Code_ The Hidden Language of Computer Hardware and Software - Charles Petzold [111]
Also in 1977, the Apple Computer Company, founded by Steven Jobs and Stephen Wozniak, introduced the Apple II. The Apple II, however, used neither the 8080 nor the 6800 but instead used MOS Technology's less expensive 6502 chip, which was an enhancement of the 6800.
In June 1978, Intel introduced the 8086, which was a 16-bit microprocessor that could access 1 megabyte of memory. The 8086 opcodes weren't compatible with the 8080, but I should note that they included instructions to multiply and divide. A year later, Intel introduced the 8088, which internally was identical to the 8086 but externally accessed memory in bytes, thus allowing the microprocessor to use the more-prevalent 8-bit support chips designed for the 8080. IBM used the 8088 chip in its 5150 Personal Computer—commonly called the IBM PC—introduced in the fall of 1981.
IBM's entrance into the personal computer market had a great impact, with many companies releasing machines that were compatible with the PC. (What it means to be compatible will be explored more in subsequent chapters.) Over the years "IBM PC compatible" has also implied "Intel inside," specifically Intel microprocessors of the so-called x86 family. The x86 family continued in 1982 with the 186 and 286 chips, in 1985 with the 32-bit 386 chip, in 1989 with the 486, and beginning in 1993, with the Intel Pentium line of microprocessors that are currently used in PC compatibles. While these Intel microprocessors have ever-increasing instruction sets, they continue to support the opcodes of all earlier processors starting with the 8086.
The Apple Macintosh, first introduced in 1984, used the Motorola 68000, a 16-bit microprocessor that's a direct descendant of the 6800. The 68000 and its descendants (often called the 68K series) are some of the most beloved microprocessors ever made.
Since 1994, Macintosh computers have used the PowerPC microprocessor that was developed in a coalition of Motorola, IBM, and Apple. The PowerPC was designed with a type of microprocessor architecture known as RISC (Reduced Instruction Set Computing), which attempts to increase the speed of the processor by simplifying it in some respects. In a RISC computer, generally each instruction is the same length (32 bits on the PowerPC), memory accesses are restricted to just load and store instructions, and instructions do simple operations rather than complex ones. RISC processors usually have plenty of registers to avoid frequent accesses of memory.
The PowerPC can't execute 68K code because it has a whole different instruction set. But the PowerPC microprocessors currently used in the Apple Macintoshes can emulate the 68K. An emulator program running on the PowerPC examines each opcode of a 68K program, one by one, and performs an appropriate action. It's not as fast as native PowerPC code, but it works.
According to Moore's Law, the number of transistors in microprocessors should double every 18 months. What are those many additional transistors used for?
Some of the transistors accommodate the increase in processor data width, from 4 bits to 8 bits to 16 bits to 32 bits. Another part of the increase is due to new instructions. Most microprocessors these days have instructions to do floating-point arithmetic (as I'll explain in Chapter 23); new instructions have also been added to microprocessors to do some of the repetitive calculations required to display pictures or movies on computer screens.
Modern processors use several techniques to help improve their speed. One is pipelining. When the processor is executing one instruction, it's reading in the next instructions, even to a certain extent anticipating how Jump instructions will alter the execution flow. Modern processors also include a