Free Proportion Calculator

The first real trap is the second dimension. A six-by-four photograph is 3:2. Put it on a thirty-by-twenty-four canvas — a standard size, and one that looks close enough — and the width scales by five while the height scales by six. On a portrait, where a viewer knows the proportions of a face, the result is a subtly stretched image that reads as wrong without anyone being able to say why. The proportional transfer overlay is the same arithmetic drawn on the surface, and the artist guide overlays cover the rest of the transfer family. The fix is to compute the height from the ratio rather than choosing a stock size and hoping, and then either accept a margin or crop the reference deliberately.

The second trap is measuring the wrong edges. A photograph's printed area and its paper are not the same rectangle, and a canvas's stated size is the stretcher bar, not the visible face — a gallery-wrapped canvas loses several millimetres to the wrap on every side. Both errors are small in absolute terms and both are multiplied by the scale factor, which is exactly the mechanism that turns a two-millimetre measurement error into a centimetre of drift at the far edge.

The third is the one this calculator exists for. In the grid method both surfaces are divided into the SAME number of cells, and the cells are different sizes on each. That is the whole trick: proportions are preserved automatically because every cell on the reference maps to exactly one cell on the surface, so you never measure a distance at all — you copy what is inside a small square into a bigger square. Knowing the scale factor does not tell you how far apart to draw the lines; knowing that one cell is 25mm on the photo and 127mm on the canvas does.

How many cells is a genuine judgement rather than a rule. For a given amount of care, more cells means more reference points and a closer copy, and also more time spent drawing and rubbing out lines. The mural scaling calculator takes the same numbers up to wall size, and the canvas size calculator works the other way, from a ratio to a stretcher. Portrait work with a lot of information in a small area — an eye, a mouth — usually wants a finer grid over that region than over a plain background, which is what the half-grid subdivision in the overlay tool is for: keep a coarse grid overall and halve it only where the detail is.

The method is old and its lineage is documented rather than folkloric. Leon Battista Alberti described a gridded veil for exactly this purpose in De Pictura in 1435, and Albrecht Dürer engraved artists using one in 1525. What has changed is not the geometry but the arithmetic: they were dividing by hand, and a calculator removes the one part of the process where a slip propagates into every cell that follows.

One more practical point about walls, because murals are where the numbers get large enough to hurt. A scale factor of thirty is normal for a sketch going onto a building, and at that factor a one-millimetre error in measuring the sketch becomes three centimetres on the wall. It is worth measuring the reference generously — in centimetres rather than millimetres, on the largest print available — precisely because the error is multiplied along with everything else.

And a note on units. There is no reason for both sizes to be in the same one; a reference measured in inches and a wall measured in metres is the normal case for anyone working from printed material in a country that builds in metric. Converting through a single base unit rather than round-tripping between two is what keeps that exact, and it is why pixels are offered alongside physical units — a digital reference has no physical size until you give it one.

It is worth separating this calculation from the one people often think they are doing, which is resolution. Scaling a PHYSICAL drawing has no upper limit — a six-inch sketch can go onto a six-metre wall, because the pencil supplies the detail as you go. Scaling a DIGITAL image does have one, because a 600-pixel-wide photograph enlarged to poster size has 600 pixels of information no matter how large you print it. The proportion arithmetic is identical in both cases; what differs is whether the extra size comes with extra information. If you are working from a phone photograph onto a large canvas, the calculator will happily give you a scale factor of twenty, and the reference will be soft long before you get there. That is an argument for a better reference, not for different arithmetic.

There is also a quiet decision hidden in the phrase "the size you are copying onto". For a stretched canvas, the honest number is the visible face, not the stated size — a gallery wrap takes several millimetres from each edge and a drawing planned to the stated size runs off around the corner. For paper, it is the image area you intend rather than the sheet, since almost nobody draws to the very edge. Both of those are a subtraction you make once, before the numbers go in, and both are the sort of thing that is obvious afterwards and invisible beforehand.

Finally, the cell count and the scale factor are independent choices and it is worth noticing that they are. The scale factor is fixed the moment the two sizes are decided; the cell count is entirely yours, and it trades drawing time against accuracy. A common working pattern is to pick the cell count so that the cell size on the SURFACE lands on a round number you can mark quickly — five centimetres, or six inches — because the grid on the big surface is the one you have to draw by hand, and marking 4.7cm intervals across a wall is how errors accumulate.