search

lichess-org / lila

♞ lichess.org: the forever free, adless and open source chess server ♞
https://lichess.org
GNU Affero General Public License v3.0
15.3k stars 2.25k forks source link

Obvious non-blunder called blunder because "Checkmate is now unavoidable" #15616

Closed xehpuk closed 10 hours ago

xehpuk commented 3 months ago

Exact URL of where the bug happened

https://lichess.org/f5uklQCg/white#66

Steps to reproduce the bug

Play this game:

1. e4 e5 2. Nf3 Nc6 3. Bc4 Bc5 4. c3 Bb6 5. d4 Qe7 6. d5 Nb8 7. a4 a6 8. O-O d6 9. Bg5 f6 10. Be3 Bg4 11. Bxb6 cxb6 12. Nbd2 Nd7 13. h3 Bh5 14. Re1 Nh6 15. Nf1 O-O 16. Ng3 Bf7 17. Nh4 g6 18. Qd2 Kg7 19. Qxh6+ Kxh6 20. Nhf5+ gxf5 21. Nxf5+ Kg5 22. Nxe7 Bg6 23. Bd3 Nc5 24. Bc2 Rf7 25. Nf5 Bxf5 26. exf5 Rg8 27. b4 Nd7 28. Re4 Rfg7 29. Rg4+ Kh5 30. f3 Rxg4 31. hxg4+ Kg5 32. Kf2 Rc8 33. Rh1 e4 34. Kg3 Rxc3 35. Rh5# 1-0

Request computer analysis.

What did you expect to happen?

33...e4 not to be called a blunder. The computer variation doesn't improve the outcome. Still mate in two.

What happened instead?

33...e4??

Operating system

-

Browser and version (or alternate access method)

-

Additional information

No response

Carbrex commented 3 months ago

Same as #14696

xehpuk commented 3 months ago

@Carbrex Reloading doesn't fix it for me, but opening the game in private mode does. 🤔

Carbrex commented 3 months ago

Screencast

See the 3 blunders shown on the graph but only 1 shown in the text at the right. I have managed to reliably reproduce the issue by using the method as mentioned in this comment https://github.com/lichess-org/lila/issues/14696#issuecomment-2013201672 .