Origin
PNG, short for Portable Network Graphics, was created in the mid-1990s as a web-friendly, patent-unencumbered replacement for GIF's still-image role. The web needed an image format that could be implemented freely, preserve exact pixels, and handle transparency without the rough edges of one-bit masks.
The result was deliberately practical. PNG did not try to be a camera-photo replacement for JPEG. It focused on portable, lossless raster graphics: the kind of file you want when a single wrong pixel or a jagged transparent edge would be visible.
Technical characteristics
PNG is lossless. Before compression, each scanline can be filtered so neighboring pixels become easier to predict; then the data is compressed with DEFLATE. When the file is decoded, the exact original pixel values return. That makes PNG excellent for screenshots, maps, charts, UI captures, and generated images.
PNG also supports full alpha transparency, indexed color, grayscale, truecolor, gamma information, and physical pixel metadata. It can store rich still images, but it is not usually the smallest choice for photographs because photographic noise and texture do not compress as neatly as flat colors and repeated shapes.
Where it fits
Use PNG when fidelity matters more than minimum size: app screenshots, interface mockups, icons with transparency, QR codes, diagrams, scanned forms, and images that will be processed again. Use JPEG, WebP, or AVIF when a large photo must load fast and tiny visual losses are acceptable.
A practical rule: if the image has crisp text, transparency, or flat blocks of color, try PNG first. If the PNG is huge and the image is photographic, convert a copy to JPEG, WebP, or AVIF and compare.
Best uses
- Screenshots and UI captures
- Transparent logos and stickers
- Generated graphics that need exact pixels
Use another format when
- Large photographic galleries where file size dominates
- Video-like animation
- Print workflows that require CMYK-specific production formats