Images stored on disk are compressed. They are typically made up of four “channels”, one for red tones, one for green tones, one for blue tones and one for alpha transparency. Each channel is a 2 dimensional array of pixels. So if you have 1024x768 sized image, the actual image has 1024x768 red pixels, each one taking 1 byte of memory, 1024x768 green pixels, 1024x768 blue pixels and so on.
Some quick math: 1024 * 768 = 768,432 bytes per channel of memory. Multiply that by 4 channels and its 3.1MB of memory just to hold the color data. So an uncompressed in-memory image at 1024x768 is going to take 3.1+ megabytes. The software is manipulating these pixels and it’s very inefficient to try and uncompress this on the fly.
But it saves disk space to keep the images compressed.
[import]uid: 19626 topic_id: 24618 reply_id: 99771[/import]