Lots and lots and lots of dots

In contrast to vector illustration programs, photo-editing programs like Adobe PhotoShop, Corel PhotoPaint, etc. work with Bitmap images. When you work with bitmap images, you can refine small details, make drastic changes, and intensify effects.

Bitmap images, also called raster or paint images, are made of individual dots called pixels (picture elements) that are arranged and coloured differently to form a pattern. When you zoom in, you can see the individual squares that make up the total image. Increasing the size of a bitmap has the effect of increasing individual pixels, making lines and shapes appear jagged.

However, the colour and shape of a bitmap image appear continuous when viewed from a greater distance. Because each pixel is coloured individually, you can create photorealistic effects such as shadowing and intensifying colour by manipulating select areas, one pixel at a time.

Reducing the size of a bitmap also distorts the original image because pixels are removed to reduce the overall image size.

Actually, the term bitmap graphics goes back to the early graphic computers which used monochrome displays. As you may or may not know, for a monochrome display, we can store the information for each pixel in one bit of memory, hence the term bitmap graphics.

Nowadays, even though the information for each pixel may be stored in 1-3 (or more) bytes of memory, the term still persists.

Basically, the main point to remember is that we do have to store the information for each and every pixel in the image, this is what causes the large memory requirements for this type of graphic.

One of the biggest advantages of bitmap graphics, is that because we can set the colour of every individual pixel in the image, we can display photorealistic pictures very easily.

One of the disadvantages of bitmap graphics, is that they do not scale very well. If we decide to make the image larger or smaller, then the program has to try and recalculate what colour the pixels should now be. Actually some of the better programs do a fairly good job of this, as long as the change in size is not too great.

If we try and enlarge bitmap images too much, we will also run into the problem of aliasing, this is the staircase effect, that usually shows up in diagonal lines, or where two different colours sit side by side. The image will also start to look blurred.
Have a look at the samples below.

Here is the original image

Here is a section magnified 200%

Here is a section magnified 400%

As you can see, the image loses clarity and is not really usable when magnified too much.

What about decreasing the size of the image? See the examples below

Here is the original image

Here is the image reduced by 50%

Here is further 50% reduction

As you can see, it isn't quite so bad when reducing an image, although some clarity is still lost.

The re-sizing above was done using the smartsizing feature of the program. What this does is tries to maintain the appearance of the image as much as possible by using antialiasing and re-sampling

Below are 2 images, both enlarged 400% - one uses smartsizing the other doesn't

With smartsizing/re-sampling

Without smartsizing/re-sampling

There is an obvious difference here, the image without the smartsizing really shows the aliasing problem. This is sometimes call the staircase effect or the jaggies. Notice also that the image is starting to look ‘blocky’

Actually this shouldn't be a problem these days, as just about all modern software does it automatically. The main reason that earlier software had the option to turn it off, was that many computers back then didn't have enough processing power and/or memory to use antialiasing.

So what does all this mean when we are using bitmap graphics?

Bitmap graphics should be created or scanned either at, or very close to, the size required for the finished product.

You can’t successfully change the size of the image in your word processor, or DTP package. You need to load the image into a suitable program and re-size it by re-sampling

If you do need to re-size an image, avoid drastic changes in size.

If you are creating a large image, you will need plenty of memory available

Vector Graphics >