Resizing images without wrecking them
Resizing changes an image more than most people expect. Here is what the three fit modes do, why enlarging is a losing game, and how to choose target dimensions.
Resizing looks trivial and is not. Where the pixels come from, in what order the operation happens, and whether the image is getting bigger or smaller all change the result.
Three ways to reconcile a target box
When you ask for a specific width and height that the image's own proportions do not match, something has to give. This tool offers three answers, and the option is named in the result so you know which one ran.
- Fit scales the image down until it fits inside the box, keeping the proportions. The result may be smaller than the box in one direction. This is the safe default for shrinking photos to a maximum size.
- Fill scales until the box is covered, keeping the proportions, then crops the overflow evenly from both sides. Use it when the output must be an exact rectangle, such as a thumbnail grid. Be aware that it removes content.
- Stretch forces the exact dimensions and ignores the proportions. Only correct when the mismatch is intentional, because it visibly distorts faces and circles.
Shrinking is fine, enlarging is not
Shrinking an image with a good resampling filter works well. Details smaller than a pixel are averaged away rather than lost at random, and the result looks clean.
Enlarging cannot invent detail. Any enlargement is a guess about the pixels that were never captured, and the result looks soft or, with sharpening, slightly artificial. If you need a larger version of an image, the only real answer is a larger source. Vector artwork is the exception: an SVG can be rasterised at any size and stays crisp, because it is instructions rather than pixels.
Order of operations
Resize before you convert to a lossy format, not after. Resizing a JPG means decoding it, resampling, and re-encoding, so the image is going through the lossy encoder once regardless. Doing the resize first means only one generation of loss, and the smaller image encodes faster and smaller.
Choosing dimensions
Match the destination rather than picking round numbers. A common set of targets:
- Full-width web image: 1600 to 2400 pixels wide is ample for a high-density display.
- In-article illustration: 1000 to 1200 pixels wide.
- Thumbnail: 300 to 400 pixels wide.
- Social preview card: 1200 by 630 pixels, so use the fill mode and check what gets cropped.
Doubling the width quadruples the pixel count, and file size follows roughly that ratio for lossless formats. Halving the width is the single most effective way to shrink a PNG.
Convert an image now
The converter reads PNG, JPG, GIF, BMP, TIFF, ICO, WebP, AVIF, HEIC and SVG, and writes PNG, JPG, WebP, GIF, BMP, TIFF, ICO, AVIF and PDF. It states for each file whether the work happened on your device or on the server.
Converters mentioned in this guide
More guides
-
Does my image stay on my device?
A direct answer, because the honest one depends on the conversion. This explains which operations are local, which need the server, and how the tool tells you which was which.
-
GIF and animation: what a still converter cannot do
Animated GIFs need a format that stores a sequence of frames. Converting one to PNG, JPG or WebP means keeping a single frame, and this tool says so instead of quietly dropping the rest.
-
What is a HEIC file, and why is it awkward?
iPhones save photos as HEIC by default. The format is better than JPG on almost every axis, and that is exactly why it causes trouble: support outside Apple devices is uneven.