Free Crosshair Grid Generator

A crosshair grid keeps the crossings and drops most of the lines. At each interior lattice point it draws a small cross — two short strokes at right angles — and nothing in between. Position information intact; image visible.

The size of the cross is the whole design of the overlay, and it is a continuous dial rather than a choice. At the minimum of 16% of a cell width, the marks are isolated ticks and the grid reads as a scattering of registration points. At the maximum of 90%, the arms of neighbouring crosses nearly touch and the field reads as a full grid interrupted by small gaps.

Everything useful is in between, and where in between depends entirely on your subject. A high-contrast graphic tolerates a big cross; a subtle tonal photograph wants the smallest mark you can still see.

The marks sit at interior points only. The construction starts one full spacing in from the top-left and stops before reaching the opposite edges, so nothing lands on the frame boundary. At a density of 14 on a square canvas that produces a thirteen-by-thirteen array — 169 marks.

Keeping the boundary clean is a deliberate decision rather than an accident of the loop. Marks sitting on the edge of a frame are ambiguous — you cannot tell whether they belong to the image or the border — and if the image will be cropped, they are the first thing lost.

The technique predates any of this by a long way. Registration marks in printing go back to the beginnings of multi-colour work: each plate carries the same marks, and the press operator aligns them so that the colours land on top of each other. Crop marks, colour bars and the corner crosses on a printer's proof are all the same idea.

In photography, the fiducial marks etched into the focal plane of aerial survey cameras serve the same purpose — they let a later measurement establish exactly where the film sat, and therefore exactly what geometry the image has. Photogrammetry depends on them entirely.

In animation, peg registration solves the problem physically: every sheet of paper has the same holes and sits on the same pegs, so drawings stack in exact alignment. Where pegs are impractical — on a background painting, on a piece of glass — pencilled crosses do the job instead.

The common thread is that all of these use marks rather than lines, and all of them do so for the same reason: a mark says here without saying anything about what lies between here and the next mark. That silence is the feature.

For artists the most common application is a portrait or figure reference where a continuous grid genuinely gets in the way. Faces are read through very small tonal transitions, and a grid line running across an eyelid or a nostril changes what you see there. A cross a centimetre away does not.

The same holds for any subject where texture carries the information — fur, fabric, water, foliage. A line across it competes with the texture at the same visual frequency; a mark does not.

For copy work and reproduction, the crosshair grid is the natural fit because the job is genuinely registration rather than composition. You are not deciding where anything goes; you are establishing correspondence between two surfaces, and marks are what correspondence is made of.

There is a limit worth naming. At low densities — below about eight — the marks are so far apart that you lose the ability to follow a row across the picture, and a continuous grid genuinely is better. The crosshair grid earns its keep from about ten upward, where the marks are close enough that the eye connects them without help.

At the top of the density range, 36 divisions of the short side, the marks become small enough that the grid approaches a dot screen. That is useful for very precise transfer of small work and impractical for anything drawn at arm's length.

The crosshair overlay page covers its uses, beside the rest of the specialty grid overlays. For a ruled grid rather than marks the rectangular grid maker is the tool, and the dot grid generator is the gentlest of the three. Setting the density is the same calculation as for any transfer grid: work back from the smallest feature you need to position. If the feature is smaller than a cell, the grid cannot tell you where it goes, and you are estimating.

A practical trick for photographs with a dominant subject: set the density by what the subject needs, not by what the background needs. Backgrounds tolerate coarse grids because errors there are invisible; faces and hands do not.

For printing, the crosshair grid has a real advantage in ink and in legibility. Fine continuous lines at low opacity often break up when printed, appearing as dashes; short marks at a slightly higher opacity print cleanly and read better on paper than the equivalent line grid.

The overlay draws over your own image rather than exporting a fixed template because placement is part of the registration. Which of the subject's features land near marks determines how useful the grid is, and shifting it half a cell can make a real difference on a subject with strong structure.

Export sizes itself from the paper format and the DPI you choose rather than from your screen. SVG keeps every mark as a path, which is what you want when the grid is a registration layer in a larger document.

Everything runs in your browser. Images you load are never uploaded, because there is no server for them to reach. The drawing, the transform and the export all happen on your own machine.