lichess.org
Donate

The moves Kf1 and Kf8 are mouseslips 100% of the time

The whole move input system should be governed by an AI that plays the most likely move (given your clicks and mouse movements and the position), or aborts if the likelihood of any single move is less than x%, where x should be a user-adjustable parameter.
I'm pretty sure that an AI assisting based upon the position would violate ToS.
Well, just your pieces near where you clicked and their legal destinations near where you let go (or made a second click) could be enough.
Just make 2 buttons with o-o and o-o-o below the board

Button is visible when
1) you are playing the game
2) castling is allowed

Helps in 960 as well :)

#33 In a time scramble, a player may not understand which moves are legal.

#34 I actually like this idea. Note that currently it is possible to enable keyboard move entry, so it's possible for someone to extend that concept to add buttons (although I don't know how you'd disable them for illegal moves, or whether they should ever be disabled).
IDK maybe we can have a pop-up? But no, not confirmation. That would be mad if we're running out of time.

But perhaps we could have an option in the settings to turn it on or off?

I don't know if the devs will incorporate this feature but if enough people think it's a good idea it might be added.
It's not a horrible idea, but don't say Kf1/f8 is a mouseslip 100% of the time, that's definitely not true.
#36
Okay, so make it choose from pseudolegal moves. Only assume that the player knows how the pieces move.

I'm thinking stuff like "just throw your mouse in the right general direction to move that knight" or rejecting moves where the mouse is too close to a different candidate square (say, barely Ba6 instead of the Ruy Lopez, or 1. e3.9, or clearly moving a pawn diagonally but not quite enough to turn it into a capture)
Kf1/f8 mouse slip originates from the early internet chess sites such as ICC and FICS.

As i recall from my peak time on ICC during the mid 2000s, this tend to recur in approximately 5% of blitz and bullet games, more frequently in bullet games obviously. Nevertheless as a matter of fact playing Kf1 when intended to castle cause positional problems most of the times it happen.

100% is an exaggeration, i'd say 3-5% is more realistic frequency.

This topic has been archived and can no longer be replied to.