TalkChess.com

Original fen: G. Costeff (Isreal)| 2nd special price | 6/11 Win k2N4/p2p4/PPp1pp2/3p2r1/3K3p/2BR4/8/6nb w - - 0 0 [Event "?"] [Site "?"] [Date "????.??.??"] [Round "?"] [White "New game"] [Black "?"] [Result "*"] [SetUp "1"] [FEN "k2N4/p2p4/PPp1pp2/3p2r1/3K3p/2BR4/8/6nb w - - 0 0"] [PlyCount "32"] ...

This study was believed to be draw but it's likely winning for black. 1K6/8/2P3P1/1R6/8/5n2/4rrPk/6b1 w - - 0 1 This position is a win for Black, but it doesn't have anything to do with the ones of your following postings, does it, Jon? 1K6/8/2P3P1/1R6/8/5n2/4rrPk/6b1 w - - 0 1 Analysis by CorChess...

Hi Eelco! Suggested repair is moving the Knight one square down :roll: ... 4K3/B1kP3N/P7/8/8/8/b7/3r3B w - - ... 94/26 262:46 +M10 1.Nf6 Be6 2.Nd5+ Rxd5 3.Bb8+ Kxb8 4.Bxd5 Ka7 5.Bxe6 Kxa6 6.d8Q Kb5 7.Qd4 Ka5 8.Qc5+ Ka4 9.Qb6 Ka3 10.Qb3+ (65.144.850.387) 4131 [Event "?"] [Site "?"] [Date "2019.02.18"...

Hi Jon! From EG january 2019, page 30. No 22071 R. Becker honourable mention White wins. 4K2N/B1kP4/P7/8/8/8/b7/3r3B w - - 0 0 [Event "StrateGems 2017; honorable mention"] [Site ""] [Date "2017"] [Round ""] [White "Richard Becker (USA)"] [Black ""] [Result "1-0"] [SetUp "1"] [FEN "4K2N/B1kP4/P7/8/8/...

We now have some Windows and macOS exe's. Thank's a lot, Michael, especially for White's and Black's contempt instead of sidewise "Defensive"- Option, numeric contempt for both sides, in game- playing mode too, is somwhat easier to understand to me. BTW, tried Monge-compile x64 only, works well wit...

Cfish-dev, in what was likely a lucky run: ... info depth 30 seldepth 55 multipv 1 score cp 325 nodes 1954637481 nps 30243969 hashfull 403 tbhits 0 time 64629 pv d5f4 f3b7 f4h3 g1g2 a8b7 g2h3 b7f3 c4e3 f3f2 c1c8 f2b2 e3c2 g8f7 c8c7 f7g6 a1f1 b2a2 f1f3 a2b1 c7c3 b1h1 f3f2 h1d5 b3b4 e6e5 d4e5 d5e6 h3...

BTW it seems as for popct.- compile, that tbs- usage cannot be switched off by Syzygy Probe Limit 0, when MCTS is on. And could it be, that working of MCTS- learning files depends on wether the opening moves are stored in a .pgn too or not? It seems to me, at positions without move history leading ...

I think very promising and interesting. So do I, Andrea, thanks again. Learning files work, as far as I could see till now, not quite as the ones I knew till now from Shredder, Hiarcs, Houdini and Stockfish, but as for tactical best move- positions, already "found" solutions, get help at "refinding...

Leelas eval is in win percentage, so it actually dropped after 42. Rc2 (SF also shows a slight plus for some reason) and shot up to 90% after 43 Rc1. I trust your judgment but Rc2 prepares Rf3 defending the pawn; I assume there are some tactical tricks that both engines missed? Ah, didn't see LC0's...

amchess wrote: ↑

Mon Jan 28, 2019 7:04 pm

Version 6.0 released:
https://github.com/amchess/ShashChess/releases/tag/6.0

Thanks a lot, Andrea!
MCTS?