21 Apr
1999
21 Apr
'99
5:43 a.m.
I have several jars of ferrite cores made by IBM sometime in the late
1950's. I don't know exactly what size they are, but they can be
hand-wired without a stereo microscope.
I know, because I built a plane of 1x64x64 when I was in high school in
1971. (I was very determined to build my own computer.) The project was
abandoned when I went away for college, which was probably a good thing,
because my analog design skills were not up to creating the drive and sense
circuitry.
As I recall, I used something like #60 enameled wire. One of the
challenges was stripping the insulation off the ends of the wires without
damaging the copper.
If anyone is seriously intending to build a demo core memory system, I'd be
glad to provide a couple of hundred cores. I'd recommend something like a
16x16 array, unless you're looking for a project to make a 20-to-life
sentence go faster.
----
John Dykstra jdykstra(a)nortelnetworks.com
Principal Software Architect voice: ESN 454-1604
Enterprise Solutions fax: ESN 667-8549
21 Apr
21 Apr
7:34 a.m.
On Wed, 21 Apr 1999, John Dykstra wrote:
challenges was stripping the insulation off the ends
of the wires without
damaging the copper.
stripx solvent or heat.
If anyone is seriously intending to build a demo core
memory system, I'd be
glad to provide a couple of hundred cores. I'd recommend something like a
16x16 array, unless you're looking for a project to make a 20-to-life
sentence go faster.
I'd love to take ashot at it say enough for an 8x8x1 plus a few spares.
Even a few I can pulse them to see how the need to be driven.
If I had the time I'd do a 8x8x8 plane, 64 bytes is enough to prove a
point!
Allison
----
John Dykstra jdykstra(a)nortelnetworks.com
Principal Software Architect voice: ESN 454-1604
Enterprise Solutions fax: ESN 667-8549
9:14 a.m.
If I had the
time I'd do a 8x8x8 plane, 64 bytes is enough to prove a
point!
Are you going to put it on a PCI board so it can be used in a modern
computer? :-) >>>
Until this recent discussion I never
really understood how core "worked",
so its been educational for me. (equal chance though I knew it well once,
etc.)
9:54 a.m.
If I had the time I'd do a 8x8x8 plane, 64 bytes
is enough to prove a
point!
Are you going to put it on a PCI board so it can be used in a modern
computer? :-) 
11:45 a.m.
A non-text attachment was scrubbed...
Name: not available
Type: text
Size: 442 bytes
Desc: not available
Url :
http://www.classiccmp.org/pipermail/cctalk/attachments/19990421/e4aa9a35/at…
10:51 a.m.
What would be HOT is to put it on a PC card (you pick
the flavor), but make
sure the size is sufficent for some REAL function like non volatile ram. I
smell a neat science project / demonstation board in this idea. Perhaps a
smaller array, but with LARGE cores where the magnetic state could have
some visual indicators.
Mike, PCI is 33mhz and core at best is 1mhz tthat would eb an expensive
slow joke.
Cores large enough for visual indication of magnetic field would be in the
10-20AMPS select curent range. Not a real possibility.
allison
11:05 a.m.
At 03:51 PM 4/21/99 -0400, allisonp(a)world.std.com wrote:
Cores large enough for visual indication of magnetic field would be in the
10-20AMPS select curent range. Not a real possibility.
How long does core stay magnetized without power?
How many bytes of battery-backed RAM are there in a typical PC?
Haven't there been contemporary memory devices that use nano-scale
cores, to eliminate the battery needs and radiation sensitivity of RAM?
- John
11:22 a.m.
On Wed, 21 Apr 1999, John Foust wrote:
At 03:51 PM 4/21/99 -0400, allisonp(a)world.std.com
wrote:
Forever, as far as I've been told.
Cores large enough for visual indication of magnetic field would be in the
10-20AMPS select curent range. Not a real possibility.
How long does core stay magnetized without power? How many bytes of battery-backed RAM are there in a
typical PC?
128?
Sellam Alternate e-mail: dastar(a)siconic.com
------------------------------------------------------------------------------
Don't rub the lamp if you don't want the genie to come out.
Coming this October 2-3: Vintage Computer Festival 3.0!
See http://www.vintage.org/vcf for details!
[Last web site update: 04/03/99]
1:26 p.m.
> >Cores large enough for visual indication of
magnetic field would be in the
> >10-20AMPS select curent range. Not a real possibility.
> How long does core stay magnetized without power?
Forever, as far as I've been told.
it is said that Fe is the most lasting element in the table,
but I'm not shure if the cores can maintain their orientation
within a black hole ...
> How many bytes of battery-backed RAM are there in
a typical PC?
128?
Hmm on several BIOS implementations there are additional RAM areas,
and dont forget the flash parts for setup information.
Gruss
H.
--
Stimm gegen SPAM: http://www.politik-digital.de/spam/de/
Vote against SPAM: http://www.politik-digital.de/spam/en/
Votez contre le SPAM: http://www.politik-digital.de/spam/fr/
Ich denke, also bin ich, also gut
HRK
11:48 a.m.
A non-text attachment was scrubbed...
Name: not available
Type: text
Size: 782 bytes
Desc: not available
Url :
http://www.classiccmp.org/pipermail/cctalk/attachments/19990421/18dee8b0/at…
1:15 p.m.
>Cores large enough for visual indication of
magnetic field would be in the
>10-20AMPS select curent range. Not a real possibility.
How long does core stay magnetized without power?
How many bytes of battery-backed RAM are there in a
typical PC?
Is my Linux system typical (using 1 Meg static as boot :) ?
Haven't there been contemporary memory devices
that use nano-scale
cores, to eliminate the battery needs and radiation sensitivity of RAM?
MRAM, based on the TMR effect - still in development, Universtity
of Kaiserslautern and SIEMENS Semiconductors (now renamed to Infineon)
cooperate in the development - Prototypes in the MBit scale are on
scedule for 2000/2001
Gruss
H.
--
Stimm gegen SPAM: http://www.politik-digital.de/spam/de/
Vote against SPAM: http://www.politik-digital.de/spam/en/
Votez contre le SPAM: http://www.politik-digital.de/spam/fr/
Ich denke, also bin ich, also gut
HRK
22 Apr
22 Apr
3:49 a.m.
Cores large
enough for visual indication of magnetic field would be in the
10-20AMPS select curent range. Not a real possibility.
Maybe so, but those are practicality issues, not feasibility. 
4:10 a.m.
At 08:49 AM 4/22/99 -0400, Allison wrote:
that answer equates to "whatever". Describe the mechanics of displaying
the orientation of a magnetic field in a closed torus? I was trying to
get someone to think on how thjis may be done impractical or not.
Yea, the field lines are contained inside the toroid, no external poles.
For an *exotic* way, the
Aharonov-Bohm effect might work. See the Scientific American article, April
1989. This, by the way, is that the magnetic potential is outside the torus
and can effect an electron in the quantum realm. Of course the effort and
equipment to do this for 1 bit would overwhelm the rest of the machine!
More on topic, the magnetic cores they used were 5 microns in diameter and
ferromagnetic, not "ferroelectric". They were 83% Ni and 17% iron, on a
silicon wafer, also shielded by niobium.
-Dave
Cores large enough for visual indication of magnetic
field would be in the
10-20AMPS select curent range. Not a real possibility.
Maybe so, but those are practicality issues, not feasibility.
8:01 a.m.
read to end...
On Thu, 22 Apr 1999, Dave Dameron wrote:
At 08:49 AM 4/22/99 -0400, Allison wrote:
that answer equates to "whatever". Describe the mechanics of displaying
the orientation of a magnetic field in a closed torus? I was trying to
get someone to think on how thjis may be done impractical or not.
Yea, the field lines are contained inside the toroid, no external poles.
For an *exotic* way, the
Aharonov-Bohm effect might work. See the Scientific American article, April
1989. This, by the way, is that the magnetic potential is outside the torus
and can effect an electron in the quantum realm. Of course the effort and
equipment to do this for 1 bit would overwhelm the rest of the machine!
More on topic, the magnetic cores they used were 5 microns in diameter and
ferromagnetic, not "ferroelectric". They were 83% Ni and 17% iron, on a
silicon wafer, also shielded by niobium.
-Dave
then there is the easy way... apply a magnetic field and see if it
switches.
Allison
Cores large enough for visual indication of magnetic
field would be in the
10-20AMPS select curent range. Not a real possibility.
Maybe so, but those are practicality issues, not feasibility.
8:28 a.m.
Cores large enough for visual indication of magnetic
field would be in the
10-20AMPS select curent range. Not a real possibility.
Maybe so, but those are practicality issues, not feasibility.
6:20 a.m.
At 10:28 AM 4/22/99 -0700, Mike wrote:
Are these arrows representing the current that cause the magnetic field
inside the toroid? The field remains at some level when the current is
removed, thus the storage.
-Dave
that answer
equates to "whatever". Describe the mechanics of displaying
the orientation of a magnetic field in a closed torus? I was trying to
get someone to think on how thjis may be done impractical or not.
Gee don't they have little arrows sticking through the middle like my
physics books show?
9:32 a.m.
New subject: Displaying core states
Somewhere in my boxes of stuff I've got one of the tape "visualizer" cards.
Its looks like a translucent grey plastic card and when you hold it over a
magnetized surface it shows the magnetic "spots" by turning white (what it
actually does is rotate a flat magnet in the card that is white on one side
and black on the other.) I got this from Edmund Scientific in the mid 70's
I believe. Anyway, some variation on this would work on cores.
--Chuck
At 08:49 AM 4/22/99 -0400, allisonp(a)world.std.com wrote:
that answer equates to "whatever". Describe the mechanics of displaying
the orientation of a magnetic field in a closed torus? I was trying to
get someone to think on how thjis may be done impractical or not.
I'm not talking about putting leds on the logig arond the cores to show
states.
Allison
Cores large enough for visual indication of magnetic
field would be in the
10-20AMPS select curent range. Not a real possibility.
Maybe so, but those are practicality issues, not feasibility.
9:35 a.m.
New subject: Displaying core states
At 11:32 AM 4/22/99 -0700, Chuck McManis wrote:
Somewhere in my boxes of stuff I've got one of the
tape "visualizer" cards.
Its looks like a translucent grey plastic card and when you hold it over a
magnetized surface it shows the magnetic "spots" by turning white (what it
There's also the $9-$20 "Magnadoodle" technology available in the toy
section of almost any store that flips from white to black after
having been exposed to a magnetic pole. I thought you had kids. :-)
- John
10:08 a.m.
New subject: Displaying core states
At 01:35 PM 4/22/99 -0500, John Foust wrote:
There's also the $9-$20 "Magnadoodle"
technology ...
I thought you had kids. :-)
Doh! That's a good thought. (And yes I have kids and yes we've gone through
our share of these :-))
Now since Allison has reminded me of what I tried so hard not to learn in
EE370 (ElectroM{agnetics/echanics} about the state of flux in a torroid it
seems that we will probably have to settle for capturing the "flip" as
opposed to reading the state directly. This would require clearing the
indicator and the core to a known state, then the capturing the
transitions. These transisitions, if I understand this correctly, would
exist as the momentary existence of a magnetic field around the torroid.
I don't know if anyone on this list has seen The Secret Life of Machines
(yeah right) but this is the kind of project Tim Hunkin really gets a kick
out of. And I've got to try to do this with nuts from a hardware store
since is the right way to do it. (I'm still in awe at how they recorded
sound on a scotch tape with rust sprinkled on it!)
--Chuck
10:59 a.m.
New subject: Displaying core states
On Thu, 22 Apr 1999, Chuck McManis wrote:
> There's also the $9-$20
"Magnadoodle" technology ...
> I thought you had kids. :-)
How do those work, anyway?
--Max Eskin (max82(a)surfree.com)
http://scivault.hypermart.net: Ignorance is Impotence - Knowledge is Power
4:53 p.m.
New subject: Displaying core states
On Thu, 22 Apr 1999, Chuck McManis wrote:
I don't know if anyone on this list has seen The
Secret Life of Machines
When does this show even come on anymore, and on what channel? It used to
appear on the Discovery channel but now all they show is stupid animal and
military weapons of mass destruction programs, which both bore me mostly
(although those monkey's can be pretty darn cute sometimes).
(yeah right) but this is the kind of project Tim
Hunkin really gets a kick
out of. And I've got to try to do this with nuts from a hardware store
since is the right way to do it. (I'm still in awe at how they recorded
sound on a scotch tape with rust sprinkled on it!)
I'm in awe of that wire recorded I got a couple months back! That thing
rocks!
Sellam Alternate e-mail: dastar(a)siconic.com
------------------------------------------------------------------------------
Don't rub the lamp if you don't want the genie to come out.
Coming this October 2-3: Vintage Computer Festival 3.0!
See http://www.vintage.org/vcf for details!
[Last web site update: 04/03/99]
23 Apr
23 Apr
9:16 a.m.
New subject: Displaying core states
On Thu, 22 Apr 1999, Sellam Ismail wrote:
> (yeah right) but this is the kind of project Tim
Hunkin really gets a kick
> out of. And I've got to try to do this with nuts from a hardware store
> since is the right way to do it. (I'm still in awe at how they recorded
> sound on a scotch tape with rust sprinkled on it!)
Really? There's a show like that? I thought they only had that stuff in
the USSR (in the USSR, there was a television show about how to build a
robot out of tin cans and magazines had plans for radio-controlled boats)!
Where can I find such wonderful sources of information?
--Max Eskin (max82(a)surfree.com)
http://scivault.hypermart.net: Ignorance is Impotence - Knowledge is Power
22 Apr
22 Apr
2:13 p.m.
New subject: Displaying core states
At 11:32 AM 4/22/99 -0700, Chuck McManis wrote:
I have a spray can of stuff that is used to make analytical measurements on
magnetic media. Spray it on, and you can see the tracks and transistions
with a microscope.
Somewhere in my boxes of stuff I've got one of
the tape "visualizer" cards.
Its looks like a translucent grey plastic card and when you hold it over a
magnetized surface it shows the magnetic "spots" by turning white (what it
There's also the $9-$20 "Magnadoodle" technology available in the toy
section of almost any store that flips from white to black after
having been exposed to a magnetic pole. I thought you had kids. :-) 
21 Apr
21 Apr
9:50 p.m.
Allison wrote:
Mike, PCI is 33mhz and core at best is 1mhz tthat
would eb an expensive
slow joke.
Not many PCI devices can support random reads in fewer than four to five
clock cycles. A core memory board would just take more cycles. So what?
Why is hooking up a small core memory to PCI any more of a joke than building
it in 1999 in general? As I recall, you proposed making about 256 bytes.
It's not like you're going to be able to run an operating system out of it
on *any* machine, modern or classic.
22 Apr
22 Apr
3:41 a.m.
Not many PCI devices can support random reads in fewer than four to five
clock cycles. A core memory board would just take more cycles. So what?
Because what would the point be for a pc with core and an access time of
1000 to 10,000ns per byte/word? not mentioning the array if hand made
with logic and read amps would likely be bigger than the mother board.
Windows would likely go belly up with memory delays like that. Of course
the array would ahve to be howmany bits wide to support PCI?
It's not like you're going to be able to run
an operating system out of it
on *any* machine, modern or classic.
Nor out of 4k in a PDP-8 according to some.
However with 256Words of core working building a simple TTL machine around
it would be of some educational value. It is however snough space to run
practical demonstration programs.
Allison
27 Apr
27 Apr
2:20 a.m.
At 08:41 22/04/99 -0400, allisonp(a)world.std.com wrote:
However with 256Words of core working building a simple
TTL machine around
it would be of some educational value. It is however snough space to run
practical demonstration programs.
There was a construction project in a local electronics mag (Electronics
Australia) around the same time as the original Altair articles describing
how to build a computer closely based on the PDP-8. This system had 256
words of memory. I should dig up the articles and this would probably be a
neat starting point for a CDS (Core Demonstration System). Please note: CDS
is a trademark of HOC2 (Huw's Old Computer Company - cf EMC2) :-)
Huw Davies | e-mail: Huw.Davies(a)latrobe.edu.au
Information Technology Services | Phone: +61 3 9479 1550 Fax: +61 3 9479
1999
La Trobe University | "If God had wanted soccer played in the
Melbourne Australia 3083 | air, the sky would be painted green"
22 Apr
22 Apr
6:22 a.m.
At 06:50 AM 4/22/99 -0000, Eric Smith <eric(a)brouhaha.com> wrote:
Why is hooking up a small core memory to PCI any more of a joke than building
it in 1999 in general? As I recall, you proposed making about 256 bytes.
It's not like you're going to be able to run an operating system out of it
on *any* machine, modern or classic.
You'd think there would've been more motivation to remove the
battery-backed nature of the CMOS RAM in today's PCs, and that
core-like replacements would make more sense in situations
where (for example) it would be difficult to replace the coin cell
or AA-pack or whatever every 12-18 months. So many failures
of PCs (and delightful amounts of "service" fees to repair them)
involve failure of the battery for the CMOS RAM.
Of course, this can lead to the discussion of real-time clock
chips with enclosed batteries with a finite lifetime...
At 08:41 AM 4/22/99 -0400, allisonp(a)world.std.com wrote:
Because what would the point be for a pc with core and an access time of
1000 to 10,000ns per byte/word?
That's hard disk territory, right? :-)
- John
7:56 a.m.
core-like replacements would make more sense in
situations
where (for example) it would be difficult to replace the coin cell
or AA-pack or whatever every 12-18 months. So many failures
of PCs (and delightful amounts of "service" fees to repair them)
involve failure of the battery for the CMOS RAM.
12-18months????? yours must be preaged about 60months when you get them.
Because what
would the point be for a pc with core and an access time of
1000 to 10,000ns per byte/word?
That's hard disk territory, right? :-)
22 Apr
22 Apr
12:12 a.m.
Mike wrote:
Actually I was thinking closer to a solenoid, and the whole array with
manual buttons that would need to be pressed, very hands on and see what
happens.
Perhaps a
smaller array, but with LARGE cores where the magnetic state could have
some visual indicators.
How would that work? Would you have an LED for each bit or something?
3:52 a.m.
> Actually I was thinking closer to a solenoid, and
the whole array with
> manual buttons that would need to be pressed, very hands on and see what
> happens.
The logical error there is that a torid is a closed field and hence cannot
operate a solinoid (open field). The magnetic path of a core is a closed
loop.
allison
8:30 a.m.
> Actually
I was thinking closer to a solenoid, and the whole array with
> manual buttons that would need to be pressed, very hands on and see what
> happens.
The logical error there is that a torid is a closed field and hence cannot
operate a solinoid (open field). The magnetic path of a core is a closed
loop.
3:35 a.m.
Mike wrote:
I can see providing a switch and a led for the circuit to set/clear the
bit. But displaying it's magnetic state like a realy might is kinda out
there. ;)
Allison
Perhaps a
smaller array, but with LARGE cores where the magnetic state could have
some visual indicators.
How would that work? Would you have an LED for each bit or something?
8:25 a.m.
Mike wrote:
I can see providing a switch and a led for the circuit to set/clear the
bit. But displaying it's magnetic state like a realy might is kinda out
there. ;)
Allison Perhaps a
smaller array, but with LARGE cores where the magnetic state could have
some visual indicators.
How would that work? Would you have an LED for each bit or something?
3:58 a.m.
At 06:45 AM 4/22/99 -0000, Eric wrote:
Mike wrote:
For a picture of a "large" core memory
(100 10 digit words for the ENIAC!)
look at:
http://www.cbi.umn.edu/photo/cbi/eniabro5.htm
and the previous pages. The wires are large enough to go to the Jones
strips on the right of the photo.
Perhaps a
smaller array, but with LARGE cores where the magnetic state could have
some visual indicators.
9219
days inactive
9225
days old
38 comments
13 participants
participants (13)
-
allisonp＠world.std.com -
ard＠p850ug1.demon.co.uk -
cmcmanis＠mcmanis.com -
dastar＠ncal.verio.com -
ddameron＠earthlink.net -
eric＠brouhaha.com -
H.Davies＠latrobe.edu.au -
Hans.Franke＠mch20.sbs.de -
jdykstra＠nortelnetworks.com -
jfoust＠threedee.com -
max82＠surfree.com -
mikeford＠netwiz.net -
pechter＠pechter.dyndns.org