Prism Paths

Routing beams in Prism Paths

Sources, mirrors, receivers

A source emits a beam of one colour in one fixed direction that you cannot change. The beam travels in a straight line until it meets a mirror, which turns it ninety degrees. A receiver absorbs a beam that reaches it, and it must be the matching colour.

Mirrors have exactly two orientations — the forward-slash and the backslash — and each simply flips between them when you interact with it. That is the only action in the game. Everything else about the board is fixed.

Levels run from a five-by-five grid upward and each carries a par: the number of rotations the intended solution requires. Some mirrors already start in their correct orientation, so par is well below the total number of mirrors on the board.

Trace backwards from the receiver

Beams are reversible, which makes working backwards the strongest technique available. Start at the receiver, note which edge a beam must enter through, and trace that line back across the grid until it meets a mirror. That mirror’s required orientation is now fixed. Repeat until you reach the source, and you have the full solution without a single trial rotation.

The second idea is to leave correctly-set mirrors alone. Some mirrors begin in their solution orientation, so touching them costs a rotation and then costs another to undo. Before flipping anything, trace the existing beam and note which mirrors it already handles properly.

On multi-colour boards, solve each beam independently first and only then check for conflicts. A mirror that lies on two beams’ paths is the crux of those levels: it can only take one orientation, so one of the two routes has to be redesigned around it. Finding that shared mirror early saves a lot of wasted flipping.

Par, and what counts against it

Par is the number of mirror rotations in the intended solution. A level with four mirrors might have a par of one or two, because the others already start correct — so par is a measure of how much has to change, not how large the board is.

Two failure states are worth recognising while you work. A beam that leaves the grid without reaching a receiver has escaped, and a beam that loops back into its own path has entered a cycle. Both mean the current mirror arrangement is wrong, and the game distinguishes them so you can tell an over-shoot from a loop at a glance.

Common mistakes

The pattern in all of these is acting before tracing.

• Flipping mirrors to see what happens. With only two orientations each, the board is small enough to solve on paper — trial and error mostly inflates your rotation count past par.
• Rotating mirrors that were already correct. Some start in their solution orientation, so touching them costs two rotations to get back to where you began.
• Solving forwards from the source. Tracing backwards from the receiver pins each mirror’s orientation deterministically, whereas forward search branches at every mirror.
• Treating a shared mirror as solvable twice. On multi-colour boards a mirror serving two beams can only take one orientation, so one of those routes must change.

What is different about this version

Every level ships with its solution orientations committed alongside its starting state, so par is a genuine measured optimum rather than an estimate. The puzzle is verified solvable at that par rather than assumed to be.

The beam tracer explicitly detects and reports cycles as well as escapes. A looping beam is a distinct outcome rather than a hang or an infinite render, which is what lets the game show you that your arrangement folds back on itself instead of simply failing to light the receiver.

Written by the Silverweb Games team from this game’s own implementation. Last reviewed July 31, 2026.