Arasan-Stooflees, TCEC Cup 5 2020

[jdart]

Arasan lost this game in the endgame after being even for a long time. Some analysis with Houdini 6 indicated to me that Ne1 might possibly draw here:

[d]6k1/6p1/p4r2/1p1Pp2P/1P2P2P/P3QNK1/8/q7 w - -

But Stockfish 11 did not find this in half an hour of search:

Code: Select all

6k1/6p1/p4r2/1p1Pp2P/1P2P2P/P3QNK1/8/q7 w - - bm Ne1; c0 "Arasan-Stooflees, TCEC Cup 5 Round 16 2020";

depth 45 seldepth 61 multipv 1 score -428 nodes 14932523138 nps 8293841 hashfull 963 tbhits 17888772 time 1800.435 pv Kg2 Qd1
depth 45 seldepth 88 multipv 2 score -432 nodes 14932523138 nps 8293841 hashfull 963 tbhits 17888772 time 1800.435 pv Nh2 Qb1 Nf3 Qc2 Ng5 Rf4 Ne6 Rf7 Ng5 Rf6 Nh3 Qb1 Kg2 Qd1 Kg3 Rf1 Nf\
2 Qxh5 Kh3 Rg1 Qc3 Kh7 Qd3 Qf7 Qe2 Ra1 Qe3 Kg8 Kg2 Ra2 Kg3 Qg6+ Ng4 Ra1 Qf3 Re1 Qf5 Re3+ Kf2 Qxf5+ exf5 Rd3 d6 Rxd6 Kf3 Rd4 Nxe5 Rxh4 Ke3 Rh3+ Ke4 Rxa3 Kd5 a5 bxa5 b4 Nc4 Ra4
depth 44 seldepth 78 multipv 3 score -405 nodes 14932523138 nps 8293841 hashfull 963 tbhits 17888772 time 1800.435 pv Kg4 Qf1 Kg3 Qc4 Ng5 Qd4 Qc1 Qd3+ Kg2 Qc4 Qe3 Qd4 Qc1 Rf8 Kh3 Qd3+ \
Kg2 Qc4 Qe3 Qf1+ Kg3 Rc8 Qd2 Qg1+ Kh3 Qa1 d6 Qc3+ Qxc3 Rxc3+ Kg4 Rd3 Ne6 Rxd6 Nc5 Rf6 Kg5 Kh7 Kg4 Rf4+ Kg5 Rf3 Nxa6 Rxa3 Nc7 Rb3 Nd5 Rf3 Kg4 Rf8 Kg5 Rf1 h6 gxh6+ Kg4 Kg6 Ne7+ Kg7 Nd5 K\
h7 h5 Kg8 Kg3 Kf7 Kg4 Rg1+ Kf3 Rc1 Ne3 Rb1 Nd5 Ke6 Nc7+ Kf6 Kg4

[Alayan]

I would trust SF11 over SF8.

And +4 is massive, beyond most fortresses eval.

The best way to resolve this is to get the position after Ne1, have SF11 analyze. If it thinks it's busted, go to the next ply and feed it to Houdini. Repeat until one of them see it was wrong. Then backtrack while keeping hash if you want to propagate it to the root.

[zullil]

Arasan lost this game in the endgame after being even for a long time. Some analysis with Houdini 6 indicated to me that Ne1 might possibly draw here:

[d]6k1/6p1/p4r2/1p1Pp2P/1P2P2P/P3QNK1/8/q7 w - -

But Stockfish 11 did not find this in half an hour of search:

Code: Select all

6k1/6p1/p4r2/1p1Pp2P/1P2P2P/P3QNK1/8/q7 w - - bm Ne1; c0 "Arasan-Stooflees, TCEC Cup 5 Round 16 2020";

depth 45 seldepth 61 multipv 1 score -428 nodes 14932523138 nps 8293841 hashfull 963 tbhits 17888772 time 1800.435 pv Kg2 Qd1
depth 45 seldepth 88 multipv 2 score -432 nodes 14932523138 nps 8293841 hashfull 963 tbhits 17888772 time 1800.435 pv Nh2 Qb1 Nf3 Qc2 Ng5 Rf4 Ne6 Rf7 Ng5 Rf6 Nh3 Qb1 Kg2 Qd1 Kg3 Rf1 Nf\
2 Qxh5 Kh3 Rg1 Qc3 Kh7 Qd3 Qf7 Qe2 Ra1 Qe3 Kg8 Kg2 Ra2 Kg3 Qg6+ Ng4 Ra1 Qf3 Re1 Qf5 Re3+ Kf2 Qxf5+ exf5 Rd3 d6 Rxd6 Kf3 Rd4 Nxe5 Rxh4 Ke3 Rh3+ Ke4 Rxa3 Kd5 a5 bxa5 b4 Nc4 Ra4
depth 44 seldepth 78 multipv 3 score -405 nodes 14932523138 nps 8293841 hashfull 963 tbhits 17888772 time 1800.435 pv Kg4 Qf1 Kg3 Qc4 Ng5 Qd4 Qc1 Qd3+ Kg2 Qc4 Qe3 Qd4 Qc1 Rf8 Kh3 Qd3+ \
Kg2 Qc4 Qe3 Qf1+ Kg3 Rc8 Qd2 Qg1+ Kh3 Qa1 d6 Qc3+ Qxc3 Rxc3+ Kg4 Rd3 Ne6 Rxd6 Nc5 Rf6 Kg5 Kh7 Kg4 Rf4+ Kg5 Rf3 Nxa6 Rxa3 Nc7 Rb3 Nd5 Rf3 Kg4 Rf8 Kg5 Rf1 h6 gxh6+ Kg4 Kg6 Ne7+ Kg7 Nd5 K\
h7 h5 Kg8 Kg3 Kf7 Kg4 Rg1+ Kf3 Rc1 Ne3 Rb1 Nd5 Ke6 Nc7+ Kf6 Kg4

Seems pretty clear to Stockfish-dev that Ne1 loses (as does everything else):

-8.09 1... Qa2 2. Nf3 Qc4 3. Ng5 Qd4 4. Qc1 Qd3+ 5. Kg2 Rf8 6. h6 gxh6 7. Qc6 Qf1+ 8. Kg3 Qf2+ 9. Kh3 Qf6 10. Qe6+ Kh8 11. Qxf6+ Rxf6 12. Ne6 h5 13. Kg2 Kg8 14. Ng5 Rf4 15. Kg3 Rg4+ 16. Kf2 Kf8 17. Ke3 Ke7 18. Nf3 Kf6 19. Kd3 Rg3 20. Ke2 Rg2+ 21. Ke3 Ra2 22. Ne1 Rh2 23. Nf3 Rg2 24. Ne1 Rg1 25. Nd3 Rg3+ 26. Ke2 Rg4 27. Nc5 Ke7 28. Kf1 Rxh4 29. Kg2 Kd6 30. Nb7+ Kd7 31. Nc5+ Ke7 32. Kg1 Rh3 33. Nxa6 Rg3+ 34. Kh1 Rxa3 35. Nc5 Rc3 36. Kg2 h4 37. Na6 Rc2+ 38. Kh1 Rc4 39. Kh2 Rd4 40. Nc5 Rxb4 41. d6+ Kxd6 42. Nd3 Rc4 (depth 50, 0:12:51)

-8.52 1... Qa2 2. Nf3 Qc4 3. Ng5 Qd4 4. Qc1 Qd3+ 5. Kg2 Rf8 6. h6 gxh6 7. Qc6 Qf1+ 8. Kg3 Qf2+ 9. Kh3 Qf6 10. Qe6+ Kh8 11. Qxf6+ Rxf6 12. Ne6 h5 13. Kg2 Kg8 14. Ng5 Rf4 15. Kg3 Rg4+ 16. Kf3 Kf8 17. Ke3 Ke7 18. Nf3 Kf6 19. Kd3 Rg3 20. Ke2 Rg2+ 21. Ke3 Ra2 22. Ne1 Rh2 23. Nf3 Rg2 24. Ng5 Ra2 25. Kd3 Rxa3+ 26. Kc2 Ke7 27. Ne6 Re3 28. Ng5 Rg3 29. Ne6 Rg4 30. Nc5 Rxh4 31. Nxa6 Rxe4 32. Nc5 Rc4+ 33. Kd3 Kd6 34. Ke2 Rxc5 (depth 51, 0:18:29)