3 Nov
2008
3 Nov
'08
9:53 p.m.
Does anyone know of any computers based on the Fairchild 9440 / 9445
cpu?
Rob
ps. Google wasn't being my friend.
Rob Borsuk
email: rborsuk at colourfull.com
Colourfull Creations
Web: http://www.colourfull.com
3 Nov
3 Nov
10:18 p.m.
On 3 Nov 2008 at 21:53, Robert Borsuk wrote:
Does anyone know of any computers based on the
Fairchild 9440 / 9445
cpu?
That'd be the Fairchild LSI version of the DG Nova (more or less)
right? ISTR that, like the TI I2L version of the 9900, sales were
pretty much in the defense racket. I recall (long ago) trying to get
a Fairchild sales rep to talk to me about the 9440 and getting
nowhere.
Perhaps some of our military computer specialists may know of an
application.
Cheers,
Chuck
11:15 p.m.
Robert Borsuk wrote:
Does anyone know of any computers based on the Fairchild 9440 / 9445
cpu?
ps. Google wasn't being my friend.
Can't help with an answer, but thanks for the question: I had never heard of
these before. Pulled out the 1982 IC Master and found some description of them
and some support chips. The support chips are marked "Preliminary" or with
pending availability dates at that time.
Amazing, I'm wondering whether it's another instance of Fairchild missing the
boat in the microproc era, or, as Chuck is suggesting, they were high-end
enough that they were limited to the military market.
4 Nov
4 Nov
5:51 a.m.
On Nov 3, 2008, at 11:15 PM, Brent Hilpert wrote:
Amazing, I'm wondering whether it's another instance of Fairchild
missing the
boat in the microproc era, or, as Chuck is suggesting, they were
high-end
enough that they were limited to the military market
Don't know if you guys found this but I did find one computer reference.
http://www.old-computers.com/museum/computer.asp?c=196
and the data sheet on bit savers for the chips(thank you Al).
If your wondering, I was looking for information on a Strobe Data Hawk
board and happen to go the the simulogics web site. The have a
product called reNOVAte which happens to emulate the Strobe Data board
and the 9440 / 9445 Microprocessor. Having never heard of that micro,
the hunt was on.
Don't you love these little distractions?
Rob
Rob Borsuk
email: rborsuk at colourfull.com
Colourfull Creations
Web: http://www.colourfull.com
12:11 p.m.
Robert Borsuk wrote:
Don't know if you guys found this but I did find one computer reference.
http://www.old-computers.com/museum/computer.asp?c=196
and the data sheet on bit savers for the chips(thank you Al).
Looks at link ... and
what do you know -- ebay stuff listed too.
A old 8080A chip ... $25 sigh... A few 8 bit micros and
the 40K core memory stack is going for about $500 so far.
1:08 p.m.
Robert Borsuk wrote:
Don't know if you guys found this but I did find one computer reference.
http://www.old-computers.com/museum/computer.asp?c=196
and the data sheet on bit savers for the chips(thank you Al).
Nice chips ... fast too!
2:39 p.m.
Robert Borsuk wrote:
On Nov 3, 2008, at 11:15 PM, Brent Hilpert wrote:
Just to put some dates on it, looking more closely at the IC Master I have
here, the 9440 is stated to have been available in early 1978, while the 9445
followed in late 1980. So the 9440 was successful enough that Fairchild saw fit
to produce the 9445 follow-on.
In that datasheet from bitsavers there are schematics (pg29-32) for a full
switches-and-blinkenlights front panel for the 9445! Funny to see that for a
microproc in 1982, presumably some legacy from the minicomputer origins.
--
I haven't looked into it in depth but it has been a point of interest to me as
to how Fairchild fell from grace in the 1980s: failing to produce a strong
contender in the microproc arena?, betting the farm on the wrong technology
(high-performance bipolar/I3L instead of CMOS?, too focussed on the high-end
military market?, too much brain drain to other companies? ...
The counterpoint is how Motorola has managed to adapt and stay on the forefront
of technology since it's inception producing car radios in the 1920's.
Amazing, I'm wondering whether it's another instance of Fairchild
missing the
boat in the microproc era, or, as Chuck is suggesting, they were
high-end
enough that they were limited to the military market
Don't know if you guys found this but I did find one computer reference.
http://www.old-computers.com/museum/computer.asp?c=196
and the data sheet on bit savers for the chips(thank you Al).
If your wondering, I was looking for information on a Strobe Data Hawk
board and happen to go the the simulogics web site. The have a
product called reNOVAte which happens to emulate the Strobe Data board
and the 9440 / 9445 Microprocessor. Having never heard of that micro,
the hunt was on.
Don't you love these little distractions?
4:13 p.m.
On 4 Nov 2008 at 11:39, Brent Hilpert wrote:
I haven't looked into it in depth but it has been
a point of interest to me as
to how Fairchild fell from grace in the 1980s: failing to produce a strong
contender in the microproc arena?, betting the farm on the wrong technology
(high-performance bipolar/I3L instead of CMOS?, too focussed on the high-end
military market?, too much brain drain to other companies? ...
I don't think that was exclusively the case. Anyone remember the
Fairchild F8? Around the same time as the 9440, the F8 was hugely
popular in embedded type applications. The "Channel F" game console
and the Video Brain used them.
But Fairchild took the product nowhere after that. A very puzzling
company, Fairchild.
Cheers,
Chuck
6:13 p.m.
Chuck Guzis wrote:
On 4 Nov 2008 at 11:39, Brent Hilpert wrote:
No - I wasn't forgetting the F8. It was an interesting architecture in the way
the functional units were spread around the physical chips. Seemed very
tortured when I first saw it, until I understood how it could be useful for
economically efficient embedded systems. They were off to a good start in that
regard in the early 70s.
Interesting to speculate whether the architecture could have grown and been
carried into the future to contend in the marketplace occupied by PICs,etc.
today, but I'm not familiar enough with it in depth to say.
I haven't looked into it in depth but it has
been a point of interest to me as
to how Fairchild fell from grace in the 1980s: failing to produce a strong
contender in the microproc arena?, betting the farm on the wrong technology
(high-performance bipolar/I3L instead of CMOS?, too focussed on the high-end
military market?, too much brain drain to other companies? ...
I don't think that was exclusively the case. Anyone remember the
Fairchild F8? Around the same time as the 9440, the F8 was hugely
popular in embedded type applications. The "Channel F" game console
and the Video Brain used them.
But Fairchild took the product nowhere after that. A very puzzling
company, Fairchild.
6:56 p.m.
Brent Hilpert wrote:
Robert Borsuk wrote:
but since MAC no longers use a 68000 style cpu, you just have NO-Comment
cpu clones
for buggy computer designs.
On Nov 3, 2008, at 11:15 PM, Brent Hilpert
wrote:
Just to put some dates on it, looking more closely at the IC Master I have
here, the 9440 is stated to have been available in early 1978, while the 9445
followed in late 1980. So the 9440 was successful enough that Fairchild saw fit
to produce the 9445 follow-on.
In that datasheet from bitsavers there are schematics (pg29-32) for a full
switches-and-blinkenlights front panel for the 9445! Funny to see that for a
microproc in 1982, presumably some legacy from the minicomputer origins.
--
I haven't looked into it in depth but it has been a point of interest to me as
to how Fairchild fell from grace in the 1980s: failing to produce a strong
contender in the microproc arena?, betting the farm on the wrong technology
(high-performance bipolar/I3L instead of CMOS?, too focussed on the high-end
military market?, too much brain drain to other companies? ...
The counterpoint is how Motorola has managed to adapt and stay on the forefront
of technology since it's inception producing car radios in the 1920's.
Amazing, I'm wondering whether it's
another instance of Fairchild
missing the
boat in the microproc era, or, as Chuck is suggesting, they were
high-end
enough that they were limited to the military market
Don't know if you guys found this but I did find one computer
reference.
http://www.old-computers.com/museum/computer.asp?c=196
and the data sheet on bit savers for the chips(thank you Al).
If your wondering, I was looking for information on a Strobe Data Hawk
board and happen to go the the simulogics web site. The have a
product called reNOVAte which happens to emulate the Strobe Data board
and the 9440 / 9445 Microprocessor. Having never heard of that micro,
the hunt was on.
Don't you love these little distractions?
9:26 p.m.
From: bfranchuk at jetnet.ab.ca
Sent: Tuesday, November 04, 2008 3:56 PM
Brent Hilpert wrote:
> I haven't looked into it in depth but it has
been a point of interest to me
> as to how Fairchild fell from grace in the 1980s: failing to produce a
> strong contender in the microproc arena?, betting the farm on the wrong
> technology (high-performance bipolar/I3L instead of CMOS?, too focussed on
> the high-end military market?, too much brain drain to other companies? ...
> The counterpoint is how Motorola has managed to
adapt and stay on the
> forefront of technology since it's inception producing car radios in the
> 1920's.
but since MAC no longers use a 68000 style cpu, you
just have NO-Comment cpu
clones for buggy computer designs.
It's been longer since Macs used 68K processors than the amount of time that
they did (1984-1995 vs. 1995-2008).
Rich Alderson
Server Engineer, PDPplanet Project
Vulcan, Inc.
505 5th Avenue S, Suite 900
Seattle, WA 98104
mailto:RichA at vulcan.com
(206) 342-2239
(206) 465-2916 cell
11:05 p.m.
Rich Alderson wrote:
It's been longer since Macs used 68K processors
than the amount of time that
they did (1984-1995 vs. 1995-2008).
That is why I said 68000 style of cpu. I just am grumbling at the LAST
mac design. IT is too bad APPLE had pick the low end stuff for consumer
products. I don't know of any well designed computers, but then I never
got into UNIX workstations.
5 Nov
5 Nov
11:05 a.m.
bfranchuk at jetnet.ab.ca wrote:
That is why I said 68000 style of cpu.
Other
than the Freescale (Motorola) Coldfire series, and the Hitachi
HD63000 (which I'm not certain actually shipped), I can't think of any
CPUs that are "68000 style" but not actually members of the M68K family.
If you're talking about the PowerPC, that's certainly not "68000 style"
in any significant sense. There's little similarity beyond both being
big-endian.
11:24 a.m.
On 05/11/2008 16:05, Eric Smith wrote:
bfranchuk at jetnet.ab.ca wrote:
Aren't all proper processors big-endian? :-)
(Well, apart from the 6502, of course, and, er, the VAX, and, well an
ARM if configured the wrong way, and I suppose a PowerPC if configured
bass-ackwards...)
--
Pete Peter Turnbull
Network Manager
University of York
That is why I said 68000 style of cpu.
Other than the Freescale (Motorola) Coldfire series, and the Hitachi
HD63000 (which I'm not certain actually shipped), I can't think of any
CPUs that are "68000 style" but not actually members of the M68K family.
If you're talking about the PowerPC, that's certainly not "68000 style"
in any significant sense. There's little similarity beyond both being
big-endian. 
3:16 p.m.
On Wednesday 05 November 2008, Pete Turnbull wrote:
On 05/11/2008 16:05, Eric Smith wrote:
No.
;)
Pat
--
Purdue University Research Computing --- http://www.rcac.purdue.edu/
The Computer Refuge --- http://computer-refuge.org
bfranchuk at jetnet.ab.ca wrote:
Aren't all proper processors big-endian? :-) That is why I said 68000 style of cpu.
Other than the Freescale (Motorola) Coldfire series, and the
Hitachi HD63000 (which I'm not certain actually shipped), I can't
think of any CPUs that are "68000 style" but not actually members
of the M68K family.
If you're talking about the PowerPC, that's certainly not "68000
style" in any significant sense. There's little similarity beyond
both being big-endian. 
4:18 p.m.
Pete Turnbull wrote:
I'm pretty sure you could get a PERQ 2T4 to behave like that if you were
crazy enough about writing the microcode...
The standard PERQ has 20 bit processor registers and 16 bit memory (at
least as seen by the processor). You can read 'MDI' (Memory Data Input)
which transfers the memory word into the low 16 bits of the CPU data
path, or 'MDX' (Memory Data eXtended) which reads the bottom 4 bits of
the memory into the top 4 bits of the data path (that is, bits 19..16).
Now, the very ratre T4 has a 24 bit processor. I assume on that machine
MDX reads the low byte of the memory into the top 8 bits (23...16) of the
data path.
Now, whether anyone wants to try wring an instruction set that makes it
middle-endian is another matter...
-tony
Aren't all proper processors big-endian? :-)
Someone really needs to design a 24-bit middle-endian CPU and settle that
argument for good :-) 
4:34 p.m.
> Aren't all proper processors big-endian? :-)
Someone really needs to design a 24-bit middle-endian
CPU and settle that
argument for good :-)
People would still argue about the sequence for the other two "bytes".
The only real solution would be massive redundancy, and have every word be
symmetrical, which would require at least 32, 48, or 64 bits.
Waitaminit! How about having a 1 bit word?
7 Nov
7 Nov
3:11 a.m.
On Wed, 05 Nov 2008 16:24:36 +0000
Pete Turnbull <pete at dunnington.plus.com> wrote:
Aren't all proper processors big-endian? :-)
(Well, apart from the 6502, of course, and, er, the VAX,
Alpha!
and, well an
ARM if configured the wrong way, and I suppose a PowerPC if
configured bass-ackwards...)
Or MIPS, like DEC did it...
Is the Cray YMP architecture big endian?
NEC SX?
--
tsch??,
Jochen
Homepage: http://www.unixag-kl.fh-kl.de/~jkunz/
5741
days inactive
5744
days old
20 comments
12 participants
participants (12)
-
ard＠p850ug1.demon.co.uk -
bfranchuk＠jetnet.ab.ca -
cclist＠sydex.com -
cisin＠xenosoft.com -
eric＠brouhaha.com -
hilpert＠cs.ubc.ca -
jkunz＠unixag-kl.fh-kl.de -
jules.richardson99＠gmail.com -
pat＠computer-refuge.org -
pete＠dunnington.plus.com -
rborsuk＠colourfull.com -
RichA＠vulcan.com