14 Jun
2020
14 Jun
'20
9:52 a.m.
Hi all.
I recently bought a mystery blinkenlight panel. Closer inspection reveals it was
manufactured by Intel in the early 70?s (1973), and some people on the book of faces
suggested it was part of a ?device multiplexer?(?)
I?m 95% confident it?s not strictly a ?computer? blinkenlight panel, but rather an
attached device, but that still hasn?t helped me narrow down what exactly it was from.
I?ve not seen any early Intel stuff as rack-mount, so i?m wondering if it was a prototype,
or maybe a piece of internal/non-commercial hardware for Intel's own use.
I?m hoping someone here might be able to shed some light on this mystery.
Pictures: https://imgur.com/gallery/lD74oSy <https://imgur.com/gallery/lD74oSy>
Thanks in advance.
Josh Rice
14 Jun
14 Jun
5:01 p.m.
On 06/14/2020 03:52 AM, Joshua Rice via cctalk wrote:
Hi all.
I recently bought a mystery blinkenlight panel. Closer inspection reveals it was
manufactured by Intel in the early 70?s (1973), and some people on the book of faces
suggested it was part of a ?device multiplexer?(?)
I?m 95% confident it?s not strictly a ?computer? blinkenlight panel, but rather an
attached device, but that still hasn?t helped me narrow down what exactly it was from.
I?ve not seen any early Intel stuff as rack-mount, so i?m wondering if it was a
prototype, or maybe a piece of internal/non-commercial hardware for Intel's own use.
I?m hoping someone here might be able to shed some light on this mystery.
Pictures: https://imgur.com/gallery/lD74oSy <https://imgur.com/gallery/lD74oSy>
Any reason the picture is upside down? Also, blurry, so it
is hard to read the labels. it seems to have two 9-bit
bytes and a memory address.
Jon
8:31 p.m.
Hi Al
Can you repeat what the text is? I can't make out what it says, even with
magnification.
Dwight
________________________________
From: cctalk <cctalk-bounces at classiccmp.org> on behalf of Al Kossow via cctalk
<cctalk at classiccmp.org>
Sent: Sunday, June 14, 2020 9:21 AM
To: cctalk at classiccmp.org <cctalk at classiccmp.org>
Subject: Re: Unknown Intel blinkenlight panel circa 1973
On 6/14/20 9:01 AM, Jon Elson via cctalk wrote:
Any reason the picture is upside down? Also, blurry,
so it is hard to read the labels.
I can't find the original eBay listing right now.
The pictures were better on there
9:05 p.m.
https://i.ebayimg.com/images/g/dysAAOSwqqJehZ-Y/s-l1600.jpg
<https://i.ebayimg.com/images/g/dysAAOSwqqJehZ-Y/s-l1600.jpg>
That?s off the listing.
Sorry the photos came out funny/upside down, i guess my phone couldn?t work out which way
up it was?
I think the photos of the board numbers are clear enough in the IMGur link though. I can
write them down if you wish. There?s no other markings on the PCB though, apart from the
3M printing on the raw PCB substrate
On Jun 14, 2020, at 8:31 PM, dwight via cctalk
<cctalk at classiccmp.org> wrote:
Hi Al
Can you repeat what the text is? I can't make out what it says, even with
magnification.
Dwight
________________________________
From: cctalk <cctalk-bounces at classiccmp.org> on behalf of Al Kossow via cctalk
<cctalk at classiccmp.org>
Sent: Sunday, June 14, 2020 9:21 AM
To: cctalk at classiccmp.org <cctalk at classiccmp.org>
Subject: Re: Unknown Intel blinkenlight panel circa 1973
On 6/14/20 9:01 AM, Jon Elson via cctalk wrote:
Any reason the picture is upside down? Also,
blurry, so it is hard to read the labels.
I can't find the original eBay listing right now.
The pictures were better on there
15 Jun
15 Jun
1:45 a.m.
On 06/14/2020 03:05 PM, Joshua Rice via cctalk wrote:
https://i.ebayimg.com/images/g/dysAAOSwqqJehZ-Y/s-l1600.jpg
<https://i.ebayimg.com/images/g/dysAAOSwqqJehZ-Y/s-l1600.jpg>
That?s off the listing.
Clearly a panel for a memory device. Data is 16 bits plus 2
parity bits. There are separate lights for the data read
bus and data write bus. The address is a bit funny :
W0 W1 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
Jon
4:41 a.m.
I can see why I was having problems. The picture was upside down. It looked like Spanish
or something. Do remember that Intel's claim to fame wasn't just micro processor.
They were one of the first to do MOS RAMs for big machines. They were more into solid
state memory systems than uPs, until after the 8080.
It clearly isn't for some 4004 or 8008. It was likely monitoring some RAM for some
mini.
Dwight
________________________________
From: cctalk <cctalk-bounces at classiccmp.org> on behalf of Jon Elson via cctalk
<cctalk at classiccmp.org>
Sent: Sunday, June 14, 2020 5:45 PM
To: Joshua Rice <Rice43 at btinternet.com>; General at ezwind.net <General at
ezwind.net>; Discussion@ <ezwind.net:On-Topic and Off-Topic Posts cctalk at
classiccmp.org>
Subject: Re: Unknown Intel blinkenlight panel circa 1973
On 06/14/2020 03:05 PM, Joshua Rice via cctalk wrote:
https://i.ebayimg.com/images/g/dysAAOSwqqJehZ-Y/s-l1600.jpg
<https://i.ebayimg.com/images/g/dysAAOSwqqJehZ-Y/s-l1600.jpg>
That?s off the listing.
Clearly a panel for a memory device. Data is 16 bits plus 2
parity bits. There are separate lights for the data read
bus and data write bus. The address is a bit funny :
W0 W1 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
Jon
6:28 a.m.
there ya? go!(as? I? said...)?ED#?In a message dated 6/14/2020 8:41:43 PM US Mountain
Standard Time, cctalk at classiccmp.org writes:?
It clearly isn't for some 4004 or 8008. It was likely monitoring some RAM for some
mini.
Dwight
6:33 a.m.
On 6/14/20 8:41 PM, dwight via cctalk wrote:
I can see why I was having problems. The picture was
upside down. It
looked like Spanish or something. Do remember that Intel's claim to
fame wasn't just micro processor. They were one of the first to do
MOS RAMs for big machines. They were more into solid state memory
systems than uPs, until after the 8080. It clearly isn't for some
4004 or 8008. It was likely monitoring some RAM for some mini.
Intel Memory Systems Division was largely responsible for saving Intel's
bacon in the early 1970s. The MPU business with the 8008 and 4004
wasn't a moneymaker initially.
However, selling DRAM assemblies for various minis (e.g. DG and DEC) as
well as S/370 add-on memory amounted to a large portion of their early
sales. IMSD was on the far end of Mathilda in Sunnyvale.
Have a look at section 7 here:
https://johncargin.files.wordpress.com/2013/12/intel-catalog-1973.pdf
--Chuck
10:09 a.m.
-----Original Message-----
From: cctalk <cctalk-bounces at classiccmp.org> On Behalf Of Chuck Guzis via
cctalk
Sent: 15 June 2020 06:33
To: dwight via cctalk <cctalk at classiccmp.org>
Subject: Re: Unknown Intel blinkenlight panel circa 1973
On 6/14/20 8:41 PM, dwight via cctalk wrote:
a
I can see why I was having problems. The picture
was upside down. It
looked like Spanish or something. Do remember that Intel's claim to
fame wasn't just micro processor. They were one of the first to do MOS
RAMs for big machines. They were more into solid state memory systems
than uPs, until after the 8080. It clearly isn't for some
4004 or 8008. It was likely monitoring some RAM for some mini.
Intel Memory Systems Division was largely responsible for saving Intel's
bacon in the early 1970s. The MPU business with the 8008 and 4004 wasn't moneymaker initially.
However, selling DRAM assemblies for various minis (e.g. DG and DEC) as
well as S/370 add-on memory amounted to a large portion of their early
sales. IMSD was on the far end of Mathilda in Sunnyvale.
I found this ad here:-
https://books.google.co.uk/books?id=9Ih5WQ-muXEC&pg=PT1&lpg=PT1&…
hird+party+memory+intel
I think if we could read the legends on the front we might get a better idea
of what system it was for...
... looks like 18-bits so something in the pdp-10 line?
Dave
Have a look at section 7 here:
https://johncargin.files.wordpress.com/2013/12/intel-catalog-1973.pdf
--Chuck
10:58 a.m.
PDP-10 was 36-bit. You?re thinking of the PDP-1, 4, 7, 9 and 15 line of computers.
However, it?s not actually 18-bit, but 2 8-bit bytes with parity for each byte. If it is
indeed an early Intel memory addon for minicomputers, i?d expect it to be for the 16-bit
DG Nova or PDP-11 line of machines.
On Jun 15, 2020, at 10:09 AM, Dave Wade via cctalk
<cctalk at classiccmp.org> wrote:
-----Original Message-----
From: cctalk <cctalk-bounces at classiccmp.org <mailto:cctalk-bounces at
classiccmp.org>> On Behalf Of Chuck Guzis via
cctalk
Sent: 15 June 2020 06:33
To: dwight via cctalk <cctalk at classiccmp.org <mailto:cctalk at
classiccmp.org>>
Subject: Re: Unknown Intel blinkenlight panel circa 1973
On 6/14/20 8:41 PM, dwight via cctalk wrote:
a
I can see why I was having problems. The picture
was upside down. It
looked like Spanish or something. Do remember that Intel's claim to
fame wasn't just micro processor. They were one of the first to do MOS
RAMs for big machines. They were more into solid state memory systems
than uPs, until after the 8080. It clearly isn't for some
4004 or 8008. It was likely monitoring some RAM for some mini.
Intel Memory Systems Division was largely responsible for saving Intel's
bacon in the early 1970s. The MPU business with the 8008 and 4004 wasn't moneymaker initially.
However, selling DRAM assemblies for various minis (e.g. DG and DEC) as
well as S/370 add-on memory amounted to a large portion of their early
sales. IMSD was on the far end of Mathilda in Sunnyvale.
I found this ad here:-
https://books.google.co.uk/books?id=9Ih5WQ-muXEC&pg=PT1&lpg=PT1&…
<https://books.google.co.uk/books?id=9Ih5WQ-muXEC&pg=PT1&lpg=PT1&dq=ibm/370+t>
hird+party+memory+intel
I think if we could read the legends on the front we might get a better idea
of what system it was for...
... looks like 18-bits so something in the pdp-10 line?
Dave
> Have a look at section 7 here:
>
> https://johncargin.files.wordpress.com/2013/12/intel-catalog-1973.pdf
<https://johncargin.files.wordpress.com/2013/12/intel-catalog-1973.pdf>
>
> --Chuck 
16 Jun
16 Jun
10:29 p.m.
On Mon, Jun 15, 2020 at 1:33 AM Chuck Guzis via cctalk
<cctalk at classiccmp.org> wrote:
On 6/14/20 8:41 PM, dwight via cctalk wrote:
I happen to have the guts for a 2MB IN-1670 memory system for the
PDP-11/70 - backplane and cards, but no box and no PSU.
https://books.google.com/books?id=fB-Te8d5hO8C&pg=PT1&lpg=PT1&d…
It's a stack of 16 memory cards covered in 4116 DRAMs and a couple of
control/management boards that cable over to the 11/70 the same way a
DEC MK11 box does.
... Do remember that Intel's claim to
fame wasn't just micro processor. They were one of the first to do
MOS RAMs for big machines. They were more into solid state memory
systems than uPs, until after the 8080.
Intel Memory Systems Division was largely responsible for saving Intel's
bacon in the early 1970s...
However, selling DRAM assemblies for various minis (e.g. DG and DEC) as
well as S/370 add-on memory amounted to a large portion of their early
sales. Have a look at section 7 here:
https://johncargin.files.wordpress.com/2013/12/intel-catalog-1973.pdf
Excellent reference. A couple years too old for my DEC memor system,
but I happen to have a tray of MSC-4 parts and some early SRAMs,
DRAMs, and EPROMs documented there.
-ethan
17 Jun
17 Jun
1:13 a.m.
On 6/16/20 2:29 PM, Ethan Dicks wrote:
It's a stack of 16 memory cards covered in 4116
DRAMs and a couple of
control/management boards that cable over to the 11/70 the same way a
DEC MK11 box does.
Oh yeah--I remember when 4116/2116 DRAMs were still new and in short
supply. The local Intel sales rep dropped off a number of 2108s
("half-good" 2116s) as a stopgap. There was a suffix on each one that
specified whether the bank (there were two "banks" in the 2116), should
be high or low.
I still have a few of those. Interestingly enough, some weren't bad
enough that prevented me from using them as 16Kb parts.
In any case, they were a BIG improvement over the 22-pin 0.400" 2107
commodity parts.
--Chuck
1:17 a.m.
I remember that shortage of memory chips. People in my office were all
blaming it the Ayatollah Khomeni!
On 16/06/2020 20:13, Chuck Guzis via cctalk wrote:
On 6/16/20 2:29 PM, Ethan Dicks wrote:
--
Nigel Johnson
MSc., MIEEE, MCSE
VE3ID/G4AJQ/VA3MCU
Amateur Radio, the origin of the open-source concept!
You can reach me by voice on Skype: TILBURY2591
If time travel ever will be possible, it already is. Ask me again yesterday
This e-mail is not and cannot, by its nature, be confidential. En route from me to you, it
will pass across the public Internet, easily readable by any number of system
administrators along the way.
Nigel Johnson <nw.johnson at ieee.org>
Please consider the environment when deciding if you really need to print this message
It's a stack of 16 memory cards covered in
4116 DRAMs and a couple of
control/management boards that cable over to the 11/70 the same way a
DEC MK11 box does.
Oh yeah--I remember when 4116/2116 DRAMs were still new and in
short
supply. The local Intel sales rep dropped off a number of 2108s
("half-good" 2116s) as a stopgap. There was a suffix on each one that
specified whether the bank (there were two "banks" in the 2116), should
be high or low.
I still have a few of those. Interestingly enough, some weren't bad
enough that prevented me from using them as 16Kb parts.
In any case, they were a BIG improvement over the 22-pin 0.400" 2107
commodity parts.
--Chuck
2:22 a.m.
On 6/16/20 5:17 PM, Nigel Johnson via cctalk wrote:
I remember that shortage of memory chips. People in my
office were all
blaming it the Ayatollah Khomeni!
The running battle between "dumping" and supply. By 1980, Japan owned
about half the world's DRAM market and US domestic producers were
leaving the business.
It was an interesting time in Silicon Valley; companies lost DRAM
inventory due to pilferage and outright burglary. One outfit who'd
carefully laid in a big stock of AT&T-manufactured DRAMs found their
parts crib empty of the things. Some started storing them in vaults and
there was at least one incident of a supply truck hijacking.
One of the other problems was that Japan was scaling up its production;
there wasn't enough domestic demand to justify expansion, so dumping was
an obvious answer. And the Japanese DRAMs were *good*. I still have
some NEC 416 DRAM here that we tested with a refresh rate of 2
seconds--that's seconds, not milliseconds.
After all, who's going to argue with someone selling Porsche 911s right
of the boat for $1500 each?
But we'd seen what happened to the US TV manufacturing market in the
70s--it simply couldn't compete with offshore production. Gone were the
Zeniths, the Packard-Bells and the Curtis-Mathes in a few years.
Things really came to a head when Korea entered the business. At that
time (1985 or so), I think the only US manufacturers of DRAM were Intel
and Micron. I think the US went a bit overboard on the anti-dumping
duties. Hyundai set up a big plant for DRAM locally in the 1990s to get
around anti-dumping penalties. The huge complex (1.2M square feet) has
sat empty since 2008. It's been passed around by various companies
(e.g. Broadcom) as a white elephant. The current owner paid $6.3
million for it in an auction. Heaven knows what will become of it; no
one seems to know.
--Chuck
3:15 p.m.
It was also clear at the time for Intel that the Japanese 4116s had 10 times better ppm
failure rates then the US made parts and they were cheaper. If it wasn't for EPROMs,
Intel might have folded then and 8080s were just taking off. I suspect if Motorola had
gotten their act together, it might be a different place today.
Dwight
________________________________
From: cctalk <cctalk-bounces at classiccmp.org> on behalf of Chuck Guzis via cctalk
<cctalk at classiccmp.org>
Sent: Tuesday, June 16, 2020 6:22 PM
To: Nigel Johnson via cctalk <cctalk at classiccmp.org>
Subject: Re: Unknown Intel blinkenlight panel circa 1973
On 6/16/20 5:17 PM, Nigel Johnson via cctalk wrote:
I remember that shortage of memory chips. People in my
office were all
blaming it the Ayatollah Khomeni!
The running battle between "dumping" and supply. By 1980, Japan owned
about half the world's DRAM market and US domestic producers were
leaving the business.
It was an interesting time in Silicon Valley; companies lost DRAM
inventory due to pilferage and outright burglary. One outfit who'd
carefully laid in a big stock of AT&T-manufactured DRAMs found their
parts crib empty of the things. Some started storing them in vaults and
there was at least one incident of a supply truck hijacking.
One of the other problems was that Japan was scaling up its production;
there wasn't enough domestic demand to justify expansion, so dumping was
an obvious answer. And the Japanese DRAMs were *good*. I still have
some NEC 416 DRAM here that we tested with a refresh rate of 2
seconds--that's seconds, not milliseconds.
After all, who's going to argue with someone selling Porsche 911s right
of the boat for $1500 each?
But we'd seen what happened to the US TV manufacturing market in the
70s--it simply couldn't compete with offshore production. Gone were the
Zeniths, the Packard-Bells and the Curtis-Mathes in a few years.
Things really came to a head when Korea entered the business. At that
time (1985 or so), I think the only US manufacturers of DRAM were Intel
and Micron. I think the US went a bit overboard on the anti-dumping
duties. Hyundai set up a big plant for DRAM locally in the 1990s to get
around anti-dumping penalties. The huge complex (1.2M square feet) has
sat empty since 2008. It's been passed around by various companies
(e.g. Broadcom) as a white elephant. The current owner paid $6.3
million for it in an auction. Heaven knows what will become of it; no
one seems to know.
--Chuck
16 Jun
16 Jun
3:41 a.m.
Josh R. wrote:
I recently bought a mystery blinkenlight panel. Closer
inspection reveals it was manufactured by Intel in the early 70?s (1973), and some people
on the book of faces suggested it was part of a ?device
multiplexer?(?)
I?m hoping someone here might be able to shed some
light on this mystery.
I suspect from looking it over, and given that Intel's early business was primarily
random access static and dynamic memory ICs, that this was the console of some kind of
memory test system.
I don't know for sure, though, it's just a guess.
I took the upside down image, flipped it, and did some image enhancement on it to make out
the legends on the indicators and switches. Based on what I see, my suspicion seems
plausible.
Rick Bensene
The Old Calculator Museum
http://oldcalculatormuseum.com
1433
days inactive
1436
days old
16 comments
10 participants
participants (10)
-
aek＠bitsavers.org -
cclist＠sydex.com -
couryhouse＠aol.com -
dave.g4ugm＠gmail.com -
dkelvey＠hotmail.com -
elson＠pico-systems.com -
ethan.dicks＠gmail.com -
nw.johnson＠ieee.org -
Rice43＠btinternet.com -
rickb＠bensene.com