We are both of the same 'generation' of programmers. In the center I worked
at there were many PDP machines around; our system was based on a Data
General Nova Eclipse 3/12 - 16-bit, 32K words of RAM with physical board
mapping to break the 32K addressing limit. We programmed in pure assembly
also, although I started a trend of doing design using integer BASIC to prove
algorithms prior to eating system time testing new code. Wrote both a
pseudo-assembler to run under DOS and a disassembler to assist with
interpreting memory dumps from the system.

In those days I was multi-lingual: I wrote in Honeywell 6000 assembly, Nova
3/12 assembly, Z-80 and 6800/6809 assembly. Now I struggle to keep up with
the changes in VB and VBA.

"Ron Rosenfeld" wrote:

On Wed, 1 Oct 2008 05:00:11 -0700, JLatham <HelpFrom @
Jlathamsite.com.(removethis) wrote:

Time is usually the price you pay for the convenience of such things; be it a
more compact way to code something or a move to a higher-order language. I
come from a background where processing time was critical (real-time air
traffic control radar systems software) and so I'm always looking to try to
make the code efficient in terms of speed. I'm sure I don't always succeed,
and in this forum I often code for clarity rather than speed. But the speed
of today's processors helps get you beyond the short comings of the languages
or the programmer's skills.

You're right about the faster processors.

There was a time when I was coding for real time acquisition of multiple
channels of physiologic data, and this was on a PDP-like machine many years
ago. Assembly language coding was the norm, then, and was required to make
things work.

But today I'm more of the opinion that "perfection is the enemy of good enough"
:-))

--ron