Posts Tagged ‘history of the computer’

History of The Computer – Codes For Data Transfer – Part 1 of 3

We have seen, in the History of the Computer series, how machine code is used to operate a computer so that it can use 1s and 0s to interpret a program’s intentions. But what about a code for human communication with the computer? We can variously write and read to and from different media, such as paper tape, magnetic tape, disk or teletype, but we need some means of converting English (or another language) into a form which can be carried on these media.

You may recall the old Western movies where the James gang shoot the telegraph operator so he can’t call the Marshall for help. Or the movies about the ’29 Wall St crash where the tycoon reads the bad news on the ticker tape and promptly leaps out of 32nd floor window. Both of these are early examples of a code being used on a medium. In this case morse code on the telegraph, and an analog code produced from a keyboard at the transmitting point, also sent by telegraph. The ticker tape, so called because of the noise made by the printing mechanism, carried the Stock Code, as used in today’s stock market, and a quote.

A later code used for paper tape, the baudot code, named after the French inventor, Baudot, was a five bit code, punched across the tape, giving a 1s and 0s pattern to make up the 26 letters of the alphabet. The tape was an inch wide and came on big reels. A sprocket hole was punched in the tape so that it could be moved through the reader.

As we know from our look at binary codes, 5 bits will give us a total of 31 characters (1+2+4+8+16) plus zero. With 26 being used for the English alphabet, the other characters are used for Carriage Return, Line Feed, etc. These are needed to operate the typewriter used to print out the information. Another character, Bel was used to ring a bell on the typewriter to alert the operator.

Later a Shift character was incorporated, similar to the Shift key on a typewriter, so that following characters would indicate numbers. Another shift code would switch back to letters. This Baudot Code was used on teletype machines to transmit data from one location to another. This was later accomplished by first punching the tape, and then transmitting it by feeding it through a ‘high speed’ reader. These speeds were gradually increased as the technology improved, and were generally expressed as the ‘Baud Rate’, roughly equivalent to bits per second, also named after Baudot. Most problems came from punching errors and synchronizing transmitter and receiver.

In part 2 we look at how codes are used in computers.

Author: Tony Stockill
Article Source: EzineArticles.com
Provided by: Digital Camera News

History of the Computer – Codes For Data Transfer – Part 3 of 3

DATA TRANSMISSION CODES

We have mentioned teletype machines, these were the backbone of operator communication with the computer, but were also used for remote input of data. A phone line could be used to link a storeman in the warehouse with the main computer using a modem (Modulator-Demodulator) in much the same way you dial-up to your ISP (Internet Service Provider).

These modems typically worked at 110 Baud, the standard speed of the teletype, and sent or received one character at a time. They modulated the carrier on the telephone line in a string of bits in serial pattern (one after the other) for the code, usually ASCII, complete with start bit, stop bits (1 or 2) and negative or positive parity bit. Other codes were, for example, SOH (Start of Header), STX (Start of Text), ETX (End of Text), used to identify the following or previous characters. Of course these ‘control’ characters had to be correctly decoded and used to control the data flow.

At the computer data terminal this serial data was read into a shift register, gated by the control characters, until all bits were received, then read out from the register in parallel (all at once) to be stored in memory. At the warehouse the data was sent from the modem to the teletypewriter for printing.

Little has changed here, except the speed of operation. The codes are much the same, only upgraded for new types of equipment. As new technology is introduced, new access and control codes will be required, but the basic data remains the same.

DISK CODES

Magnetic disks and, later optical disks use similar recording techniques to magnetic tape, with one major difference – they record in serial format. The blocks on the disk incorporate an address code as well as the data and other codes we have mentioned, as the data on a disk is randomly accessible. We can specify an address and read the data from it. The same applies to an optical disk, which is used mainly for its capacity and portability. The data codes are of standard format.

PRINTER CODES

Printer codes used were a standard set of ASCII characters, and were restricted to the character set etched on the print drum, or band in use. A line printer was used, and one line of data was sent at a time, to be held in the print buffer and read out as required when the appropriate character was under the hammer. Once again control characters such as line feed and form feed were used.

Matrix printers were used for small scale print jobs usually accompanying data terminals away from the central computer. Once again the character set was restricted to the basic ASCII codes, and Control codes with printers from different manufacturers able to be used due to the standard codes (Interfaces were a different matter!)

Contrast this with the laser or inkjet printer today, Virtually any font and character set, color and pictures too. Transmission by Network, USB, Parallel or Serial port, or Wireless. Once again mainly control code differences to suit the medium and different handling for serial or parallel.

Author: Tony Stockill
Article Source: EzineArticles.com
Provided by: Programmable pressure cooker

History of the Computer – Codes For Data Transfer – Part 2 of 3

COMPUTER CODES

Various codes have been used in computing, EBCDIC, Fieldata etc., with various bit combination’s, and total bits between 5 and 8. This often depended on which proprietary computer system you were using, and what type of device you were using. Punched cards had a different code again. A standard code, ASCII, was eventually introduced. The American Standard Code for Information Interchange has 8 bits, and is used interchangeably on almost all systems today.

PAPER TAPE CODES

Paper tape was upgraded to 8 bits across, plus the sprocket, and used to boot computers, and to transfer data from one machine to another. For instance a NC, or Numerical Control, could use paper tape to relay instructions to a machine for cutting steel sheet shapes.

PUNCH CARD CODES

Punch cards had been used since the 19th century, and a couple of different formats had emerged, round holes, rectangular holes, 80 per card, 90 per card, different codes, etc. Once again it depended on whose computer you were using IBM, Univac, Control Data etc. However, the basic principle was the same. The ‘standard’ came to be the 80-column card, with 12 rows for each character. The 1 to 9 rows were numbers 1-9, with the 0, 11 and 12 rows being used for various special characters, shifts etc.

MAGNETIC TAPE CODES

Magnetic tape, initially on 10 inch reels of half inch wide tape, 2400 ft long, used variations of the computer codes. Initially 6 bits plus parity (7) were written across the tape. With the introduction of bytes this increased to 9 bits to incorporate the 8-bit byte plus parity.

Most variations here have been in the recording system, that is how the data is written to and read from the tape. These include NRZ (Non Return to Zero), PE (Phase Enabled) and, latterly GCR (Group Controlled Recording). The density possible has increased, so that the capacity of the tapes even in the present day small cassettes has greatly increased.

Transport methods, initially ‘Stop-Start’, to record each character, now more commonly ‘Streaming’, have developed in parallel. The tape has normally been written in blocks of various lengths with codes being required to control these, such as ‘Start of Block’, ‘Start of Data’ etc. Other codes are used to verify the data. The DAT (Digital Audio Tape) tapes used today use a different form of digital coding, but data codes are much the same.

As well as the parity bit in each character, there can be a parity bit, for example, for all the ’3′ bits in a block. This is known as ‘longitudinal parity’, or parity along the tape, a bit for each ‘row’ of bits in the block. You can also have a CRC or, cyclic redundancy check, which considers all the bits in the block, in a complicated formula.

All these check codes are necessary because magnetic tape is an inherently unreliable medium, as you probably know from listening to your old cassettes! This has caused ECC or error correction codes to be developed, similar to those used in memory. In fact there are so many extra codes on the tape you wonder if there is enough room for data.

In part 3 we look at data transmission, disk, and printer codes.

Author: Tony Stockill
Article Source: EzineArticles.com
Provided by: Digital TV, HDTV, Satellite TV