Aspect Ratio Calculator - Reduce, Identify and Scale Ratios

An aspect ratio states the relative width of a shape to its height, written as two numbers separated by a colon, and it describes the shape rather than the size. Writing width first is a convention that became dominant because screens and images are discussed in landscape, so 16:9 always means sixteen units wide for every nine units tall. The order matters: 16:9 and 9:16 describe genuinely different shapes, the first landscape and the second portrait, which is why video platforms ask you to state both when you upload the same footage. Because the ratio is scale-free, any two rectangles sharing a ratio are similar figures with identical angles, which is exactly what lets you scale a 3000 pixel photograph down to a 300 pixel thumbnail with no distortion of shape.

Find the greatest common divisor of the two numbers and divide both by it. For 1920 and 1080 the greatest common divisor is 120, since 1920 divided by 120 is 16 and 1080 divided by 120 is 9, giving 16:9. The quickest way to get the divisor is the Euclidean algorithm: repeatedly replace the larger number with the remainder of dividing the smaller into it. So 1920 divided by 1080 leaves remainder 840, then 1080 divided by 840 leaves 240, then 840 divided by 240 leaves 120, then 240 divided by 120 leaves zero, so the last non-zero value, 120, is the divisor. Reducing matters whenever you want to name a format rather than describe one particular resolution. It also exposes near misses: 1366 by 768 reduces to 683:384 because the divisor is only 2, so that screen is not truly 16:9 even though it looks close, and the 5 pixel difference is real.

Almost, but not exactly, and the distinction matters when you are writing code or specifying hardware. Sixteen ninths equals 1.7777777777777777 repeating, so 1.7778 is a rounded approximation that is off by about 0.0000222. For a 1080 pixel tall image that error is roughly 0.024 pixels, entirely invisible. For anything measured in points or millimetres it accumulates into a visible discrepancy over long edges. The safe practice is to keep the ratio in fractional form, 16 / 9, wherever the language or format allows it, since that form is exact by definition. Use the decimal when an external system demands a number, and use the reduced integer pair when a human needs to recognise the format. The percentage form, 177.78 percent, is simply the decimal multiplied by 100 and is the form most display specifications quote.

Fix the dimension you control and let the other follow from the ratio. If your source is 3000 by 2000 and you need a width of 1280, the scale factor is 1280 divided by 3000, which is 0.42667, and the height becomes 2000 multiplied by that factor, giving 853.33. In this calculator you type the required width into the Target width field and the matching height and scale factor appear immediately. The same logic works in reverse when height is the constraint. Two practical warnings. First, the computed height may not be a whole number, and most software will round it, which introduces a very slight error in the final ratio; if exactness matters, choose a target width that is a multiple of the divisor. Second, rounding the height down rather than to nearest slightly changes the shape, so for repeated resizes prefer scaling from the original file instead of resizing an already resized copy.

The pixel aspect ratio is the shape of a single pixel, usually written as two small numbers, while the display aspect ratio is the shape of the whole picture as viewed. They differ because early broadcast and digital video deliberately used non-square pixels to fill wider screens with fewer transmitted lines. A 720 by 576 frame whose pixels are 12:11 has a pixel aspect ratio of 45:32 once you combine them, but because each pixel is taller than it is wide, the displayed image looks like 4:3, which is its display aspect ratio. Modern displays and nearly all digital content use square pixels, so the pixel aspect ratio is 1:1 and the distinction has largely disappeared, but you will still meet it in DVD frame rate tables, archival video documentation and camera sensor mode descriptions. Keep it separate from device pixel ratio as well, which is the count of physical screen pixels per layout pixel, written 2x or 3x, and is about rendering sharpness rather than geometry.

Because ISO chose the one ratio that lets you halve a sheet and get a smaller sheet of exactly the same shape. A4 measures 210 by 297 millimetres, and 210 multiplied by the square root of 2 is 297, so the ratio of the short side to the long side is 1 to 1.41421. Cut A4 across the middle and you get two sheets of 148.5 by 210 millimetres, which is A5 to within rounding, and its ratio is still 1 to 1.41421 because halving preserves the proportion by construction. Any other ratio would fail this test, which is why the dimension involves an irrational number. Two consequences follow. The area halves cleanly, so A4 is 1/16 of a square metre and A5 is 1/32 of it. And the proportion can never be reproduced exactly on a device that only addresses whole dots, so print services round the millimetre dimensions and accept a microscopic error.

Because the greatest common divisor of 1366 and 768 is only 2, so the reduced form is 683:384, and 683 is not divisible by 16. A true 16:9 panel at 768 pixels tall would be 1365.33 pixels wide, which no display can address, so manufacturers settled on 1366 and accepted a ratio of 1.7786 against the ideal 1.7778, a difference of about one part in a thousand, or roughly a single pixel across the whole width. The calculator surfaces exactly this: entering 1366 and 768 gives a simplified ratio of 683:384, a decimal of 1.7786 and no known-format label, whereas entering 1920 and 1080 gives 16:9 and the label to match. Recognising that distinction matters when you are cropping content for a specific screen, since forcing a true 16:9 crop from a 1366 by 768 source either trims two pixels of width or shifts the frame very slightly.