What is the PNG format?

This material is intended to inform the simple users about the features of the PNG format. So, it will pass over some details like for example the freedom from patents, as this has no connection to the end-user, being more of a concern for the developers.

The PNG format was intended to be used like a substitute for the GIF and TIFF formats. So, it was created with two main ways of usage in view:

1. World Wide Web
2. Image editing

PNG can be widely used for the web, having several advantages over the GIF format: the possibility of transparency by using alpha-channels, gamma-correction (control of image brightness on any platform) and two-dimensional interlacing (a method of progressive display). PNG is also better compressed than GIF, but the difference is small. PNG was intended to be a single-image format, so it cannot contain animation. But there are formats derived from PNG, like MNG and APNG, which support animation.

As PNG’s compression is fully lossless and it supports up to 48-bit images or 16-bit grayscale, manipulating the image would not degrade its quality unlike JPEG. And unlike TIFF, the PNG image is readable in every PNG-supporting application. But JPEG can be compressed to a much smaller file, with the quality close to the original, and the TIFF’s black and white images compressed in Group 4 fax compression are far better than the 1-bit grayscale PNG images.

Being a rester format, like GIF and TIFF, it represents the image like a two-dimensional array of pixels. PNG is not an explicitly vector format, so the images stored in PNG cannot be arbitrarily scaled to any possible size without distortion. In order to be able to do this, one should better use SVG or PostScript. There are also some private extensions the PNG format that add support for the vector information in addition to classic PNG pixel information (for example the extensions from Macromedia Fireworks), but no valid PNG can omit the pixel data.