In Ludo Ghar, landing on an opponent captures it to base, except on safe squares. The hypothetical example below separates the destination from intermediate squares; it does not describe every Ludo edition.
Capture depends on the final square
Imagine six ordinary board positions labelled A, B, C, D, E and F in forward order. Your token starts at A, while an opponent’s token is at D. There is no blockade, and the token at D belongs to an opponent rather than a partner.
With a roll of three, your token travels A to B, then C, then D. Because D is the destination and is an ordinary square occupied by the opponent, the move captures that token. Record your token at D and the opponent in its starting yard.
Passing is only movement through visited squares
Now suppose the roll is four. The route is A to B, C, D and finally E. D is visited during the move, but E is the final square. Since the token does not end on D, this move is passing, not capturing. The opponent’s token remains on D, while your token ends on E.
This distinction is easiest to check by recording the position before the roll and the position after it. Intermediate squares describe the route; the final square determines whether an ordinary-square capture can occur.
Compare two separate records
For the three-step trial, write: mover D, opponent base. For the four-step trial, write: mover E, opponent D. Reset the original A-and-D arrangement before each trial. Otherwise the second trial would mistakenly start from the first trial's final state rather than test a different roll against the same position.
A quick reading test
Ask three questions after any move: Where did the token start? Which squares did it pass through? Where did it finish? Only then check whether the final square was occupied, empty, ordinary or safe. This prevents treating every crossing of an opponent’s location as a capture.