This thread has become shanghaied down a path different that the one I
intended to take.
My point was that you didn't have to write code when you bought a computer
that was intended to be used as a computer in order just to get it to run your
OS and applications. It didn't take a rocket scientist to figure out that you
could buy an FDC from the same vendor as provided your CPU and minimal I/O,
and then you could be off and running. It didn't matter whose memory you
used, so long as it worked at the speed required by the CPU. If you wanted to
pick just any combination of boards, yes, you were on your own. It took a
real fool to put himself in that situation since the cost wasn't much
different. between manufacturers of similar cards for the simple reason that
the manufacturers had to compete for your business. Sure, some vendors
"locked you in" with their odd-ball hardware/software combinations, but you
didn't have to fall into that trap.
When you opened the box with your COCO, what useful work would it do with the
$399 you had just spent? Could you write a letter? Could you write and
compile a Fortran program? Could you save your work in any meaningful way?
Given that you had a printer, could you attach it and use it? What software
was there, that you could install and use? How and where would you install
it? When you finally decided you had to build your own hardware and write
your own software, wouldn't it have been easier to use a wirewrap card and a
CPU chip and start from scratch rather than having to work around all that
stupid, Stupid, STUPID hardware they used? ... hardware you had to work
around, probably had to disable, before you could do anything sensible? That
certainly was the case if you wanted a reasonable display. That certainly was
the case with the TRS-80. Still, people did it. People paid in cash and
endless evenings and weekends. Now, if they wanted to be amused, making
something into what it was not, that's fine, but if you wanted a computer, the
hardware didn't cost less if you used one of the "toy"-based systems, and the
software wasn't any more available than if you'd used a "real" computer. It
probably would have cost less to use an Apple with a 6809 board in it (yes,
there were some of those) than to make a COCO into a comparable computer.
IIRC, RS computers were always priced near the top, while their quality was
near the bottom. For something equal to, say, and Apple][,. I'd say you'd pay
nearly twice what a well-purchased Apple][ would have cost for a RS product of
nearly similar actual capabilities. RS never did build something genuinely
intended for expansion though, did they?
Dick
----- Original Message -----
From: "Jeff Hellige" <jhellige_at_earthlink.net>
To: <classiccmp_at_classiccmp.org>
Sent: Thursday, April 25, 2002 5:39 AM
Subject: expansion differences (was Re: Micro$oft Biz'droid Lusers)
> >Open the box. Open the S-100 mainframe. Open the individual card boxes,
> >extract the cards and plug them into the mainframe. Open the carton the in
> >which the FDD enclosure traveled. Extract the enclosure, insert the 8"
floppy
> >drives. Attach the cables to their respective sockets, at each end, attach
> >power cables, power on the Terminal, power up the S-100 box, Power up the
> >FDD's, insert the boot disk, watch the lights bling, ... WHAT? no display?
> >... oh yeah... attach the serial cable to the terminal. repeat the
process...
> >and 64k CP/M 2.2 is booted. Total elapsed time, 4 weeks of shopping, 1
week
> >for shipping, 1.5 hours fiddling with boxes, cables, enclosures, etc. 45
> >seconds to boot the 1st time. ... done... Cost? well, the two Mistubishi
8"
> >DSDD drives cost $479 each, the terminal cost $753, shipped the CCS
hardware,
> >including CPU, FDD, 64KB DRAM, 4-additional serial port card, 4 parallel in
+
> >4 parallel out card, mainframe, FD box (the one I sent you) $1479, I think.
I
> >don't remember what the shipping cost was, but that was in '79. Later the
> >next year I attached an ST-506 drive.
>
> Looking at the above and then looking at how one would expand
> a CoCo or any of the early 'home' micros, I fail to see the
> difference other than the orientation and placement of the exansion
> buses. Both required you to add expansion cards for greater
> functionality. One happened to house the cards internally while the
> other didn't. You list a total of $3190 for the system above, that's
> a far cry from the $599 for the TRS-80 Model I or $399 for the
> original CoCo. Another thing to remember is that you list a system
> made up primarily of one vendor's cards, that being California
> Computer Systems. This made things quite easy, especially concerning
> booting CP/M. Take a more typical hobbyist S-100 bus machine though,
> with it's mix-and-match selection of S-100 bus boards, and the new
> owner will likely have to spend quite a bit more time than 45 seconds
> to get it to boot the OS for the first time. Did you have to expand
> the CoCo or other 'home' 8bit machines to do useful things? No. Did
> you have to spend time expanding most S-100 bus machines in order to
> do anything useful? Yes. The higher level of integration coupled
> with the much lower price is what opened the market up. Both systems
> are 'real' computers, both capable of similar uses, they're just
> aimed at different markets.
>
> Jeff
> --
> Home of the TRS-80 Model 2000 FAQ File
> http://www.cchaven.com
> http://www.geocities.com/siliconvalley/lakes/6757
>
>
Received on Thu Apr 25 2002 - 18:36:49 BST