Effects of Using Less Colours in the Image or Display

WARNING: This page will only make sense if viewed with a 24 bit (True-Color) display. The effects may be seen to some extent with a 16 bit display but this is not optimal.

You might find yourself working with a reduced number of colours, or bit depth, because you need to keep the files sizes small or because your display is not able to display a photographic quality 24 bit image. The changes in the images here are caused by changes in the way the files themselves were made, but a similar effect is caused by changes in the display.

24 bit Image

Here is the full 24 bit image of the Matterhorn with a size in pixels of 200 by 150. There are 8 bits for red, green and blue, giving 256 possible values for each of the three colour components to the image. This gives a possible 16,777,216 different colours, which is more than the eye can usually distinguish. There are 30,000 pixels in the image, which with three bytes per pixel, makes the file about 88 KBytes in size.

8 bit Image

This next image takes the most used 256 colours in the picture and maps the whole picture to these colours. Dots of colour in groups together are used to make it appear that there are more colours visible than the 256. Using the most popular 256 colours from an image is known as having an adaptive palette and it can work very well for most images. The size of the file has now shrunk to 29 KBytes. Of course, a photographic black and white image can be built using only 256 shades of grey.

Quarter resolution

As an alternative, if you wish to quarter the size of your file, you could reduce the number of pixels in the image to 100 by 75 and then either have it on-screen at quarter size or, as here, blow it back up. You should be able to see the pixilation this causes. The file, with half the number of rows and half the number of columns of pixels, will be about 22 KBytes in size.

3 bit Image

Now we move down to only having eight colours available but again chosen as the most popular in the picture. The groups of coloured dots, known as a dither pattern, are more visible here, especially on the smooth snow face at the left of the mountain.

1 bit Image

Finally this is a monochrome image. Dither patterns of deep black and peak white dots only are used to make up the image and the dithering is so visible as to be distracting. The file size has come down to about 3 KBytes.

One irony of using a dithered image like these, with different coloured dots in groups, is that they do not compress very well with run-length encoding or JPEG because neither system compresses scattered dots very well. JPEG compression is discussed on this page.