Please implement null moves

One must be careful with the threat button. It may not do exactly what is expected, depending on what is expected.

https://lichess.org/analysis/fromPosition/8/8/1b6/8/8/3k4/5r2/4K3_w_--97_1 with threat button: #-2
https://lichess.org/analysis/fromPosition/8/8/1b6/8/8/3k4/5r2/4K3_b--_98_1 : +0.0

One must be careful with the threat button. It may not do exactly what is expected, depending on what is expected. https://lichess.org/analysis/fromPosition/8/8/1b6/8/8/3k4/5r2/4K3_w_-_-_97_1 with threat button: #-2 https://lichess.org/analysis/fromPosition/8/8/1b6/8/8/3k4/5r2/4K3_b_-_-_98_1 : +0.0

mkubecek

#12

@corvusmellori said in #11:

lichess.org/analysis/fromPosition/8/8/1b6/8/8/3k4/5r2/4K3_b_-_-_98_1 : +0.0
Looks like a cached bogus value. Usually this can be seen when you let the engine run long enough. But here the position after Bc5 is shown as "0.0" with depth 99, even if it also shows Kd1 as the only legal reply (with 0.0) but the position after Kd1 is correctly evaluated as "#-1".

In any case, this is definitely not a problem of the "show threat" function as what "show threat" tells you is correct. It's the normal analysis what is wrong.

@corvusmellori said in #11: > lichess.org/analysis/fromPosition/8/8/1b6/8/8/3k4/5r2/4K3_b_-_-_98_1 : +0.0 Looks like a cached bogus value. Usually this can be seen when you let the engine run long enough. But here the position after Bc5 is shown as "0.0" with depth 99, even if it also shows Kd1 as the only legal reply (with 0.0) but the position after Kd1 is correctly evaluated as "#-1". In any case, this is definitely not a problem of the "show threat" function as what "show threat" tells you is correct. It's the normal analysis what is wrong.

nadjarostowa

#13

Haha, you got me there... took me a moment to find the reason for the difference.

The draw evaluation is perfectly correct.

It is debatable if the threat evaluation with #2 is correct or not. I think it is a clear indication that at the moment there is no null move inserted, but instead simply switched the colour flag, without incrementing the ply counter.

Haha, you got me there... took me a moment to find the reason for the difference. The draw evaluation is perfectly correct. It is debatable if the threat evaluation with #2 is correct or not. I think it is a clear indication that at the moment there is no null move inserted, but instead simply switched the colour flag, without incrementing the ply counter.

mkubecek

#14

Ah, OK. Not sure why the analysis board takes the move count into account for evaluation but still shows the moves like 1... Bc5.

corvusmellori

#15

@mkubecek said in #14:

Not sure why the analysis board takes the move count into account for evaluation but still shows the moves like 1... Bc5.
Because the FEN I set had the move counter set to 1. You could change that by changing the move counter.
lichess.org/analysis/fromPosition/8/8/1b6/8/8/3k4/5r2/4K3_b_-_-_98_130

@mkubecek said in #14: > Not sure why the analysis board takes the move count into account for evaluation but still shows the moves like 1... Bc5. Because the FEN I set had the move counter set to 1. You could change that by changing the move counter. lichess.org/analysis/fromPosition/8/8/1b6/8/8/3k4/5r2/4K3_b_-_-_98_130

mkubecek

#16

Then I guess the most weird thing is that it is allowed to make the two counters inconsistent this way.

corvusmellori

#17

@mkubecek said in #16:

Then I guess the most weird thing is that it is allowed to make the two counters inconsistent this way.
I don't see why it should care. I can set impossible-to-reach positions in the FEN as well. There wouldn't be a lot of benefit in the lichess analysis board refusing to analyze positions with impossible combinations of 50-move-counter and full-move-counter; if someone has chosen such a FEN, presumably she doesn't care about the inconsistency.

@mkubecek said in #16: > Then I guess the most weird thing is that it is allowed to make the two counters inconsistent this way. I don't see why it should care. I can set impossible-to-reach positions in the FEN as well. There wouldn't be a lot of benefit in the lichess analysis board refusing to analyze positions with impossible combinations of 50-move-counter and full-move-counter; if someone has chosen such a FEN, presumably she doesn't care about the inconsistency.

This topic has been archived and can no longer be replied to.