0

In the position shown in the image, Stockfish (via the lichess interface) shows a top line of 91 ... Rxe5 92. Bxe5 Kxe5 93. Kc5 Kf5. 94 Kd5 Kf4 and so on. It reports an evaluation of +0.1 for the line.

FEN: 8/8/4k3/4P1r1/2KB4/8/5N2/8 b - - 18 91

After Black's 92 ... Kxe5 the position is drawn due to insufficient material. What does the evaluation +0.1 mean in that context? (And why does Stockfish continue the line beyond that point?) Does the evaluation shown result not only from the displayed line but somehow also the alternatives that White has to 92. Bxe5 ? Obviously it's not a question of any practical importance, but it's curious not to see 0.0 when the line is literally a draw.

Improve this question
4
  • 1
    I have Stockfish 15 and 15.1 engines (with NNE) installed on my PC. Both immediately give an evaluation of +0.0, with and without tables for finales.
    – Damien
    Jul 6 at 18:32
  • 1
    Disabling NNE doesn't make any difference ...
    – Damien
    Jul 6 at 18:48
  • @Damien OK thanks! So what does this suggest? A difference between Stockfish 14 and 15? (of what nature?) Or something that involves the way Stockfish is embedded into the lichess interface?
    – James Martin
    Jul 7 at 8:24
  • 1
    In your diagram, Lichess indicates a search depth of 22. I made a test on an old PC with SF13 installed. It needed to go up to a depth of 28 to get an evaluation of 0.0. I also went to Lichess (SF14) and made same analysis again. It did not show the good result at a depth of 55. However, by forcing a deeper analysis, it finally provided the correct 0.0 evaluation. It seems that SF doesn't know the rule of unsufficient material and needs a higher depth to be convinced that this is a draw.
    – Damien
    Jul 7 at 13:58

0

Reset to default

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Browse other questions tagged or ask your own question.