5 Oct
1998
5 Oct
'98
9:19 a.m.
< Sam wrote:
< > Anyway, the F14 CADC SLF chip can be classified as a single-chip
< > microprocessor. I say "can" because you've get a different
definition
< > microprocessor depending on whom you ask.
<
< OK, let us know the definition, so that we can decide whether the LSI-1
< chip set also qualifies as a microprocessor. Then we can settle the tri
< question regarding which microprocessor-based Unix system came first.
Dec sold the LSI-11 and later chip based systems as microcomputers.
Source for this is the bindings of not less than 7 databooks and
technical descriptions. Also the T-11 member of the family is a single
40pin dip with one die inside.
If that werent' the case the fairchild F8 would not qualify as a single
chip CPU due to the need for multiple chips and the same would apply to
the predecessor to the RCA CDP1802.
allison
5 Oct
5 Oct
10:48 a.m.
A non-text attachment was scrubbed...
Name: not available
Type: text
Size: 338 bytes
Desc: not available
Url :
http://www.classiccmp.org/pipermail/cctalk/attachments/19981005/9dfb275b/at…
11:20 a.m.
Dec sold the LSI-11 and later chip based systems as
microcomputers.
They sold the *systems* as "microcomputers". That has little bearing on
whether the LSI-11 MOS VLSI chip set can be considered to be a
microprocessor.
Source for this is the bindings of not less than 7
databooks and
technical descriptions.
Yes, I've got the whole set as well.
Also the T-11 member of the family is a single
40pin dip with one die inside.
Still not relevant. The T-11 was definitely a microprocessor, but it came
much later than the LSI-11. AFAIK noone ran Unix on a T-11, but even
if they did, it certainly didn't predate running Unix on an LSI-11.
If that werent' the case the fairchild F8 would
not qualify as a single
chip CPU due to the need for multiple chips and the same would apply to
the predecessor to the RCA CDP1802.
Certainly the F8 qualifies as a single chip CPU. So did the 8080, 6800,
6502, Z-80, etc. Just because it took some additional chips to make a
system doesn't mean that it didn't have a single-chip CPU.
We haven't established a precise definition for microprocessor, but what
you seem to be talking about is a microcontroller, which is a complete
system on a chip (i.e., CPU, memory, and I/O).
You'll get very few people to agree that the memory and I/O must be present
for a chip to qualify as a microprocessor. This is evident from the word
itself: 'micro', meaning small, and 'processor', meaning something that
processes information. Nothing in either the 'micro' or 'processor'
portion
of the word seems to imply that any memory or I/O are included.
AFAIK, the first true microcontroller (CPU, memory, and I/O on one monolithic
IC) was the Intel 8048.
Eric
11:41 a.m.
On 5 Oct 1998, Eric Smith wrote:
Certainly the F8 qualifies as a single chip CPU.
I thought the F8 was two-chip (ALU and control logic on two different
chips). Mostek made a microprocessor version of the F8 (3840?).
We haven't established a precise definition for
microprocessor, but what
you seem to be talking about is a microcontroller, which is a complete
system on a chip (i.e., CPU, memory, and I/O).
My definition is ALU and control logic on a single chip.
-- Doug
12:27 p.m.
Doug Yowza <yowza(a)yowza.com> wrote:
I thought the F8 was two-chip (ALU and control logic
on two different
chips). Mostek made a microprocessor version of the F8 (3840?).
No, the entire F8 CPU was on a single chip. I'm not sure about a Fairchild
part number; Mostek second-sourced it with the part number MK3850.
The confusion may stem from an unusual feature of the F8 bus. Each F8
support chip maintained a local copy of the program counter and data pointer,
in order to reduce the necessary bus bandwidth. However, the CPU was
complete in that it contained the master copy of both pointers. Note that
the HP Saturn processor used in the HP-71B handheld computer in 1984,
and in all HP handheld calculators since 1986, uses the same principle.
Mostek also made single-chip-microcontroller variants in the MK3870 series.
My definition is ALU and control logic on a single
chip.
Seems basically reasonable, although "control logic" is a bit vague. Your
definition could apply to hardwired processors that do not execute stored
programs.
Eric
12:06 p.m.
A non-text attachment was scrubbed...
Name: not available
Type: text
Size: 181 bytes
Desc: not available
Url :
http://www.classiccmp.org/pipermail/cctalk/attachments/19981006/7efb09a9/at…
12:57 p.m.
On 5 Oct 1998, Eric Smith wrote:
Ahhhhhh! So we finally have some consensus. I wish people would realize
this about Holt's F14 microprocessor. The 5 other chips in the set (PMU,
PDU, SL, RAS, ROM) made up a whole SYSTEM, but the SLF could perform
usefully with just the RAS (RAM) and/or ROM.
If that
werent' the case the fairchild F8 would not qualify as a single
chip CPU due to the need for multiple chips and the same would apply to
the predecessor to the RCA CDP1802.
Certainly the F8 qualifies as a single chip CPU. So did the 8080, 6800,
6502, Z-80, etc. Just because it took some additional chips to make a
system doesn't mean that it didn't have a single-chip CPU. We haven't established a precise definition for
microprocessor, but what
And you never will.
you seem to be talking about is a microcontroller,
which is a complete
system on a chip (i.e., CPU, memory, and I/O).
This is what many people get confused about when trying to define a
"microprocessor" or more precisely, "single-chip microprocessor". A
"computer-on-a-chip" is quite a different beast.
You'll get very few people to agree that the
memory and I/O must be present
for a chip to qualify as a microprocessor. This is evident from the word
itself: 'micro', meaning small, and 'processor', meaning something that
processes information. Nothing in either the 'micro' or 'processor'
portion
of the word seems to imply that any memory or I/O are included.
I think to truly understand where the controversy lies in what was the
"first" "microprocessor", one has to get themselves in the mindset of
the
computer engineers back in the late 60s, early 70s. A "microprocessor",
or a CPU on a chip, was the next logical progression. There was no magic
in creating it really, it was simply a matter of time before the
technology advance to a state where it could be accomplished. Hoff,
Faggin and Mazor, by virtue of their employer having mass-marketed a
product that crammed an entire Central Processing Unit on a single chip,
take license to define what a "microprocessor" is or should be. The
definition is somewhat arbitrary. They just happened to garner the
prestige that enable them to this day to influence that definition, which
isn't necessarily proper, but more self-serving.
AFAIK, the first true microcontroller (CPU, memory,
and I/O on one monolithic
IC) was the Intel 8048.
Are you sure?
Sam Alternate e-mail: dastar(a)siconic.com
-------------------------------------------------------------------------------
Ever onward.
Coming in 1999: Vintage Computer Festival 3.0
See http://www.vintage.org/vcf for details!
[Last web site update: 09/21/98]
9576
days inactive
9576
days old
6 comments
5 participants
participants (5)
-
allisonp＠world.std.com -
ard＠p850ug1.demon.co.uk -
dastar＠ncal.verio.com -
eric＠brouhaha.com -
yowza＠yowza.com