GIFConverter is a shareware graphics utility that enables you to work with raster images. These are the kinds of images you would use with a black and white or color paint application, or that you get from your scanner. Using GIFConverter, you can view these images, change their appearance, save them in nine different file formats, and print them.
GIFConverter was originally developed in response to a challenge by Larry Wood, administrator of the Graphics Support Forum on CompuServe. He was looking for somebody to develop an application for the Macintosh that supported GIF, the Graphics Interchange Format. (“GIF” is officially pronounced “JIF,” like the peanut butter. “GIFConverter” has a hard “g” as in “gift” because I named it before I ever heard the word GIF spoken.) You can use GIF files on practically any computer, and for that reason you will see thousands of GIF files on BBSes and major online services like Compuserve, GEnie, and America Online.
GIF Files themselves can contain one or more images, each with an associated color map. An image is one way a computer can represent a picture. Images are more like photographs than like line drawings. Line drawings are made up of mathematical elements like lines, circles, curves and polygons. An image is made up of a rectangular arrangement of points.
An example of an image, greatly magnified
These points in the images are called pixels. “Pixel” is verbal shorthand for “picture element.” The image is actually a rectangular array of pixels. Think of a sheet of graph paper. You can color each square on the graph paper so that each square is one color. If you stand back from the graph paper, you don’t see the individual squares, you see the overall effect. On your computer screen, each pixel is 1/72” of an inch (or so) on a side. There are 5,184 pixels in every square inch of the screen. A 9” 512 x 342 pixel Macintosh screen has 21,888 pixels. A 13” 640 x 480 pixel Macintosh screen has 307,200 pixels.
You may have a 300 dpi printer. The dpi stands for “dots per inch,” and a dot is just another name for a pixel. These printers have 90,000 pixels per square inch.
GIFConverter can work with images that have up to eight bits per pixel. The number of bits per pixel determines how many separate colors the image can contain. Images in GIFConverter can contain up to 256 colors each (this is because 2 8=256).
When images have 256 colors or less, they often use a color map or color table. This allows the images to use a specific set of 256 colors. A color monitor can really display 16,777,216 colors. That corresponds to 24 bits per pixel, but a 24-bit-per-pixel image takes three times as much memory as an 8-bit-per-pixel image. Also, working with 24-bit-per-pixel images is much slower.
The color map or table is simply a list of RGB colors. Each entry in the list has a color value for the red, green, and blue values that should be sent to your screen (hence RGB). Red, green, and blue are called the additive primaries because all colors can be represented as a mixture of red, green, and blue.
The computer then uses the value of each pixel in the image to look up the color in the table. If it finds a pixel that has the value 5, it looks at color #5 in the table, and sends that RGB value to the screen. Sometimes a color table used in this way is called a Color Look-Up Table or CLUT.
Color tables are often handy for making quick color adjustments for the image. Instead of changing the thousands of pixels in the image, a program like GIFConverter need change only the 256 color entries in the color table to get the desired effect.
Dithering and halftoning are two ways to adapt an image to a particular set of colors. This set of colors can be more than just black and white. Both of them depend on the human eye’s ability to blend details.
The most common use of halftoning is in the printing of photographs in newspapers and magazines. A photograph has a range of intensities, but a printing press has only two—either there is ink on the page or there is not. Halftoning converts a image into a field of spots. The size of each spot corresponds to the darkness [1] at that point. Light areas have small spots spaced a distance apart. Areas that are almost black have spots that grow together and touch. In the darkest areas, the white spaces between the spots disappear entirely.
On computers, we simulate the spots with groups of pixels. See the examples of halftoning methods in the section entitled The Display.
Dithering is another method of reducing the number of colors. In dithering, the computer chooses colors so that the average effect you see corresponds to the average color for an area of the image. Dithering does not have discernable spots like a halftone, but can sometime have other undesirable patterns.
The original Macintosh screen had only one bit per pixel. It could only show two colors, black and white. Color Macintoshes can display two, four, or eight bits per pixel (4, 16 or 256 colors) with a color table. If you have the right hardware, your Macintosh may display what it calls thousands or millions of colors. These are 15 or 24 bits per pixel, but use RGB values directly instead of a color table.
Now that we’ve defined some terms, we’re ready to look at the organization of a GIF file:
A GIF file can contain one or more images. Each image has a color table that defines up to 256 colors. The GIF file has a background color that’s used to fill the spaces between images, if any.
The first GIF standard was called GIF87a. There is a new standard, GIF89a, which defines some extensions to the GIF file format. GIFConverter currently supports only the comment text extension to the GIF format.
A GIFConverter document in memory is organized the same way as a GIF file. The images appear in a window, and you can call up a separate window for the comment text.
[1]Darkness is the inverse of intensity. We print with black inks on white paper, so darkness is appropriate here.
Copyright ©1998 by Kevin A. Mitchell. All rights reserved.