Aspect Ratio Calculator
Calculate and convert aspect ratios for images, videos, and screens. Find dimensions, resize proportionally, and maintain quality.
An aspect ratio calculator reduces a width and a height to their simplest whole-number ratio and works out the matching dimensions when you resize. The default 1920 by 1080 gives 16:9, 1.78:1, 2.07 megapixels, 2,073,600 total pixels, Landscape and the display standard Full HD; type a New width in the resize box and the New height follows in proportion.
Compare your aspect ratio side by side with the most common ratios to see which one it matches.
About Aspect Ratio Calculator
The ratio is the two dimensions divided by their greatest common divisor, computed with Euclid's algorithm as you type. That is why 1920 by 1080 collapses neatly to 16:9 while 1921 by 1080 stays 1921:1080 - the numbers share no common factor, so there is nothing to reduce. The decimal form beside it, width divided by height to two places, is the number to compare when the whole-number ratio looks unhelpful: both of those examples read 1.78:1.
Four figures sit under the ratio. Megapixels is width times height divided by a million to two decimals, Total pixels is the same product written out with separators, Orientation is Landscape, Portrait or Square from a simple comparison, and Display standard is a lookup of exact pixel pairs - 1920x1080 is Full HD, 3840x2160 is 4K UHD, 2560x1440 is 2K QHD - which returns Custom for anything not on that list. A preview rectangle beside them is scaled so its longest side is 160 px.
The resize box works both ways. Enter a New width and the New height becomes new width times original height divided by original width, rounded to the nearest whole pixel; enter a New height and the width is derived the same way. Rounding is the thing to watch: at small sizes a rounded pixel can move the ratio slightly, so resize once from the original rather than chaining several steps.
The Presets tab holds 24 fixed sizes grouped as Displays, Social media, Photography and Cinema, from HD 720p and 8K UHD to Instagram Story, Pinterest Pin, DSLR Standard and Anamorphic. Their labels are the conventional names of those formats, so a cinema preset labelled 2.39:1 at 2560x1080 pixels calculates as 2.37:1 here; the label is the format's name, the calculation is the arithmetic on the pixels you actually have.
Use Cases
How to use
On the Calculator tab, type the original size into Width (px) and Height (px); the results update as you type.
Read the ratio and, beside it, the decimal form, which is the number to trust when the whole-number ratio does not reduce.
Check Megapixels, Orientation, Total pixels and Display standard underneath for the size rather than the shape.
Type the size you want into New width or New height in the resize box; the other value is derived in proportion and rounded to a whole pixel.
Open the Presets tab and filter by Displays, Social media, Photography or Cinema to load a known size instead of typing one.
Use the Compare tab to see how your ratio sits against the common ones.
Pro Tips
- A ratio like 1921:1080 is not a bug. The tool divides by the greatest common divisor, and coprime numbers cannot reduce, so read the decimal ratio beside it - 1.78:1 - to see that the shape is still effectively 16:9.
- Display standard only recognises exact pixel pairs. 1920x1080 returns Full HD, but 1920x1081 returns Custom, so treat that field as a quick identity check on standard sizes rather than a classification of the shape.
- Resize once from the original numbers. Each conversion rounds to a whole pixel, so going 1920 to 1280 to 640 can land a pixel away from where a single step from 1920 to 640 would.
- Use Megapixels, not the ratio, when the question is print size or storage. 6000 by 4000 and 3000 by 2000 are both 3:2, but one is 24.00 megapixels and the other 6.00.
- Preset names are the industry names for those formats, not the arithmetic. Load a cinema preset and the calculated ratio may differ in the second decimal from the name on the button; the calculated one is what your pixels actually are.
Troubleshooting
The ratio shows huge numbers such as 1921:1080.
The two dimensions share no common factor, so dividing by the greatest common divisor changes nothing. This is arithmetically correct. Read the decimal ratio next to it instead, and if you need a clean ratio, adjust one dimension by a pixel or two.
Display standard says Custom for a size I know is standard.
The field matches exact pixel pairs against a short lookup list. A size that is one pixel off, or a standard the list does not carry, falls through to Custom. The ratio and the decimal ratio are still correct; only the name is missing.
The resized height is one pixel away from what I calculated.
The tool rounds to the nearest whole pixel, because a fractional pixel cannot exist. Over a single conversion the difference is invisible; if you are chaining resizes, go back to the original dimensions each time rather than resizing the result.
A preset's ratio label does not match the ratio the tool then calculates.
The label is the conventional name of that format and the calculation is the arithmetic on its pixels, and the two can differ in the second decimal - a 2560 by 1080 cinema preset works out as 2.37:1. Trust the calculated figure for your own file.
Frequently Asked Questions
The proportional relationship between an image's width and its height, written as two numbers separated by a colon. 16:9 means sixteen units across for every nine down. It says nothing about size: 1280 by 720 and 3840 by 2160 are both 16:9, one is a ninth of the area of the other.
Divide both dimensions by their greatest common divisor. For 1920 and 1080 the greatest common divisor is 120, so 1920/120 is 16 and 1080/120 is 9, giving 16:9. The decimal form is simpler still: width divided by height, which is 1.78 for the same pair, and that is what this page shows beside the ratio.
Multiply the new width by the original height and divide by the original width. For 1920 by 1080 scaled to a new width of 1280: 1280 x 1080 / 1920 = 720. The resize box does exactly that in both directions, rounding the answer to the nearest whole pixel.
It depends on the medium. Screens and video are mostly 16:9, with 21:9 and 32:9 on ultrawide monitors; still cameras are usually 3:2 for full-frame and 4:3 for Micro Four Thirds; phones shoot 9:16 vertically; cinema uses 1.85:1 and 2.39:1. All of those are in the Presets tab with their pixel sizes.
Match the placement rather than the platform. The presets here cover the common ones - a square 1:1 post at 1080 by 1080, a 9:16 story or TikTok video at 1080 by 1920, a 2:3 Pinterest pin at 1000 by 1500, a 16:9 YouTube thumbnail at 1280 by 720. Networks change these, so check the current spec before a campaign.
Scale down rather than up, and do it in one step from the original file. Enlarging invents pixels that were never captured, so detail cannot come back; reducing discards them, which is safe. This page gives you the target numbers, but the actual resampling happens in your image editor.
Width times height divided by a million, shown to two decimals. It is the pixel count, not the ratio: 6000 by 4000 is 24.00 megapixels, and cropping that to 16:9 leaves fewer pixels even though it looks wider. Use it when you care about print size or file weight rather than shape.
Because the exact pixel pair is not in the tool's lookup table. It recognises a fixed list - VGA, SVGA, XGA, HD 720p, Full HD, 2K QHD, 4K UHD, 5K, 8K UHD, UWQHD and a couple of mobile sizes - and returns Custom for everything else, including sizes that are perfectly standard shapes but non-standard dimensions.
No. A tyre's aspect ratio is its sidewall height as a percentage of its section width, and a wing's is span squared divided by area. Both are ratios of two measurements, but neither is a width-to-height picture ratio, and neither is calculated by this page.
Yes, indirectly. The arithmetic does not care about units: enter 30 and 20 and you get 3:2, the same as a 6000 by 4000 photograph. The megapixel and total pixel figures will then be meaningless, because they assume the two numbers are pixel counts.