How do slash and backslash mirrors turn the ray?

A slash mirror maps right to up, up to right, left to down, and down to left. A backslash mirror maps right to down, down to right, left to up, and up to left.

What happens if the ray enters a loop?

The tracer stops when the same cell and incoming direction would repeat, labels the outcome as a loop, and lets you change the layout unless it was your second distinct miss.

Is this a full physics or optics simulator?

No. It is a deterministic grid puzzle using ideal cardinal rays and 45-degree mirrors; it does not model materials, wavelength, refraction, intensity, or physical measurements.

Why does placement feel harder on later levels?

Each later level adds one more mirror to the route. By the fifth level you must hold a four-turn path in mind and keep stray mirrors outside the active path from confusing the trace.

Place Mirrors in a Ray Grid: A Placement Walkthrough

A mirror placement puzzle is a logic grid where you drop slash and backslash tiles into empty cells so that a straight ray bounces its way from a marked entry to a single OUT edge. In the five-by-five Mirror Reflection Puzzle, every level hands you a fixed starting cell, a required exit side, and an empty board, and your only decision is which diagonal mirror to place in which cell. Empty cells let the ray pass straight through, slash mirrors turn it ninety degrees one way, and backslash mirrors turn it the other way. Because the eight turn combinations follow the law of reflection — the angle in equals the angle out, measured from the diagonal surface normal — you can predict every bounce before tracing. The goal across all five levels is to land the ray on the exact OUT edge, not just any boundary, and each level adds one more mirror to the route so placement thinking has to stretch further. Done carefully, all five grids award a perfect 1,000-point run.

place mirrors puzzle — Place Mirrors in a Ray Grid: A Placement Walkthrough

The five-by-five grid you'll be placing mirrors on

The board is exactly twenty-five cells laid out as five rows by five columns. One cell on the edge is marked as the entry point, and one full side of the grid is flagged as the OUT target for that level. Every other edge is treated as a wrong exit that ends a trace as a failed attempt. You place mirrors by selecting a cell and choosing a slash, a backslash, or leaving the cell empty. Tap or click a cell to cycle it through empty, slash, backslash, and back to empty, use the arrow keys to move the selection, type the matching character to set the tile, press 0 to clear, and press Enter to trace. Cells are large enough to be touch-friendly and carry accessible labels so you always know what you have placed. The full board, the trace reducer, and the score live in your browser — there is no server calculation, no third-party image, and no random seed.

The slash and backslash turn rules

The placement rule is fixed and auditable. The ray always travels along one of the four cardinal directions — right, up, left, or down — and a mirror rotates it to the perpendicular direction. The table below shows every turn mapping you can rely on when deciding where to place each tile.

Incoming direction	Slash (/) turns ray to	Backslash (\) turns ray to
Right	Up	Down
Up	Right	Left
Left	Down	Up
Down	Left	Right

Slash and backslash are mirror opposites: if a slash turns a right-traveling ray up, a backslash turns the same ray down. The two tile types together cover all four ninety-degree exits. With this mapping memorized, you can predict any cell's effect before tracing, which makes placement decisions much faster.

How to place mirrors on a ray puzzle grid

Open the level and locate the marked entry cell and the OUT side flagged for that board.
Select a cell along the most likely route, then place a slash or backslash with the orientation buttons, by tapping, or by typing the matching key.
Press 0 to clear a cell back to empty whenever a tile is no longer helping the route.
Press Enter or choose Trace ray to run the simulation and read the highlighted path cell by cell.
If the ray loops or exits through the wrong edge, change only the mirror cells the trace actually visited, then retrace.
When the ray exits through the exact OUT side, the level clears and you move to the next board.

Work backward from the OUT edge

A useful placement habit is to start from the OUT side and work backward. Ask which direction the ray must be traveling when it lands in the target cell, then choose the mirror on the cell just before the exit so that it produces exactly that incoming direction. Continue backward one cell at a time, picking the mirror that maps the previous direction into the new one. This reverse path is a guaranteed plan: the forward trace will follow it as long as no other mirror interferes.

Because the grid only has five rows and five columns, you can usually sketch the entire route on paper in under a minute, then move forward through the board and place each mirror exactly where the reverse plan predicted. On longer levels, place only the turns you can justify and trace early. The highlighted path will show the first place your prediction diverged, which is much faster than guessing and rebuilding from scratch.

Reading the trace after each placement

After you trace, the puzzle highlights the cell-by-cell route and writes the outgoing direction on every visited cell, so you can see exactly what each placement did. A wrong-edge trace stops the moment the ray leaves the grid through a non-target side, which means the last highlighted cell is the one whose mirror turned the ray the wrong way, or the empty cell that let the ray continue past where it should have turned. A loop trace stops at the cell where the same incoming direction would repeat, which usually means a small mirror cluster is bouncing the ray between two cells forever.

Because mirrors outside the active path do not change the ray, keeping the board sparse also makes the intended route easier to audit. Many placement runs fail because the player forgets an old mirror that the new trace no longer visits, so clearing those stray tiles before each retrace saves time.

How placement difficulty grows across the five levels

The five fixed boards progress on purpose. The first level asks for a single mirror and one turn from entry to OUT, the second adds a second mirror so the route has two turns, the third adds another, the fourth another, and the fifth asks for a four-mirror path through the grid. None of the layouts shuffle between runs, so the placement thinking is reproducible: once you know the route for a level, you can place those exact mirrors on every fresh attempt.

This slow ramp means you can practice placement intuition on level one before the grids get crowded. By the time you reach level five, the same habit — work backward, place one justified turn, trace, adjust — covers the longer path without adding hidden rules, random shapes, or new controls.

Scoring and what stops a placement run

Every solved level awards exactly 200 points, and all five levels together produce a perfect score of 1,000. A failed trace counts only when the current mirror layout is distinct from the previous failed layout; checking the same wrong layout again does not consume another attempt. A second different failed layout ends the run, while Restart restores level one, an empty placement map, zero mistakes, and zero points. Completed and failed states are frozen, so extra clicks or key presses cannot alter the recorded result.

The placement decisions you make are the only input that matters. There is no timer, no random seed, no network calculation, and no third-party image, so the entire board and trace reducer run locally in the browser. The eight turn mappings can be audited exactly because the equal-angle rule that anchors them matches the law of reflection rather than a custom rule, which is why a single memorized table is enough to predict every bounce.

A practical placement routine

In short, place mirrors by working backward from the OUT side, commit one justified turn at a time, retrace after each change, and clear stray tiles the new trace no longer visits. The five fixed levels teach this habit incrementally, and the deterministic scoring means your placement work shows up directly in the final 1,000-point total.