[Guest Blogger] 40 Years in the Field (Part 10)

 

Graham Jones (Surrey, British Columbia, IT Pro)

My entry into the PC world started with the Commodore 64. I bought the “kit”; C64 computer, 1541 disc drive and 1084 RGB monitor (good monitor which did service for many years). All excited I took the “kit” home and set it up only to find myself in constant competition with my 10 year old son. Suffice it to say that it was very clear where his career was going to lie. With many teenagers you can’t get into their bedroom for the “debris” on the floor. In his case it was computer “debris”. Like most enthusiasts I was more interested in how it worked and “tweaking” it than actually using it. I found myself interested in software copy protection schemes. Can you say “cracking”? Because the 1541 had its own 6502 processor that opened up various anti-copying possibilities. Equally it meant that it could be “programmed”. In fact, if you connected 2 1541’s together (they were daisy chained off the computer) you could start copying from one to the other and actually disconnect them from the computer. However, my interests were in the in-memory schemes once the software had been loaded. Unlike some I wasn’t interested in “cracking” the schemes for profit. It was pure technical curiosity. By way of example, I discovered that with one scheme the software actually overwrote itself in memory several times eventually producing the real executable code. The tools were a disassembler, lots of time, learning when to hit the “break” key and knowledge of the 6502 processor instruction set. As you probably already guessed my son eventually “won” the “kit”. He wrote a C64 program for his school Science Fair but they had to disqualify him because nobody knew how to judge it (:. To this day there are many C64 enthusiasts (eg. https://www.c64.com/, https://www.zzap64.co.uk/c64/c64emulators.html, etc.) who write emulators for modern PC’s, play C64 games, etc..  It would be some time before I had another home PC. My son moved on to the Amiga 500 and writing his own OS while in High School.

Things were jogging along fairly nicely until the demise of Prime was rumoured. The user community, which by now was quite large, insisted that PDMS be made available on another platform just in case. In fact at one point a source code escrow agreement was put in place in case the vendor decided against it. DEC VAX was the choice. So the next big decision was “do we stay or move?”. We decided to move which was just as well because Prime did eventually fold. We had a hard time even giving the Prime away. I think that it eventually went to a college. Of course changing hardware meant learning a new OS. The big task was converting all of the “add-ons” that had been developed. Although the DEC command language was one of the best it didn’t have all of the features of PRIMOS. On the other hand I believe that VMS was one of the best OS’s that was ever written. If DEC had made VMS “open source” and available for other hardware who knows what might have happened. Fortunately, the PDMS programming language had substantially improved by this point and so we focused our attention there. Clearly it was better to work within the software than try and “wrap” something around it.

PDMS would go through 2 more major hardware changes with the increase in computing power and graphics processing in hardware. The first major change was a move to Unix based Silicon Graphics (SGI) workstations. This also meant a move from a central computer connected to essentially “dumb” terminals to a central DB which was partitioned by geographic sections of the plant and by engineering discipline, eg. piping, electrical, etc.. This permitted multiple people from different disciplines to work on the project at the same time but only permit one person per partition to make updates. Any number of partitions could be used as “read only” for reference purposes. A nightly “interference check” to look for spatial clashes was run in batch mode followed by clean-up the following day. Improvements in the graphics processing power eventually took us into real time rotation of colour shaded models (hidden surface removal), which were used for project team design reviews and “wowing” the clients before and during projects.

pdms.jpg

Wire frame representation was adequate for most design work and could be rotated much faster. The eventual move was into the PC world when processing power went up and costs came down. The initial challenge was the multitude of graphics cards and their different drivers. Plus, as all of us who have been through this will know, a PC is a PC, right? Wrong!!! We now had very much cheaper hardware but more technical problems and an ever changing landscape. It is only in the last few years that “standards” has made things much more predictable.

One of the benefits of a large user base is the exchange of ideas. A few years after we started working with PDMS an annual “worldwide” user conference was inaugurated held somewhere in Europe (usually a nice location :)) which I attended. The most valuable part was the presentation of papers by the member companies. Most years we presented a paper of some kind. The people who consistently astounded the rest of us with the “jaw dropping” things that they had done was Electricité de France (EDF) which is the French nuclear power industry (ie. oodles of money!). Incidentally they were the only company running on IBM and had paid the vendor for the port. Each year a different company hosted the event. One of the conferences was held in England and we were invited to the SGI research labs to see “virtual reality” demos. I got to walk around a North Sea oil rig model built in PDMS wearing a “virtual reality” headset. It was really COOL and eerily real! If you stood at the top of stairs and looked down you really felt like you were going to fall and you ducked if something got too close to your head. The other demo was a group of us sitting in a room surrounded by 3 screens (stitched together by the computer) using a joystick to control the “walkthrough”, which was equally impressive for the purposes of doing team design reviews. Eventually there were enough NA clients that an annual conference was started here. I found myself on the NAPUG Board and hosted a conference in Vancouver. Some of the other “lousy” venues were Orlando, New Orleans, Marina del Rey (near LA where we had an earthquake that threw me out of bed) and San Francisco.

While all of this was happening with PDMS other computer based engineering activities had sprung up both within our office and via our parent company which was no longer CIL but a large UK based engineering company. In part 11 I will outline the developments in the application of “data processing” techniques which made “Concurrent Engineering” more of a practical reality.

Cheers

Graham J.