Saturday, November 10, 2007

How the Microprocessor Runs Software

to see previous matter visit: http://www.lessin1b.blogspot.com/
A Personal Computer's microprocessor is a complex collection of thousands or even millions of microscopic on/ off switches called transistors. The transistors are laid out in a microchip along circuits made up of superfine traces of aluminium. When the transistors are arranged in certain patterns, parts of the microprocessors are designated to hold data and others become capable of manipulating that data in various ways.
The on/off states of the transistors lend themselves handily to representing binary numbers, the form in which data and software code are stored. In the binary system, there are only two digits, 0 and 1, and these represent all numbers. The decimal number 1 is also 1 in binary, but 2 is 10, 3 is 11, 4 is 100, 5 is 101, 6 is 110, 7 is 111, and so on. A transistor that's turned off represents a 0; a transistor that's turned on stands for a 1. Each single 0 or 1 is called a bit. Eight bits make up a byte, and 1,024 bytes make up a kilobyte (K).
A computer sees both software code and data as a series of binary number. If you could shrink yourself to the size of a bacterium, go inside a chip, and see how the transistors in the microchips that make up a PC's random-access memory (RAM) are switched — on or off —you'd find a vast panorama of 0s and 1s, but you'd have no way to determine which portions of that array represent code and which represent data.

JOKES

JOKE1: Barber: ‘Sir, how would you like your hair cut?’
Customer: ‘off.’

JOKE2: The beaver said to tree: ‘It’s good to gnaw you.’

JOKE3: Adam was The world’s first book –keeper. He turned over a leaf and made an entry.

Intel's 80386, 80486, and Pentium microprocessors, along with their imitators, are 32-bit processors, which means that they can manipulate binary numbers that contain up to 32-bits. The largest 32-bit binary number is 11111111111111111111111111111111, which translates to the decimal number 4,294,967,295. Performing arithmetic on any number larger than that requires breaking the number into smaller components, performing the mathematical operations on the parts, and then combining the results.
The number 4,294,967,295 is significant in another way. All memory is identified with an address; 4,294,967,295 represents the upper limit to the amount of memory that the processor can address. Each byte of data in a PC's memory chips is numbered to create a physical, or real, address that represents the location of the actual transistors making up that byte in RAM or ROM chips. Because software applications can't know the physical addresses on the memory on every PC, a different addressing scheme — called a logical address -is used by the software. The microprocessor, in combination with the PC's operating system, uses logical addresses, which are symbolic rather than physical, and translates those logical addresses into physical addresses when reading and writing data.
JOKES

JOKE: I went to Australia recently-but it was closed.

JOKE: A man visited a large town in Australia and asked one of the locals: ’Does this town have any night life?’
‘Yes, ’replied the local, ‘but she’s ill today.’

Despite the ability of modern processors to address more than 4 million kilobytes RAM, most PCs depend on operating-system tricks when they must use more 640 kilobytes. The processor used in the first IBM PC —an Intel 8088 — is a 16 – bit processor, which means that it can manipulate a maximum of 16 bits of data. That means the largest address of a memory location can be only 11111111111 (16 bits filled with the value 1), which allows only 64 kilobytes of RAM to be directly by the 8088.
To make more RAM usable, the creators of MS-DOS — the operating system that was designed to work with the 8088 —devised a segmented memory scheme that combines two 16-bit binary numbers in a way that creates a 20-bit address. With this method. DOS can use the 8088 to address up to 1,024 kilobytes, or 1 megabyte (MB), of RAM. (Of that 1 megabyte, however, only 640 kilobytes is used by DOS for running programs. The rest of the megabyte of memory is left for various hardware devices that require memory addresses.)
The 80286, the processor that followed the 8088, can directly address up to 16 megabytes of RAM, but is hindered by the DOS limitation on the amount of RAM that can be addressed. The Intel 80386 and newer processors can use even more memory, but face the same limitation when used in real mode, a way of operating developed to mimic how an 8088 chip works. (The 80386, 80486, and Pentium chips also have a protected mode that takes full advantage of their capacities to address memory directly and to use special instructions. Protected mode requires OS/2, Windows NT, or some other advanced operating system.) In real mode, special programs called memory managers act as an extension to DOS, allowing the processors to use megabytes of RAM while continuing to use the DOS segmented memory scheme.
Newer operating systems, such as OS/2 and Windows NT, use a flat memory scheme that allows them to address directly the full range of memory that 32-bit processors are capable of using.

to see next matter please visit: http://www.systemsoftwarelession1a.blogspot.com/

No comments: