Not working training studies. • page 1/1 • Lichess Feedback • lichess.org

KupferUndNickel

However a human player knowing this endgame would most probably not try this approach, but chose the more resiliant approach of defending their pawn - trying to get white to make a mistake and stalemate him.
Therefor it it kind of pointless to do the exercise, as you do not get to train the proper technique.

I was wondering if it were possible to script these parts or promt Stockfish to make the more resiliant moves - even if they lead to a faster checkmate.
I know you guys do amazing work, with little financial ressources. But if you find the time - i think it would be very helpful for many players, to fix these training parts.
The easyest way i think would be to just copy the scripted part and delete the arrows and marks guiding the player - so the player would have to execute the strategys by memory.

Hey - i really like the training-section an use it to try and improve my endgames. Hovever, when i try to practice theoretical endgames it can be quite frustrating as the engine doesnt chose the most resiliant defense, as it already knows the outcome of the endgame. If you for example practice Queen vs Rook pawn with King assistance on short side, the engine knows that the most resiliant defense leads to a checkmate by perfect play from white - therefor it abandons the pawn and run's their king to avoid checkmate for as long as possible. However a human player knowing this endgame would most probably not try this approach, but chose the more resiliant approach of defending their pawn - trying to get white to make a mistake and stalemate him. Therefor it it kind of pointless to do the exercise, as you do not get to train the proper technique. I was wondering if it were possible to script these parts or promt Stockfish to make the more resiliant moves - even if they lead to a faster checkmate. I know you guys do amazing work, with little financial ressources. But if you find the time - i think it would be very helpful for many players, to fix these training parts. The easyest way i think would be to just copy the scripted part and delete the arrows and marks guiding the player - so the player would have to execute the strategys by memory.

Royals_Fan22

@KupferUndNickel said ^

Hey - i really like the training-section an use it to try and improve my endgames.
Hovever, when i try to practice theoretical endgames it can be quite frustrating as the engine doesnt chose the most resiliant defense, as it already knows the outcome of the endgame.
If you for example practice Queen vs Rook pawn with King assistance on short side, the engine knows that the most resiliant defense...

Well, Stockfish can't just be changed to a script, but there is an engine called Maia designed to play at a human's skill level, and also different rating levels. @maia5 @maia1 @maia9 I haven't tried endgames agianst these, but it might work.

@KupferUndNickel said [^](/forum/redirect/post/pgZRkk1i) > Hey - i really like the training-section an use it to try and improve my endgames. > Hovever, when i try to practice theoretical endgames it can be quite frustrating as the engine doesnt chose the most resiliant defense, as it already knows the outcome of the endgame. > If you for example practice Queen vs Rook pawn with King assistance on short side, the engine knows that the most resiliant defense... Well, Stockfish can't just be changed to a script, but there is an engine called Maia designed to play at a human's skill level, and also different rating levels. @maia5 @maia1 @maia9 I haven't tried endgames agianst these, but it might work.