3 ms·

Has anyone actually checked that a modern chess engine believes black is winning in the Penrose position? I find it very hard to believe.

shric 5y ago
Stockfish 12 (280920) on my phone rates black as up 19.71
perihelions 5y ago
Lichess cloud compute has that position. It still thinks black is winning at depth 59 (Stockfish 11 HCE). https://lichess.org/analysis/8/p7/kpP5/qrp1b3/rpP2b2/pP4b1/P3K3/8_w_-_-_0_1 https://lichess.org/analysis/8/p7/kpP5/qrp1b3/rpP2b2/pP4b1/P...
- shric 5y ago
  Indeed, and same with Stockfish 13 running locally.
dwohnitmok 5y ago
It looks like Stockfish still thinks black is winning. I'd be curious to see if Leela Chess Zero also thinks black is winning.
- gweinberg 5y ago
  Is that 59 ply? 59 moves without a pawn move or capture means the game was drawn nine moves ago. Here's what's confusing me: if the chess engine has an initial numeric value for the current position, and the positions going out to the end of its search all have the same value or worse, shouldn't it be pretty trivial to conclude it is not making progress?
  - shric 5y ago
    Yes, I guess Stockfish doesn't bother to consider a draw on the 50 move rule. It does, iirc, go for a draw if it has no better via threefold repetition.
    - perihelions 5y ago
      Stockfish does plan around the 50-move rule. Here's a quick demo (it is silly): https://lichess.org/analysis/8/1p3K2/kPp5/2P5/2P5/1p6/1P1Q4/8_w_-_-_98_1 https://lichess.org/analysis/8/1p3K2/kPp5/2P5/2P5/1p6/1P1Q4/... It's not visible on the board itself, but this position's 50-move rule counter is set to 98 ply -- two ply short of a draw. (That's the 2nd-to-last-field in the FEN representation -- the editable text field just below the board, and also in the URL). The only way to avoid that draw, as Stockfish does in fact suggest, is to a sacrifice a piece in a specific and ridiculous way, to compel black to re-capture on the 100th ply (captures reset the 50-move rule). If you edit the 50-move counter down, Stockfish instead shows you the simple mate-in-3. Nakamura famously tricked a different chess engine, relying on this 50-move draw avoidance (as one component of the hack) -- a sibling thread is discussing that example: https://news.ycombinator.com/item?id=27189718 https://news.ycombinator.com/item?id=27189718

shric 5y ago

Stockfish 12 (280920) on my phone rates black as up 19.71

perihelions 5y ago

Lichess cloud compute has that position. It still thinks black is winning at depth 59 (Stockfish 11 HCE). https://lichess.org/analysis/8/p7/kpP5/qrp1b3/rpP2b2/pP4b1/P3K3/8_w_-_-_0_1 https://lichess.org/analysis/8/p7/kpP5/qrp1b3/rpP2b2/pP4b1/P...

shric 5y ago
Indeed, and same with Stockfish 13 running locally.

dwohnitmok 5y ago

It looks like Stockfish still thinks black is winning. I'd be curious to see if Leela Chess Zero also thinks black is winning.

gweinberg 5y ago
Is that 59 ply? 59 moves without a pawn move or capture means the game was drawn nine moves ago. Here's what's confusing me: if the chess engine has an initial numeric value for the current position, and the positions going out to the end of its search all have the same value or worse, shouldn't it be pretty trivial to conclude it is not making progress?
- shric 5y ago
  Yes, I guess Stockfish doesn't bother to consider a draw on the 50 move rule. It does, iirc, go for a draw if it has no better via threefold repetition.
  - perihelions 5y ago
    Stockfish does plan around the 50-move rule. Here's a quick demo (it is silly): https://lichess.org/analysis/8/1p3K2/kPp5/2P5/2P5/1p6/1P1Q4/8_w_-_-_98_1 https://lichess.org/analysis/8/1p3K2/kPp5/2P5/2P5/1p6/1P1Q4/... It's not visible on the board itself, but this position's 50-move rule counter is set to 98 ply -- two ply short of a draw. (That's the 2nd-to-last-field in the FEN representation -- the editable text field just below the board, and also in the URL). The only way to avoid that draw, as Stockfish does in fact suggest, is to a sacrifice a piece in a specific and ridiculous way, to compel black to re-capture on the 100th ply (captures reset the 50-move rule). If you edit the 50-move counter down, Stockfish instead shows you the simple mate-in-3. Nakamura famously tricked a different chess engine, relying on this 50-move draw avoidance (as one component of the hack) -- a sibling thread is discussing that example: https://news.ycombinator.com/item?id=27189718 https://news.ycombinator.com/item?id=27189718