Laser Maze

Bend the beam onto every sensor

A fixed emitter fires a single laser across the grid, and your job is to route that one beam so it passes through every ringed sensor on the board. You do it with mirrors. Tapping a mirror flips it between its two diagonal tilts, and each tilt kicks the beam ninety degrees one way or the other. Some mirrors are already fixed in place; others you carry as spares and drop into empty slots yourself, tapping to rotate them and tapping once more to pick one back up.

The best part is that the beam re-traces the instant you change anything, so feedback is immediate. There is no commit button and no penalty for experimenting: flip a mirror, watch where the light goes now, adjust again. Fiddling is free. What costs you time is fiddling without a plan. Two or three flips per idea is plenty; if you have flipped ten mirrors without a theory, stop and read the board again.

Think in rows and columns

Here is the observation that turns random flipping into actual solving: the beam only ever travels along rows and columns. Every mirror turns it exactly ninety degrees, so the whole path is a zigzag of horizontal and vertical runs. A sensor gets lit when one of those runs passes through it, which means every sensor on the board must share a row or a column with some segment of your beam.

That gives you a quiet checklist before you touch anything. For each sensor, ask which row or column the beam would have to travel along to reach it. Two sensors in the same column are a huge hint, because one vertical run can light both; the puzzle is really asking you to steer the beam into that column and let it fly. A sensor that lines up with nothing else is the expensive kind, demanding its own dedicated turn, and turns are exactly what your mirrors provide. Count the turns the route needs, count the mirrors you have, and the shape of the solution often appears before you have placed a thing. On dense boards it helps to trace the route with a finger before flipping anything, emitter to first corner to second, naming the mirror that makes each turn.

How to solve one

• Start from the emitter and follow the beam's current path with your eye, then find the first mirror that could redirect it toward a sensor you have not lit.
• Work out which sensors are hardest to reach and solve those first; the easy ones often fall into place around them.
• Remember a single beam has to hit them all in one continuous path, so think about the order the light visits the sensors, not just each sensor alone.
• Use your spare mirrors to make a turn the fixed ones cannot. A well-placed spare is often the whole solution.

Every level is checked to be solvable before you ever see it, so if you are stuck the answer is genuinely there. You have not been handed an impossible board, only one you have not cracked yet. When the last sensor lights, the level clears and a trickier grid takes its place, usually with more sensors or a longer route to plan around.

A puzzle with a pedigree

Mirror-and-beam puzzles have been around longer than you would guess. Home computer players were spinning mirrors in a game called Deflektor back in 1987, steering a beam through mazes on 8-bit machines. In 2005 the idea landed on the tabletop as Khet, a chess-like board game invented by a Tulane University engineering professor and his students, where the pieces carry real mirrors and each turn ends with an actual low-power laser fired down the board.

The physics being borrowed is the law of reflection: light leaves a mirror at the same angle it arrived. Tilt a mirror at forty-five degrees to the beam, as every mirror here is, and that law bends the light through a clean right angle, which is why the grid stays so readable. Nothing curves and nothing is random. So when a board has you stuck, trust the geometry. Follow the beam out from the emitter and find the first turn that sends light somewhere useless. That mirror, more often than not, is the one that is lying to you.