Got a question....PDP? VAX?

From: Jerome Fine <jhfine_at_idirect.com>
Date: Mon Mar 15 08:54:35 1999

>Lawrence Walker wrote:

> On 13 Mar 99 at 1:11, Buck Savage wrote:
> > Larry:
> > Given your start in computing in the mid-fifties, you must now be
> > in your mid-sixties. I was one who started in the early 70's (as
> > early as my 12-th year) and am now 41. The machine was
> > an IBM 370/155 with 1MB of main memory, and the language was
> > APL (APL*PLUS from Scientific Time Sharing).
> Pretty close, I'll begin my 63rd year this summer. I never spent any great
> length of time on computers untill the early 80s and then the late 80s till
> now. Being closer to it ,you've likely seen much more computing than I.
> I did use a slide-rule at college tho. :^))

Jerome Fine adds his $ Can 0.02:

While I am a tad shy of the lofty age that Larry has attained (I am now
a magic 60 years young), I started with computers back in 1958 when
I had a summer job with an insurance company. In 1960, I actually
started to write programs and ran the IBM 650 at the University of
Toronto when I had dedicated time during the mid-night shift.

I can't remember when the IBM 704/709 systems were introduced
(never used one - I believe they were built with vacuum tubes), but
after trying an IBM 1620 for a while (I used the 1401 as well, but
only as an input device for the =>) the University of Toronto acquired
an IBM 7094 (a speeded up 7090 which was the transistor equivalent
of the 709). The IBM 7094 was used around 1963-65.

Naturally, all of the above systems used 80 column punched cards
and programs were written, then punched onto the cards which were
fed into the systems. The output was punched cards and printed paper.

> > Well, it is true that the machines of the 50's were the so called
> > Unit Record devices and the low quality mainframes of the day,
> > like the 7090, the 1620 (if I get the numbering correct) and the
> > 1401 (all of IBM fame). I had the privilege of working with such
> > Unit Record devices, and even to have programmed a few of the
> > old style plug boards.
>
> > I am greatful that those days are over!
> >
> > William R. Buckley
> >
> Actually, I got a kick out of wiring up plug boards. Tangible logic.
> Why I prefer Hardware over Software.

I had the opposite reaction. I got a kick out of getting the program
to work correctly. Why I prefer Software over Hardware. That is
NOT a criticism!! Just a personal preference. I think that part of the
reason is that once a program is perfected (small ones can easily be
perfected and shown to be perfect), it never goes wrong! But with
hardware, the possibility always exists!

NOW, I must admit that I have had far more software bugs over
the years than hardware bugs and the actual number of hardware
bugs on a computer which was supposed to be running has been
less than the fingers on one hand. And I have had my share (in the
early days when we programmers think we can't do anything wrong)
of blaming the hardware when it was my own program that was at
fault. In general, if you can't easily prove immediately that the hardware
is the problem, then 99.9% of the time it is not.

The only time I ever really had a difficult one to figure out was when
I shifted from a PDP-11 to a VAX. The program in assembler on
the PDP-11 was using the ADD instruction in a non-standard manner.
As a result, the overflow and carry bits had an unusual meaning. On
the VAX, the result was very slightly different with respect to those
bits and the program had to alter the algorithm very slightly as a result.

Sincerely yours,

Jerome Fine
RT-11/TSX-PLUS User/Addict
Received on Mon Mar 15 1999 - 08:54:35 GMT