@Kingscrusher-YouTube #20
"Much more likely in the current rules of lichess, is for someone with basically "insufficient mating material" of a knight or bishop left only to aimlessly move them around (especially with premoves which are not btw in the FIDE handbook), and to win on time with that single knight or bishop. "
True enough, but I don't see that as being qualitatively different from playing on when down a huge pile of material but still having a pawn.
PS: Love your videos!
This is where the inconsistency arises.
To say "A single pawn does count as 'sufficient mating material' - the pawn could technically queen.", which could likely be on it's first rank against an armada of opponent pieces could be a win, but to say "Just a single knight one move away from checkmate is not sufficient mating material" is contradictory.
All of these arguments stem from 'potential', does player A have the 'potential' to win.
More thought needs to be put into this. I think that in the 1 knight or 1 bishop scenario Stockfish could be elected to adjudicate the position for possible mates in N (with accurate play from both sides) for the side with only a bishop or only a knight.
Addressing point #28 - one more thing here to mention in the case of more material but very very slow progress being made ...
If one side has no clue how to mate say with B+N against King, the system should be aware of the 50 move draw rule. So if the other side presses "draw" - a check should be made of the 50 move draw rule - to again rule out cases where one side is not making "reasonable progress" to try and win the game, and has technically exceeded the 50 move draw rule.
Cheers, KC
Clarkey #32
If you are really worried about this case - (which i haven't seen in playing 15 years of blitz chess myself) , you could let the system check if there is an imemdiate mate in one from a single knight or bishop, and in that case declare the person a winner with the single knight or bishop if their opponent runs out of time.
The 50 move rule is already exercised automatically.
Addendum to point #34
Rather than worrying about a theoretical case which might happen once in 10,000 times, I would be more concerned with daily breach of FIDE rules with the winning on time with a single knight or bishop. This must happen on a daily basis with the volume of games being played. This reflects winning on time and not on position and in fact "no reasonable progress" often being made to win on time.
It is a case sometimes of "lesser evil" choices.
E.g. although Electronic Cigarettes are "Bad" because they have nicotine in them, they avoid all the other thousands of chemicals. So maybe the EU shouldn't try and regulate them too hard given that each year thousands of people actually die of cigarette smoking.
Where there is exists pratical risk, you can use a "lesser evil" approach to decide strategy - you chose something which may have a downside, but which is less than the alternative.
Cheers, KC
It's not that I don't see your point, KC. It's that I don't want to see these rules altered and have more issues arise. Chess is an infinite game, so the rules that are applied must stretch equally as far. Taking a best guess will not do, so an accurate adjudication process must take place in my opinion.
Actually if someone really did have a mate in one with a single knight or bishop left, I would suggest just checking their "premove" - and giving them the win, if the premove was a checkmate. So even if the other side ran out of time, the person with the knight or bishop, but with the intended premove mate declared the winner.
Premoves btw are part of online chess and not in the FIDE handbook. But you can use them as the qualifier if needed.
If you wanted more sophistication, then you could calculate the probability of a checkmate with that last knight or bishop. But I really don't think these cases would happen with any frequency compared to the current issue.
Maybe it's 1/1000000, but if it can happen, it must be accounted for. So my current opinion is: If player A has only a knight or a bishop and their opponent times out:
--If SF claims that player A can mate in N moves with accurate play from both sides, it's a win for player A.
--If SF claims that player A cannot win in N moves, the position is draw.
Does this with agree with you?
That would be amazing, but I would be concerned about degreadating the performance of the site if a load more CPU checks are needed for caculating the mating possibilities. If the user experience isn't damaged that much, then yes that would be ideal.
If the user experience is risked because the servers CPUs are frying to calculate mates in everyones blitz games where one side has a single knight or bishop left, then perhaps the simpler idea of the single knight or single bishop check along with the premove check if needed might be really good instead.
