Return to the Little Kingdom_ Steve Jobs and the Creation of Apple - Michael Moritz [80]
Wozniak returned to his lab bench at Hewlett-Packard and started to attack two entirely different problems. One centered on devising a circuit that would display color. The other was devoted to reducing the number of chips on the board by simplifying the memory. The Apple computer had two sets of memory. One—an 8K-byte board of chips—served the microprocessor. Another—composed of shift registers (an older, slower form of memory)—served the black-and-white display. In an effort to reduce the number of chips, Wozniak wanted to find a way to let one memory serve both the computer and the display. He investigated the way a picture is displayed on a television and found that a raster scan spends two thirds of its time moving across the screen spraying the phosphor with electrons from left to right, and one third of its time whipping back from right to left. Armed with that knowledge Wozniak decided to force the microprocessor and the display to share the same memory. While the raster moved across the screen, taking bits from the memory, the microprocessor was barred. And while the raster whipped back, the microprocessor plunged in. It was the sort of approach that had been the subject of debate at the Homebrew and a writer in the newsletter had wondered in August 1975 whether the timing for a display might be solved for members if “a circuit could be published which read from a microcomputer memory while the computer isn’t using it.” To make the allocation work, Wozniak had to slow down the speed of the microprocessor. “All the computer was supposed to do was play games, so nobody would ever know. It was funny. Just by thinking of a couple of unrelated issues, out came a simpler design.” Wozniak had virtually added color for free and designed a computer which, though it had about half the number of chips of the first machine, was more powerful.
Wozniak also wanted to expand the capacity of the Apple. Much of the power of minicomputers had sprung from slots in the motherboard that housed smaller printed circuit boards. The slots were a crucial part of the design because it meant that the computers could be expanded to perform a large variety of tasks. smaller printed circuit boards that plugged into the slots might contain more memory chips, a connection to a printer or to a telephone. Some of the most successful minicomputer companies had encouraged smaller firms to make peripherals that would work with the computer. So the slots tended to bring benefits to everybody: to the computer manufacturer who could boast about the machine’s many attributes and the subindustry that it created; to peripheral manufacturers who would make a new product; and to the customer who came to own a machine that could do more than one job. Part of the reason the Altair made such an impression on the hobbyists was that it mimicked a minicomputer with slots. Wozniak liked the notion of slots—“I was used to computers with twenty slots that were always filled up with boards”—and decided that his color computer should have eight slots. Jobs disagreed, and the difference of opinion turned into one of their most protracted arguments. Wozniak recalled, “All Steve saw was a computer that could do a couple of things—write basic programs and play games. He thought you might add a printer and maybe a modem but you would never need more than two slots. I refused to let it go with two slots.
While Jobs and Wozniak wrangled about the number of slots, they made regular appearances at the Homebrew Club whose meetings ran like a motif through the development of the color computer. At the hobbyist sessions Wozniak collected a couple of teenage camp followers: Randy Wigginton who in the summer