Basic Microprocessors Architectural Concepts The concept of microprocessors
Basic Microprocessors Architectural Concepts The concept of microprocessors and buildings are similar. Most try to do the same job, but the internal structure which they use to do the job is different. That is, their architecture is different. Examples: wooden poles and Greek column may be not necessary
Architectural Differences n The Intel 8051 8 -bit single-chip microprocessor and the Intel 80486 32 -bit microprocessor will both add two numbers. The way they add the two numbers are quite similar. But if you want to compute the cosine of an angle, they perform and the parts they use will be very different.
Architectural Differences n Major differences: n n n n n Length of data word Size of memory which can directly address Speed of executing an instruction Number of registers available Different types of registers Different types of instructions Different types of memory addressing mode Different types of support circuits Compatibility with development tools
Word Length Longer word length tends to have more features and extensive instruction sets to solve more problems faster but its microprocessor and support components cost also increases. n From the figure below, the number of support circuits typically grows with the number of bits of the data word. The choice of microprocessor word length should be made to get the job done without overkill. n
Supporting Circuits Vs Word Length
4 -bit Applications Toys: robots remote-controlled cars n Calculators: financial, scientific n Simple, intelligent consumer product controllers: microwave ovens, telephone dialers, TV remote control n Computer peripherals: keyboard, scanners, simple printers, clocks n
8 -bit Applications Toys: video games, programmable robots n Complex, intelligent consumer product controllers: VCR control and programming, security systems n Computer peripherals: video displays, higherspeed printer, modems, disk controllers n Industrials: robotics, logic analyzers, disk drive tester, digital oscilloscope, smart voltmeters n
16 -bit, 32 -bit, and 64 -bit n n n 16 -bit, 32 -bit, and 64 -bit are the heart of the personal computers (PC) 1981 – 8088 introduced by IBM, is a 16 bit internally and externally it uses an 8 bit bus 80286 – is a 16 -bit microprocessor 80386, 80486 and Pentium – is a 32 -bit microprocessor Other areas such as six-axis-robotic arm
Motorola 68000 family n n Used for a number of advanced personal computers such as Apple Macintosh It also used for Unix-based industrial microcomputer system
Why 32 -bit? n n n Why is 32 -bit microprocessor faster in these situations? A 32 -bit word gives the user resolution of 1 part in 4 billion (4, 294, 967, 295). The 16 -bit word has a resolution of 1 part in 65, 536 only. Both the Intel 80386 and Motorola 68020 series have full 32 -bit parts. Some calculations can be performed in a single rather than multiple-precision operation
32 -bit Data Handling Each time the 32 -bit microprocessor takes data from its memory, it gets four times as much data as the 8 -bit microprocessor n Increasing the microprocessor’s speed is difficult. Therefore the speed increased by doubling the word length is a very real improvement. n
Addressable Memory Early 4 -bit microprocessors such as Intel 4004 had limited memory addressing, only 16, 384 memory locations available (214=16, 384) n Most 8 -bit microprocessors use 16 address lines, can therefore address 216=65, 536 n Most single-chip microcontroller can address 1 to 8 K locations of ROM for program storage and 64 to 512 RAM locations for data. n
How Much Memory-Addressing? Example: Simple Alarm Clock Application n TIME: AM/PM, 10 s hours, 10 s minutes, 10 s seconds, 1 s seconds (7 memory locations) n TIME SET: similar to TIME (7 memory locations) n ALARM: similar to TIME (7 memory locations) n CONTROL: Alarm On/Off, Alarm Set, Time Set, Clock Run (4 memory locations) n
How much ROM and RAM needed? Typical Alarm Clock Application can be written in less than 1000 instructions and stored in ROM n Total=7+7+7+4=25 locations, 1 byte each of RAM needed n So a 4 -bit microprocessor with 64 RAM will do the job n
Microprocessors’ Speed The speed which a microprocessors executes its instructions n Speed can be measured in two different ways: n Microprocessor’s clock speed – frequency of the clock oscillator n Number of instruction it can execute per second – usually expressed in MIPS – million instruction per second n
MIPS can only tell for a fairly large program n See figure in the next page n If we can double the word size and keep the clock frequency the same, the MIPS rate is nearly doubled n On average, a 8 -bit microprocessor must access memory at least twice to fetch a complete instruction n
MIPS Table
Increase of Clock Frequency The clock frequency is largely depends on the semiconductor manufacturing technology. It depends on how many transistors can be put on the IC n The processes used to build the very dense microprocessors today are able to make parts as close as a few tenths of one millionth of an inch n
Semiconductor Technologies n Most microprocessors are constructed using one of the following semiconductor technologies: NMOS (N-channel metal oxide semiconductor) n CMOS (Complementary metal oxide semiconductor) n n NMOS processors typically allow below 20 MHz. Some VLSI cannot be produced by NMOS
CMOS processors typically will operates at higher frequency than NMOS n CMOS technology are much lower in power n
1988 Computer Food Chain Mainframe Supercomputer Minisupercomputer Work- PC Ministation computer
1997 Computer Food Chain Massively Parallel Processors Mainframe Minisupercomputer PDA Server Supercomputer Minicomputer Work- PC station
Why Such Change? n n n Performance n Technology Advances n CMOS VLSI dominates older technologies (TTL, ECL) in cost AND performance and is progressing rapidly n Computer architecture advances improves low-end n RISC, superscalar, RAID, … Price: Lower costs due to … n Simpler development n CMOS VLSI: smaller systems, fewer components n Higher volumes n CMOS VLSI : same device cost 10, 000 vs. 10, 000 units n Lower margins by class of computer, due to fewer services Function n Rise of networking/local interconnection technology
Other Architectural Characteristics n Parallel Processing Both microprocessor working on a job which can be done by a single microprocessor but which can be done faster with two microprocessors n Co-processing The co-processor work with the main processor to perform special functions. Example: 80387 numeric data co-processor provides logarithms, square root function, etc
Other Architectural Characteristics. . Cache Memory Is a fast memory located very close to the microprocessor. Not only fetches instruction from the memory but also fetch the next to fill the cache memory n Pipelining Like an assembly line, each instruction is processed a little at each station. This allow faster, more dedicated logic at each station n
Other Architectural Characteristics. . n Wider Buses Double-wide accumulator allows the user to process certain function faster because it eliminates the need to perform an intermediate storage in memory
The Microprocessor’s Register The registers are temporary data storage devices found inside the microprocessor n Some have special function and others are used for general purpose data storage n The programmer need enough registers to achieve specific task n If the register is not enough the programmer must transfer the data into a temporary memory location for storage when the task is complete n
The Microprocessor’s Register. . The data is then retrieved from the temporary location when the task is going to use the register again. Each of the data transfers from register to memory or from memory to a register are wasted instructions. n Many different types of registers in a microprocessors. Some make the job of selecting the memory address for program data much easier. n
Microprocessor’s Instructions Additional microprocessor instructions can only be in the microprocessor if the microprocessor has the necessary logic circuits to execute the instruction. n This means that more instructions mean a bigger microprocessor, and bigger microprocessors are more complex. n Example: 8051 has 111 different instructions and Z 80 has 178 (for 16 -bit data transfer). n
Types of Memory Addressing Mode Different kinds of ways the programmer can refer to data stored in memory. n Example: 8 -bit microprocessor uses direct addressing mode (1 byte instruction + 2 bytes memory address). Some microprocessor uses index addressing which use 1 bytes for instruction and 1 byte for either added to or subtracted from the index register. n
Types of Support Circuits Other external circuit which enable microprocessor communicate with the outside world, such as UART – serial I/O n Others such as arithmetic coprocessor (8087/80287), RAM controller, clock generator, programmable interval timer, floppy disk controller, serial communications controller, parallel input/output controller, etc. n
Development Tools n n n Some microprocessors only have assemblers. They do not have any high-level language. A microprocessor development system includes the software, which allows a programmer to write and modify programs; the hardware tools, which one to allow to examine the microprocessor. In-circuit emulator is a test instrument which allows one to test the microprocessor system to examine how the microprocessor is functioning in the circuit. One can stop in the middle of executing program and examine the contents of memory and registers.
Summary If possible we make microprocessor go faster by increasing its clock frequency. This is not done without expense, because parts which work at higher frequencies cost more. n If we cannot speed up the clock frequency, there a number of architectural techniques which add hardware to the microprocessor but which result in faster operation. n
- Slides: 34