The Complete Manual of Typography: About Fonts
In this chapter:
TWO KINDS OF FONTS: OUTLINE AND BITMAPPED
FONT FORMATS: POSTSCRIPT, TRUETYPE, AND OPENTYPE
CROSS-PLATFORM COMPATIBILITY ISSUES
FINDING THE CHARACTERS YOU NEED
IDENTIFYING FONT FORMATS (MAC AND WINDOWS)
The Two Basic Kinds of Fonts: Outline and Bitmapped
Digital devices—computer monitor screens, desktop printers, imagesetters—create images out of dots. The simplest way to create type for one of these devices is to draw a picture of every character as an array of dots and store these drawings in a font. Then all a device has to do to image the type is to copy those dots into place on the screen or page. When this technology was first figured out, each one of those dots was represented by one bit of computer data—a simple yes/no choice of whether to image a dot or not. Images created from these predrawn, prearranged arrays of dots were called bitmaps, and fonts using this trick were called bitmapped fonts.
Bitmaps are a clever and simple approach, but the more dots a bitmap contains, the more computer data it requires. As the resolution of the device increases or the size of the character images increases, the number of dots grows geometrically: Doubling the size of a character quadruples the number of dots. You also need a separate set of bitmaps—a separate font—for every size of type you want to create. And the bitmaps designed for one resolution will appear much smaller when imaged on a device with a higher resolution, where the dots are much smaller (see Figure 4.1). To image a single typeface at the same range of sizes on a computer screen, a desktop printer, and an imagesetter, then, would require hundreds of bitmapped fonts.
Figure 4.1 In a bitmapped font, each dot in the drawing of a character is mapped against one pixel of the output device. As the device's resolution increases, the dots of the bitmapped font become smaller and smaller, and the character follows suit. In this illustration, a character bitmapped for the screen appears at left. If this bitmap is imaged on a desktop printer, it shrinks to the size in the center. At a typical imagesetter resolution, it shrinks to the size on the right.
The solution is to store the descriptions of the characters as a set of outline drawings. Outline fonts, which do just this, store character images as outlines described mathematically as a series of curves and straight-line segments. (These line segments are sometimes called vectors; and the fonts based on them, vector fonts.) These outlines can be mathematically scaled to any size without distorting the shapes or proportions of the characters. The scaled outlines are then colored in with dots of the size created by the device that the type is being imaged on: around 100 dots per inch (dpi) for a computer screen, approximately 600 dpi for a desktop printer, and well over 1,000 dpi for imagesetters.
While it was once common for operating systems to use bitmapped fonts for screen display, they now generate screen type from the same outline fonts used for high-resolution printing. Some fonts may contain sets of hand-tuned bitmapped screen fonts for use at small sizes because they're more legible than those generated by your computer. But these embedded screen fonts are not apparent to the user, and you don't have to concern yourself with them.