The Quick Start Guide to the GIMP, Part 2

This second article in the series introduces us to the basic features of the GIMP, a Linux power tool for the graphics artist.

Last month, in the first of this four-part series on the GIMP, we took a quick tour around the GIMP's installation and system requirements. This month we'll cover a few basics about how the GIMP works, which types are included and how they are used, and a bit about supported file formats. This information is somewhat condensed, and you may wish to read through it once, then read it again as you begin playing with the GIMP and its windows.

How Does the GIMP work?

The GIMP is a raster graphics tool, meaning that it operates on images as collections of individual points called pixels. Each pixel is made up of a number of channels. The number of channels depends on the image type being processed. For example, RGB (and RGBA) images consist of four channels, one each for levels of Red, Green and Blue and one called the Alpha channel which is used to determine the transparency for a pixel. Transparency for individual pixels becomes important when working with layers. Layers are a feature of the GIMP which permit an artist to create pieces of an image separately; the pieces can be manipulated on their own without affecting the rest of the image. This is useful, for example, when creating cover art for a magazine or CD. The image on the cover of the November issue of Linux Journal had many layers. The text for the word “Graphics” is made up of 4 layers, each combined with the layer below in a different way to create the final 3D effect. We'll talk about layers in more depth next month, when we cover the Image Window in Part 3 of this series.

The GIMP supports three types of image formats: RGBA, Grayscale and Indexed. RGBA was described in the previous paragraph. RGB images are just like RGBA except they do not contain an Alpha channel. Grayscale images are similar to RGBA images except they only function with one channel—Gray—that allows varying levels of gray from black to white. Indexed images use a color palette to determine which colors are available, and each pixel's color is represented by an index value into that palette. For example, a value of 112 for a pixel means that the pixel will be the color specified by the 112th entry in the color palette. Each channel of an RGBA image uses 8 bits, allowing for 256 shades of each channel's color or transparency level. The grayscale channel uses 24 bits, supplying about 16 million shades of gray. Indexed images use 8 bits to define the color palette index, meaning a color palette consists of 256 unique colors.

Most processing work should be done in RGB (note that the GIMP usually refers to RGBA simply as RGB) or Grayscale modes to allow greater flexibility. Some image file formats, most notably the GIF format, save images as indexed images. If you open a GIF file for processing with the GIMP, you should consider first converting it to RGB. Afterwards, if a GIF file is required (say, for use on a web page), you can convert the image back to indexed. Conversion of image formats from within the GIMP is done using the Image Window menus, which we'll discuss at length next month.

GIMP Windows

When you start the GIMP, you'll find it provides a single, small window with a set of buttons, called the Toolbox. The Toolbox is the starting point for creating images and is made up of a menu bar, the tool buttons and the foreground/background color block. If you are familiar with Adobe Photoshop then the look and feel of the Toolbox should be quite familiar to you. Figure 1 shows the default Toolbox configuration.

The GIMP provides a number of other windows with which you should become familiar. The first is the Image Window, in which an image is displayed. There can be more than one of these windows open at a time. You can open the default Image Window by placing the cursor over the Toolbox and typing ctrl-N (N is for New Window). A small dialog box will open that allows you to specify some parameters for this new window. Accept the defaults for now so that you can see an example Image Window. These windows display an image as it will look in its final form based upon the layers that are currently marked as visible. Image Windows are resizeable and scrollable. It is possible, through the use of zooming features, to have an image that is larger than the display area of an Image Window as well as an image which is smaller than the display area.

Image Windows are made up of a number of smaller window features: rulers, menus and scrollbars. The rulers, along with their guides, provide a convenient method for determining location and size of areas of the image. All image windows have rulers on the left and top sides of the window, although these can be turned off using one of the pop-down menus.

The pop-down menus in the Image Window are not obvious—you have to hold down the right mouse button while the cursor is in the Image Window to open the menus (opening a menu is also known as “posting” the menu). These menus contain a wealth of options for viewing, selecting and manipulating the image. While many of the options are available directly from the Toolbox, many others, like the plug-in filters, are only directly accessible from pop-down menus. You'll probably want to familiarize yourself with the layout of these menus, since you'll be using them quite often.

The scrollbars appear on the right side and the bottom of the Image Window. They are used to move the image around the display area when the full image cannot be displayed within the current width and height of the window.