Kitz ADSL Broadband Information
adsl spacer  
Support this site
Home Broadband ISPs Tech Routers Wiki Forum
 
     
   Compare ISP   Rate your ISP
   Glossary   Glossary
 
Please login or register.

Login with username, password and session length
Advanced search  

News:

Author Topic: A historical tale  (Read 4391 times)

waltergmw

  • Kitizen
  • ****
  • Posts: 2776
A historical tale
« on: April 27, 2010, 12:12:19 AM »

Gentlefolk,

SinBin had become confused with bits and bytes so I thought a practical illustration might help. That led me on to some historical ramblings which are posted below the link to a photograph of paper tape.

Kind regards,
Walter

http://www.scribd.com/Unipunch-Open-amp-Paper-Tape-2265/d/30540707


The picture shows a unipunch with the retaining lid off, a section of paper tape and an adhesive patch used to repair torn tape.

The paper tape has 8 holes across it, sometimes known as channels, plus a sprocket hole which was originally used to pull the tape through the automatic punch and then the tape reader. Latterly it was used in optical or capacitive readers as the strobe to read the paper tape character or byte. i.e. a row of holes.

To recap one hole is a digital bit and a row of 8 bits make one byte. The least significant or low order bit is on the right and the most significant or high order bit is on the left. In a binary word, the bits have increasing values from right to left of 1, 2, 4, 8, 16, 32, 64,128.

So reading the set of bytes near the patch the top row represents the digit 1, the next 2, the third 3 and the fourth 4 etc. so you now know how to convert binary to decimal !

Paper tape was also used to enter alpha-numeric data according to a standard code. The one illustrated in the top two groups is American Standard Code for Information Interchange or ASCII, pronounced askey. In this illustration bit 8 is always added to make every character even parity, i.e. bit 8 is added to every character which would otherwise have an odd number of bits. Bits 6 and 7 are used to differentiate between upper case, lower case and numeric characters. The centre group starting at the top represent the upper case characters A, B, C, D & E. The top group are the numbers zero to 9. The “all ones” character was the delete or null character so you could blot out an error. At this time there were no editors available so source code tapes in ASCII were edited by running the tape through a teleprinter reader and punching a new tape at the same time. Where a change was required the input tape was stopped, new characters were punched into the output tape from the keyboard, and the old tape was then pulled through to the point where the rest of the data was correct again. I wrote probably one of the first computer tape editors in the UK on a radar system using its tabular display. Later when floppy disks and then hard discs with a staggering capacity of 2.4 Mbytes were introduced, proper editors used a teleprinter keyboard for the editing commands such as C/JACK/FRED to change the first occurrence of JACK to FRED from where the cursor was. To find the names of the files within a folder it had to be printed out using a command such as
TI:=DK3:[154,154]*.*/LI i.e. list to the console the contents of disk 3’s User Identification Code 154,154 all file names with all version numbers.

Although very high speed tapes were used in Bletchley Park, the industry norm was 1000 characters (or 10 Inches of tape) every second. So the maximum data input speed of the tape reader was 8000 bits per second.

This was the technology in use towards the late 1960s. Before that, standard GPO teleprinters were used with 5 channel tape I think with a maximum speed of 10 characters per second. With only 5 channels it was not possible to produce even a full alphabet with the numbers and some basic punctuation. Two precious characters were used for case normal and case shift which provided a workable letter vocabulary. The printers often produced nonsense on incoming messages so the trick was to punch a tape from the incoming message, replay it and then inch the tape forward before adding a case normal or shift from the teleprinter keyboard.
Logged

silversurfer44

  • Kitizen
  • ****
  • Posts: 4421
  • Lord Muck
    • Ben Novice Weather
Re: A historical tale
« Reply #1 on: April 27, 2010, 08:10:34 AM »

Very interesting Walter. I did know about the tape basics but never knew how it all worked.
Your writings show the relevance to today's bits/buckets and so-on. Do I take it you were involved in the production of the tapes etc?
I find the subject quite fascinating as I knew about such things, but were never involved. I was not fortunate to have the hands on experience.
Logged
Colin II : It's no good being a pessimist, it wouldn't work anyway.

waltergmw

  • Kitizen
  • ****
  • Posts: 2776
Re: A historical tale
« Reply #2 on: April 27, 2010, 08:28:43 AM »

Hi Silversurfer,

I was indeed involved in air defence systems which were one of the few areas that could afford the technology.

For many years I had small burn marks on my fingers from winding the tapes up with a hand operated modified grinder wheel and then electric winders.

It never fails to amaze me how technology has progressed over such a short period. E.g. an iPhone with GPS and Sat Nav has far more computing power and facilities in a shirt-pocket sized device than the UK could then afford to defend itself, using room-fulls of equipment consuming kilowatts !

Kind regards,
Walter
Logged

silversurfer44

  • Kitizen
  • ****
  • Posts: 4421
  • Lord Muck
    • Ben Novice Weather
Re: A historical tale
« Reply #3 on: April 27, 2010, 09:20:47 AM »

Room sized equipment indeed, brings back a few memories. I used to service office machinery a few years back.
One of the places I frequented had the photocopier in what was the mainframe room. Fully air conditioned, whilst the rest of the office staff roasted. It wouldn't surprise me to hear I had more processing power on my desk right now than was in that roomful of very expensive equipment.
Ooh nostalgia, I bet you have many tales to tell Walter.
Logged
Colin II : It's no good being a pessimist, it wouldn't work anyway.

geep

  • Reg Member
  • ***
  • Posts: 452
    • My ST546 Statistics
Re: A historical tale
« Reply #4 on: April 27, 2010, 07:28:59 PM »

Hi Walter,
You've triggered some memories with your historical tale.

I remember programming an Elliott 905 which didn't have much memory. One program I wrote used paper tape as temporary storage, with the tape looped from the punch back to the reader. As the tape reader was MUCH faster than the punch, I had to have delay loops in the program to prevent the reader from breaking the tape. Some trial and error to get the timing right, with several tape breakages when I didn't.

At one time I could even read the tape directly by looking at it, without bothering to print it out on the old ASR33 teletype at 10 characters per second.

The airborne version of the 905, the 920, was used in Nimrods. It used mylar tape for robustness. Mylar tape could really do you damage.
 
And the Elliott 803 used the 5 hole teletype tape. I programmed it in machine code, 2 instructions in each 39 bit word (think about it!). Ah, happy days with the B-Line modifier.

Not to mention the Data General Nova. In the days before BIOS, you had to enter manually a bootstrap loader via 16 switches - one for each bit for the instruction word. I think the loader was about 20 or so instructions. I used to be able to remember the whole lot in my fingertips.

Cheers,
Peter
Logged

silversurfer44

  • Kitizen
  • ****
  • Posts: 4421
  • Lord Muck
    • Ben Novice Weather
Re: A historical tale
« Reply #5 on: April 27, 2010, 07:48:36 PM »

@Peter, I believe you describe where the boot up process came from.
i.e. pulling the system up by the boot straps.
Logged
Colin II : It's no good being a pessimist, it wouldn't work anyway.

waltergmw

  • Kitizen
  • ****
  • Posts: 2776
Re: A historical tale
« Reply #6 on: April 27, 2010, 08:16:16 PM »

@ Peter,

I had a brief time with some Elliott machines on a "man portable" military radar system I think called GL161.
The electronics used Germanium semiconductors which had to be oil-cooled. The support fibreglass framework to carry all the heavy multicore screen cables and oil hoses had to be seen to be believed!

I used Marconi Myriad 24 bit machines, then a horrible Plessey XL 9** built in telephone exchange racking, but then progressed to DEC PDP 8s and 11s.

** upon one occasion we discovered that there was a bug in the compiler which added the bit 24 modifer when it shouldn't have been there. We reported this bug to the manufactures and in due course I had to visit their factory to witness test the new version. They had developed a compiler tester with a test source tape which had to be fed in twice without any differences. If successful it then produced a binary tape from the punch which then was fed into the reader for verification purposes. This process took the best part of an hour if the tapes weren't torn. I watched this process with all due solemnity and respect. We were all very relieved when it passed its test. I then produced about 6 inches of tape containing a single pseudocode instruction and asked that it be compiled using the new version. Out popped the binary with the three characters making up the 24 bit word with the most significant (modifier) bit still set. I was not particularly polite to the senior departmental manager as I pointed out that they had failed to fix the bug we had told them about. The said manager grumped off to phone my manager with the complaint that I had failed to show him sufficient respect and had made him look very foolish !

You may have realised that I punched the tape yesterday from memory !

Kind regards,
Walter
Logged
 

anything