CSIRO Computing History, Appendix 7: Artefacts
This page/post is under construction.
Last updated: 11 Sep 2019.
Robert C. Bell
Appendix 7: Artefacts
These are artefacts and publications held by Rob Bell at Clayton. Some are held under his desk, and some are in the basement of Building 001 – IT Build room archive area. Archive boxes series A and B are held in the IT Build room archive area, while series C boxes are in the Records store in building 001, room RCB.02.
Sample punched cards
Sample magnetic tapes/cartridges
Sample coding forms
Special ruler for use with 11” x 15” paper
Rock and three-ringed binders
Abacus (small and large)
Model of Cray Y-MP4
Cray Y-MP4 front panel
Cray Y-MP CPU (costing about $250k around 1995)
Circuit boards from Cray Y-MP IOS – Cray X-MP technology
Cray J90se CPU board
CPU from NEC SX-6
Commemorative book of the build of Cray Y-MP2/216 SN 1409
CDC Manuals – complete set of user manuals from 1985 for the CDC Cyber 76 (SCOPE 2)
CDC Manuals – complete set of user manuals from 1990 for the CDC Cyber 205 (VSOS)
CRS/DCR/Csironet News – almost complete set
DCR/Csironet Cybarites – technical newsletter
DCR/Csironet Staff News
Microfiche – Csironet manuals
Microfiche – Csironet minutes of DAD (Design and Development) Committee
Microfiche – Cray UNICOS manuals from about 1990
On-line (cherax=ruby ~bel107/*tests*): captured output of monitor job from around 2004 to present – gives snapshots of various systems.
On-line (cherax=ruby ~bel107/UNICOS): captured output of monitor job from around 1990 to 2004 – gives snapshots of various systems.
On-line (cherax=ruby:~bel107/Csironet): captured files from Csironet systems, including Rob Bell’s software projects, and files dating back to the late 1960s.
A list of artefacts with brief descriptions can be found at https://hpc.csiro.au/users/70294/CSIRO.Artefacts_list.docx
John Morrissey has a module from the CDC Cyber 76 processor. Here are two pictures:
Here is a montage of programming tools from the 1970s and 1980s:
- a coding form: programs were written on such forms, and then transferred to cards by the author or by a ‘punch girl’ – apologies, but that was the term used, I’m afraid. This particular one was a CDC Fortran – the rigid column-limited form of old Fortran is evident.
(C in column 1 for a comment, columns 1-5 for statement numbers, column 6 for a continuation flag, columns 7-72 for statements, and columns 73-80 for sequence numbers (rarely used in the CSIRO environment).
- an automatic pencil – these followed on from clutch pencils and conventional pencils, and were a considerable (?) saving in that they were fine enough not to need sharpening.
- an eraser – most people coded in pencil, as mistakes were common. Tony Davies at Aspendale used to code using a fountain pen.
- a flow-chart template – flow charts were on the way out by the 1970s, as structured programming became the rallying cry for software development processes. “GOTOs were considered harmful!” Here is the first page from a flowchart from 1974.
- a punched card – this one is a Fortran one, and fairly unusual as being full-coloured – a plain buff colour was more common.
- a highlighter pen – these came into use around the end of the 1970s. Most output was on 11″ by 15″ paper, and a highlighter pen was ideal for highlighting important information or errors without obscuring the print.
- a special ruler, primarily for counting character positions on printed output. Output was printed 10 characters per inch horizontally, and 6 characters per inch vertically, and the ruler provided a quick way to determine the position of a character relative to the start of a line on a page without having to count each character. The ruler saved a lot of effort when trying to set up output using pre-Fortran 77 format statements:
98 FORMAT(29H OPTIMUM RELAXATION FACTOR = ,1PE18.11,24H RESIDUAL MUL $TIPLIER = ,1PE18.11)
Here are some examples of pages of output (“printout”) on 11” x 15” fanfold sprocketed paper as used in line printers
a) Two pages showing a table of prime numbers to 100 000. I think this was produced on a CDC 3400 undergoing acceptance testing for DSD in about 1965, and has a clever encoding scheme(see the key at the top of the second page).
b) A listing of numbers, with a line-printer graph beside it. This was run on the Monash CDC 3200 in 1969.
c) A sample output from the Cyber 76, in 1983. This particular job wrote fixed length records of source code to tape for transport to another site running an IBM mainframe. SCOPE2 job output started with a dayfile, which gave a summary of all the commands executed and the results of running the commands, with time stamps and running totals of CPU time used. Also included was a detailed summary of the charges!