C-64 vs the world (vintage flamebait) (was Re: Micro$oft Biz'droid Lusers)

From: Richard Erlacher <edick_at_idcomm.com>
Date: Wed Apr 24 10:21:59 2002

see below, plz.

Dick

----- Original Message -----
From: "r. 'bear' stricklin" <red_at_bears.org>
To: <classiccmp_at_classiccmp.org>
Sent: Wednesday, April 24, 2002 12:27 AM
Subject: Re: C-64 vs the world (vintage flamebait) (was Re: Micro$oft
Biz'droid Lusers)


> On Tue, 23 Apr 2002, Richard Erlacher wrote:
>
> > Waidaminnute, waiddaminute ...
>
> I think you need to wait a minute, there, speedy.
>
> > The year I got my first PC/AT (I used 'em at work all the time, but didn't
get
> > one at home for a while.) I bought a 25 MHz clone (NEAT architecture)
with 8
> > MB RAM and a pair of 700 MB ESDI drives for $1080. That was with a
1kx768x16
> > VGA display board. I hacked it to work with a fixed-frequency
(sync-on-green)
> > monitor I had sitting around. It was WONDERFUL. That was in early '89.
>
> This struck me as being blatantly wrong, since it seemed to me (if I
> recalled) that in early 1989 your 8 MB of RAM would've run nearly $1k by
> itself.
>
A Denver area company called "Orca" (still in business, AFAIK) had prices much
lower. The motherboard cam with 4 MB in DIP form, and I had it maxxed out
using SIPPs. I know they had the "best" price we could find because I
designed a 5-ported 16/64MB buffer memory board for a high-capacity helical
scan tape drive (basically a modified VCR) sold by Honeywell Test Instruments
division) that used 1MB/4MB SIMMs. We bought the memory in sufficiently small
quantity (these drives were not that popular due to their price) 1MB SIPPs
cost about $35 at the time, though Arrow wanted $115. The price of a 1 MB
SIMM was not much higher than that ~$35 price in the Computer Shopper. Have a
look there and you'll see what I mean.
>
> So I checked the January 1990 issue of "Byte". These are directly from the
> cheap ads at the back, and representative of prices a year (loosely)
> after your conjectured purchase date:
>
I think your perception of dates and prices may be skewed somewhat by a number
of factors. I'll have a look at the CPU on that '286 board (yes, I still have
it.) to get a date code, but I'm comfortable that the 25MHz parts were
available in the '87-'88 time frame. Motherboard chipsets capable of
operating at that speed were not available until '89, and had to be selected
for 25 MHz operation. Further, the coprocessors, though they actually
operated at a lower rate, were not available at sufficient speed to accomodate
25 MHz CPU's until '89. I know I bought the '286 system BEFORE I went to work
at Honeywell, only because I ultimately bought a '386 motherboard and a bunch
of RAM from the folks with whom I'd established that procurement arrangement
on behalf of Honeywell. The '386 board +RAM cost less than the '286 board had
cost, since I got it at "cost," whatever that meant.
>
The ads in the back of BYTE, though interesting, were seldom the best-priced
ads. They were terribly expensive and nobody offering decent prices could
afford them. Further, they were 90-days or more out of date by the time they
were printed, so they were utterly useless if you were shopping in real time.
Computer Shopper was also plagued by long delay from submission to printing.
However, once you had an 800-number, you could easily shop for the best
prices, often ranging over a 300% span on memory and disk drives. It's too
bad you can't call back in time.
>
> Imprimis "766 MB" $3126 (p.303)
> Maxtor 8760 2579 (p.303)
> Micropolis 1558 1695 (p.311)
> ST4144R 605 (p.303)
>
The prices you show for the first two are quite a bit on the high side. A
couple of years earlier they'd have been in-range, but not in '89-'90. Maxtor
was selling their 1.2 GB drives in high volume by then. IIRC, Miniscribe was
plagued by financial woes, culminated by their "bricks" scandal, which
restored their liquidity, oddly enough, but ultimately led to their demise.
>
I was really busy at Honeywell at the time, so I had an associate of my former
business partner, who even today is still in the retail business, acquire and
integrate this stuff for me. He got the drives (Miniscribe 9760's) from
Miniscribe, since he was a former salesman from there. The second two of the
drives you list above, BTW, are not 700 MB types. The 1558 is about 300,
IIRC, and the ST4144R is the RLL that famous Seagate model that had extra
hardware built in just to prevent its use with RLL. The 4144R was, surprise,
surprise, 144 MB. The one (its MFM version) I'm thinking of was a ~96MB drive
which had the same head/cylinder spec's and looked exactly like the 4144R.

I gave the remaining Miniscribe 9760 to Emanuel Stiebler a couple of years
ago, IIRC, since my Lark Associates controller had gone tits up, and it was
the one that made the large drive useable under DOS by splitting it into
several pseudo-physical drives, which meant one had to use two controllers for
two drives. The other Lark Assoc. controller went with the other drive. WD
controllers, at least the ones I have, can't do that.
>
> 1Mx9, 30-pin SIMM $ 99 (p.312)
>
competely out of line for '89-90. I was very sensitive to memory pricing due
to what I was doing at the time. I started on the VLDS buffer memory board in
September of '89 and finished it in April of '90. The price I found for
HONEYWELL, completely unaccustomed to buying components on the "commodity"
component market, was about $35 in the limited quantity in which they bought
them. That was the "going rate" from "Orca."
>
> Base 25 MHz 80386 $ 850 (p.311)
>
I had a Harris 'C286-25. These cost about $385 including the 4 MB RAM, IIT
coprocessor, and CPU on a C&T NEAT chipset. I'd previously worked for a
company that had some pretty slick software to support the NEAT in order to
allow the use of the "adapter space" to load drivers. I'd found that the
'286/25 outperformed the '386/25 by about 10% despite the increased data path
width. Quite a surprise, but that's why I went with the '286.
>
> I also supposed, being the bargain-conscious person you are, that you
> might have been speaking of a 25 MHz 80286. I didn't offer a price because
> nobody was selling them, in 1989, OR in 1990. Harris did eventually manage
> to push the 80286 to 25 MHz, but I don't believe it happened until at
> least 1992.
>
While I was at Martin Marietta, we routinely used the 25 MHz CMOS part from
Harris in satellite hardware in '87-'88. These were selected parts, but were
common and readily available once a supply channel was established. The board
I bought was populated with chips labelled "25 MHz," though the CPU was
labelled 20 MHz. It worked fine, though I replaced the CPU with a selected 25
MHz part I had gotten as a sample some years earlier.
>
> I admit some margin of error is to be expected, but these prices---sampled
> fully eight or nine months after your stated date---do not come close to
> corroborating your statements. In consideration of this, I must humbly
> submit that you are totally full of shit, or at the very least have made a
> grave typographical error.
>
> As for the rest of you honking geese, here's an excerpt from a letter
> printed in the "Chaos Manner Mail" column, from the same issue of "Byte":
>
> "Dear Jerry,"
>
> "Over the years I've read several of your philippics against Unix.
> Today I have time to write a letter, so I'm going to put my two bits in.
>
> "Most Unix debates, when parsed, are recognizable as religious disputes
> between cult insiders and cult outsiders. The position of the insiders is,
> 'If you would only believe in Unix, you would be saved from the twin
> demons of low productivity and ugliness.' The position of the outsiders
> is, 'I can manage my productivity a damned sight better than you can, and
> I spit on your aesthetic pretensions---get lost.' Such conversations get
> really boring fast.
>
> [typos my responsibility]
>
> Can we please move on?
>
Clearly it's an old argument. I'm just as tired of it as you, but every time
I jump someone for introducing this "Of you wold only..." argument as
unrealistic, a hundred "honking geese" jump in to defend their
job-security-producing ultra-cryptic command-line-based baby, which, if it
were so great, wouldn't need them to defend it. That's the way it is with
religious arguments, I guess. I admit that I'm weak and can't let these
irrational arguments pass.
>
> ok
> r.
>
>
Received on Wed Apr 24 2002 - 10:21:59 BST