26 Apr
2002
26 Apr
'02
6:04 a.m.
-----Original Message-----
From: Richard Erlacher [mailto:edick@idcomm.com]
> That pretty much describes commodore disk drives,
yep. The
An external IDE drive would be hazardous thing to use
because
of cable length.
Didn't think of that -- you're probably right.
> handled all of the complexities of disk I/O, so
that the CPU didn't
> need to -- so there are good points to them.
What you may have had, then, is a toy with a computer
as a
peripheral. That
was probably quite a bit after the time reference of 1980.
Actually, I don't remember what CPU was in the 1541 drives, but I
seem to recall that it was actually more powerful than the computer.
I also remember somebody working on a way to get code into the drive
to be executed, but that's kind of fuzzy.
What I am focused on is where the intelligence to run
the I/O
resides. Once
So does the computer have to come with it, or simply allow space for
it on the inside of the case?
peripherals made great deal of sense. However, it
resulted
in ugly and
awkward packaging, which was addressed even later with more
elegant interface
standards, say, by 1986, with the standardization of SCSI.
Well, I certainly won't argue with that.
> Would you also have considered 9-track tape
"mass storage" for the
> time?
I'd tread lightly around that subject, unless your
Atari or
whatever, had a
9-track drive in '80 or so.
Obviously not, I'm just curious as to whether it would need to
be some form of "fast" disk-like storage, or if a high-capacity
tape would be ok.
Chris
Christopher Smith, Perl Developer
Amdocs - Champaign, IL
/usr/bin/perl -e '
print((~"\x95\xc4\xe3"^"Just Another Perl
Hacker.")."\x08!\n");
'
26 Apr
26 Apr
7:40 a.m.
Actually, I don't remember what CPU was in the
1541 drives, but I
seem to recall that it was actually more powerful than the computer.
More like "about the same" -- the 1541 is a 1MHz 6502, and the C64 is also
approximately 1MHz (slightly less for PAL, slightly more for NTSC).
The 1571 is 2MHz IIRC, but then the 128, which was its intended controlling
machine, can also run at 2MHz.
Many of the PET drives had multiple 650x CPUs, each handing different aspects
such as FDC operation, GCR coding, etc. When the 2031 and the 1541 arrived,
Commodore assigned all the tasks to one overworked CPU instead -- the CPU
alternates between FDC and IP mode on interrupts.
I also remember somebody working on a way to get code
into the drive
to be executed, but that's kind of fuzzy.
That's easy to do with M-W/M-R commands to alter drive RAM; it's just slow
to do transfers back and forth over the IEC bus, which fouls any useful
attempt at cooperative processing (best to hand the drive a separate task
to do all on its own), and there's not much RAM there -- only a few kilobytes,
most of which is being used by the drive itself, of course.
Several monitors featured "transparent" access to drive RAM where the
monitor would treat drive RAM as if it were main memory. The communication
speed, of course, invariably wrecked the otherwise clever illusion.
--
----------------------------- personal page: http://www.armory.com/~spectre/ --
Cameron Kaiser, Point Loma Nazarene University * ckaiser(a)stockholm.ptloma.edu
-- Not sun-worshippers: Son-worshippers! -- Uhura, Star Trek "Bread &
Circuses"
10:07 p.m.
see below, plz.
Dick
----- Original Message -----
From: "Christopher Smith" <csmith(a)amdocs.com
To: <classiccmp(a)classiccmp.org
Sent: Friday, April 26, 2002 9:04 AM
Subject: RE: "Toy" computers (was Re: Micro$oft Biz'droid Lusers)
Well, this is not much more than a preference of mine, apparently, but I
prefer my microcomputers to be small, i.e. not much bigger than a couple or
three loaves of sandwich bread next to each other on the table. I prefer to
have the interface to the I/O and memory, and whatever other hardware is
intended to be used with the computer to live within the main box (formerly
called mainframe, not meaning a computer requiring several acres of
floorspace) and that the interfaces be either permanent or removable, but
resident in that box. The peripherals can live outside. It's a convenient
way for microcomputers to work. However, the industry's getting away from
that. In a year or two, I expect PC's intended for typical home use will come
in what Sun and others often nicknamed a "pizza box" enclosure that has
external attachments for a monitor, keyboard, mouse, USB, network, SCSI or
fiberchannel, a printer, a phone line, a couple of serial ports, and nothing
else. The interfaces will, of course, still reside in the box, but you won't
have the liberty of picking and choosing your own as you do now. A technology
upgrade will often mean replacing the whole shootin' match, but with enough
volume that won't be so painful. Since every such computer will be EXACTLY
alike, Billy won't have a problem writing software to work pretty much the
same with all of 'em.
I wouldn't presume to make that choice for you. I'd point out, however,
that
if you use tape, you'd get better performance with several drives, just as
with floppy disk, one drive can be a mite tedious.
-----Original
Message-----
From: Richard Erlacher [mailto:edick@idcomm.com]
> That pretty much describes commodore disk
drives, yep. The
An external IDE drive would be hazardous thing to
use because
of cable length.
Didn't think of that -- you're probably right.
> handled all of the complexities of disk I/O,
so that the CPU didn't
> need to -- so there are good points to them.
What you may have had, then, is a toy with a
computer as a
peripheral. That
was probably quite a bit after the time reference of 1980.
Actually, I don't remember what CPU was in the 1541 drives, but I
seem to recall that it was actually more powerful than the computer.
I also remember somebody working on a way to get code into the drive
to be executed, but that's kind of fuzzy.
What I am focused on is where the intelligence to
run the I/O
resides. Once
So does the computer have to come with it, or simply allow space for
it on the inside of the case?
peripherals made great deal of sense. However,
it resulted
in ugly and
awkward packaging, which was addressed even later with more
elegant interface
standards, say, by 1986, with the standardization of SCSI.
Well, I certainly won't argue with that.
> Would you also have considered 9-track tape
"mass storage" for the
> time?
I'd tread lightly around that subject, unless
your Atari or
whatever, had a
9-track drive in '80 or so.
Obviously not, I'm just curious as to whether it would need to
be some form of "fast" disk-like storage, or if a high-capacity
tape would be ok.
8117
days inactive
8118
days old
2 comments
3 participants
participants (3)
-
csmith＠amdocs.com -
edick＠idcomm.com -
spectre＠stockholm.ptloma.edu