It is easy to forget that new technology is often worth more than the
high initital cost. When I was a young analyst crunching numbers, we
used mechanical calculating machines with thousands of moving parts.
They had to be on maintenance contracts to keep them running for $60
(that's 1970 $) per month. I did some research when electronic
calculators became available. One model actually used a B/W CRT, making
it a huge thing. I recommended that we buy some much smaller ones using
individual Nixie numeric display vacuum tubes and discrete logic on
complex circuit boards. You can buy equivalent computing power (except
they were 10 place machines) for less than $5 today. However, they were
a bargain, as they paid off their $600+ purchase price in lack of
maintenance in about 10-11 months and lasted for years. They were also a
big boost to productivity because they were FAST. We used to joke that
one guy with a particularly slow mechanical model kept it because he
could crank in a full length long division problem, leave work 15
minutes early and no one would know he left early.
I later got one of those supposedly very expensive XTs at work. I think
it paid for itself in increased productivity in a few months. The 10mb
HD was pure luxury after living with floppies. I didn't order the 3/5"
floppies because all other computers in the company had 5.25" floppies
and data exchange was still by sneaker net. Then we got PC3270s, IBM PCs
with a card to connect them direct to the SNA network and mainframe
timesharing, where I had created many applications. That was heaven,
both resources on one screen at the flick of a key. Then I had thin
Ethernet put in our department while IT was sitll wasting time with
Token Ring (well, it ws IBM, and nobody ever got fired for buying Blue,
right?). We had star topology Ethernet (the current form before it was
standardized) before IT figured out how far behind the 8 ball they were.
The point is that each of these technologies was apparently very
expensive at the time and each enabled both increases in productivity
and even doing things that just weren't economically feasible before and
were worth way more than they cost. I'm talking ratios of thousands to
one, in some cases.
Moose
chucknorcutt@xxxxxxxxxxx wrote:
>Hmmm. My memory is apparently not so good. (big surprise, eh?). I was
>just digging around and discovered that the IBM PC/XT (model 5160)
>announced in March 1983 offered support for (optional) 3-1/2" 720 KB
>floppy drives. They've been around longer than I remember but since the
>PC/XT in pretty basic form would cost about $7,000 list probably nobody
>ordered the optional floppies :-)
>
The olympus mailinglist olympus@xxxxxxxxxx
To unsubscribe: mailto:olympus-request@xxxxxxxxxx?subject=unsubscribe
To contact the list admins: mailto:olympusadmins@xxxxxxxxxx?subject="Olympus
List Problem"
|