sense, as FinalGen doesn't calculate this depth): 61.Kf5 Nd5 62.Ra3 Ne7+ 63.Ke4 Ng6 64.Rh3 Nf8 65.Kf5 Ne6 66.Rg3+ Kf8 67.Ke4 Nc5+ 68.Ke5 Nd7+ 69.Kf5 Nb6 70.Rb3 Nd7 71.Rd3 Nc5 72.Rc3 Nd7 73.Rc7 Ke7 74.Rc6 Nf8 75.Rxh6 and we're down to six pieces, so you can use the Nalimov tablebases from here onward.
07 Mar 15
2 edits
Originally posted by StansiYes, after making the move Rb7 I did have the feeling of "maybe I shouldn't have done that", which is never a pleasant sensation. Looking at Crafty's annotation of the game the assessment drops from +1.85 to +0.25 with that move. It also does at 78. Re7, apart from that the assessment bounces between +1.85 and +1.6 so I think that based on the evidence I drifted into a draw twice, but otherwise it was won.
The position was a draw after 79.Rb7 (but Black needed to reply Nd3+). It was still a win after 70...Nf8. Between these moves it's unclear, as FinalGen isn't able to solve these positions fully (just says White wins or Draw). Maybe someone with access to the Lomonosov tablebases can tell you the status of these positions.
Here's an optimal line from the di ...[text shortened]... f8 75.Rxh6 and we're down to six pieces, so you can use the Nalimov tablebases from here onward.
Judging by what it says on their website, FinalGen's really for pawn endings, it can only cope with 1 piece per side, so if there is no winning line not involving a pawn promotion leaving two pieces for white (say) it cannot calculate the position. It also, quite sensibly, has some internal checks that prevent it from giving a complete solution if the position is too complex. It'll calculate distance to conversion rather than distance to mate. But since the whole sequence takes less than 50 moves, 15 until the pawn is taken and another 25 to checkmate, it's not really an issue.
The line you've given from the position at move 61 is:
and according to the 6 man tablebases on the Shredder site it's a win in 25 more moves. The correct plan looks to be to force the black king away from the h-pawn using the rook along the g-file forcing it to choose between defending the f-pawn and then push the knight around until the h-pawn can be taken. I was fixated on controlling the 7th rank in the game.
Great - thanks very much for the help!