• The ISA bus, which is an older low speed bus.
• The PCI bus, which is a new high speed bus.
• The USB bus (Universal Serial Bus), which is a new low speed bus.

As mentioned earlier, I/O busses are really derivatives from the system bus. On the system board it ends in a controller chip, which forms a bridge to the two other busses. The essential in modern PC’s is fast busses. Let us compare the two primary I/O busses:

Clearly, there is a vast difference between the capacity of the two busses. All in all, the busses have a very central placement in the PC’s data exchange. Actually, all components except the CPU communicate with each other and with RAM via the different I/O busses. Here you see a demonstration of this logic:

• The system bus, which connects the CPU with RAM
• I/O busses, which connect the CPU with other components.

The point is, that the system bus is the central bus. Actually, it connects to the I/O busses, as you can see in this illustration:

You see the central system bus, which connects the CPU with RAM. A bridge connects the I/O busses with the system bus and on to RAM. The bridge is part of the PC chip set.

3 different I/O busses
The I/O busses move data. They connect all I/O devices with the CPU and RAM. I/O devices are those components, which can receive or send data (disk drives, monitor, keyboard, etc. ). In a modern Pentium driven PC, there are two or three different I/O busses:

• The ISA bus, which is oldest, simplest, and slowest bus.
• The PCI bus, which is the fastest and most powerful bus.
• The USB bus, which is the newest bus. It may in the long run replace the ISA bus.

BIOS is a typical link in the IBM compatible PC design. The BIOS programs control hardware, the user (programmer) controls hardware via a call to BIOS.

BIOS typically occupy 1 MB, and the programs are saved ROM chips on the system board.

During start-up, BIOS is read from ROM chips. That information is supplemented with the system data saved in CMOS. Furthermore, there is BIOS code on the expansion cards. The expansion cards are external hardware, as interpreted by the system board, and the BIOS code, which is linked to the expansion card, must be included in the configuration.

Therefore, this expansion card ROM is read during start-up, and the program code is woven together with other BIOS data. It is all written into RAM, where it is ready for the operating system, as you can see here:

Otherwise, the BIOS routines are not always in use. They can be regarded as basic program layers in the PC. Many programs routinely bypass BIOS. In that case, they “write direct to hardware”, as we say. Windows contains program files, which can be written directly to all kinds of hardware – bypassing BIOS routines. One example is the COM ports. If you use the BIOS routines connected with them, you can transmit only at max. 9600 baud on the modem. That is insufficient. Therefore, Windows will assume control over the COM port.

• You hear the fan motor starting. There are one or more cooling fans in the PC. They produce a whirring sound.
• After a few seconds, text starts to scroll on the screen.
• Now the PC tests and counts the RAM. You see a number on the screen. It increases in size.

To understand the working of the PC, it is useful to study the PC start-up process. Those are events, which take place from power-on until the PC is ready to work. Remember, the PC can do nothing without receiving instructions. These instructions are commands, which are sent to the CPU. During start-up, the PC reads the commands in this sequence:

• First it receives commands from the ROM chips. Those chips are inherent in any computer. They contain the POST and BIOS instructions, which we will look at shortly.
• Next, the operating system is read from the hard disk (or from floppy drive A). This is called the boot process.

The ROM chips
ROM (Read Only Memory). The ROM chips are on the system board. They contain system software. System software are instructions, which enable the PC to coordinate the functions of various computer components.
The ROM chips contain instructions, which are specific for that particular system board. Those instructions will remain in the PC throughout its life. They will usually not be altered. Primarily, they are start-up instructions. There are different parts in the start-up instructions. For most users, they are all woven together. You can differentiate between:

• POST (Power On Self Test)
• The Set-up instructions, which connect with the CMOS instructions
• BIOS instructions, which connect with the various hardware peripherals
• The Boot instructions, which calls the operating system (DOS, OS/2, or Windows)

All these instructions are in ROM chips, and they are activated on by one during start-up. Let us look at each part.

POST
Power On Self Test is the first instruction executed during start-up. It checks the PC components and that everything works. You can recognize it during the RAM test, which occurs as soon as you turn power on.
As users, we have only limited ability to manipulate the POST instructions. But certain system boards enable the user to order a quick system check. Some enable the user to disable the RAM test, thereby shortening the duration of the POST. The duration of the POST can vary considerably in different PC’s. On the IBM PC 300 computer, it is very slow. But you can disrupt it by pressing [Esc].
If POST detects errors in the system, it will write error messages on the screen. If the monitor is not ready, or if the error is in the video card, it will also sound a pattern of beeps (for example 3 short and one long) to identify the error to the user. If you want to know more of the beeps, you can find explanations on the Award, AMI and Phoenix web sites.
POST also reads those user instructions, which are found in CMOS:

CMOS
CMOS (Complimentary Metal Oxide Semiconductor) is a small amount of memory in a special RAM chip. Its memory is
maintained with electric power from a small battery. Certain system data are stored in this chip. They must be read to make the PC operable. There may be 100 to 200 bytes of data regarding date, time, floppy and hard disk drives, and much more. CMOS data can be divided in two groups:

• Data, which POST can not find during the system test.
• Data, which contain user options.

For example, POST cannot by itself find sufficient information about the floppy drive(s). Floppy drives are so “dumb,” that
POST cannot read whether they are floppy drives or not, nor what type. About the same goes for IDE hard disks, while EIDE hard disks are a little more “intelligent,” However, POST still needs assistance to identify them 100% correctly.
The same goes for RAM: POST can count how much RAM is in the PC. However, POST cannot detect whether it is FPM, EDO or SD RAM. Since the CPU and BIOS reads data from RAM chips differently, depending on the RAM type, that type must be identified.

Configuration
The PC must be configured, be supplied with this information. That is done in the factory or store, where it is assembled. This information is stored in CMOS, where they stay. CMOS data only need to be updated, when different or additional hardware components are installed. This could be a different type hard disk or floppy disks or an new RAM type, Often he user can do this.
Other data in CMOS contain various user options. Those are data, which you can write to CMOS. For example, you can adjust date and time, which the PC then adjusts every second. You can also choose between different system parameters.
Maybe you want a short system check instead of a long one. Or if you want the PC to try to boot from hard disk C before trying floppy disk A, or vice versa. These options can be written to CMOS.
Many of the options are of no interest to the ordinary user. These are options, which regard controller chips on the system board, which can be configured in different ways. Ordinarily, there is no need to make such changes. The system board manufacturer has already selected the optimal configurations. They recommend in their manuals, that you do not change these default settings.
We can conclude, that CMOS data are essential system data, which are vital for operation of the PC. Their special feature is, that they are user adjustable. Adjustments to CMOS are made during start-up.

Here is a list of the PC components. Read it and ask yourself what the words mean.. Do you recognize all these components?

• Mainframes are the very largest computers – million dollar machines, which can occupy more than one room, An example is IBM model 390.
• Minicomputers are large powerful machines. They typically serve a network of simple terminals. IBM’s AS/400 is an example of a minicomputer.
• Workstations are powerful user machines. They have the power to handle complex engineering applications. They use the UNIX or sometimes the NT operating system. Workstations can be equipped with powerful RISC processors like Digital Alpha or MIPS.
• PC’s are the Benjamin’s in this order: Small inexpensive, mass produced computers. They work on DOS, Windows, or similar operating systems. They are used for standard applications.

The point of this history is, that Benjamin has grown. He has actually been promoted to captain! Today’s PC’s are just as powerful as minicomputers and mainframes were not too many years ago. A powerful PC can easily keep up with the expensive workstations. How have we advanced this far?

The PC’s success
The PC came out in 1981. In less than 20 years, it has totally changed our means of communicating. When the PC was introduced by IBM, it was just one of many different micro data processors. However, the PC caught on. In 5-7 years, it conquered the market. From being an IBM compatible PC, it became the standard.

If we look at early PC’s, they are characterized by a number of features. Those were instrumental in creating the PC success.

• The PC was from the start standardized and had an open architecture.
• It was well documented and had great possibilities for expansion.
• It was inexpensive, simple and robust (definitely not advanced).

The PC started as IBM’s baby. It was their design, built over an Intel processor (8088) and fitted to Microsoft’s simple operating system MS-DOS.
Since the design was well documented, other companies entered the market. They could freely copy the central system software (BIOS) and the ISA bus, since they were not patented. Slowly, a myriad of companies developed, manufacturing IBM compatible PC’s and components for them.
The Clone was born. A clone is a copy-machine. A machine, which can do precisely the same as the original (read Big Blue – IBM). Some of the components (for example the hard disk) may be identical to the original. However, the Clone has another name (Compaq, Olivetti, etc.), or it has no name at all. This is the case with “the real clones.” Today, we differentiate between:

• Brand names, PC’s from IBM, Compaq, AST, etc. Companies which are so big, so they develop their own hardware components.
• Clones, which are built from standard components. Anyone can make a clone.

Since the basic technology is shared by all PC’s, I will start with a review of that.