10 Mar
2005
10 Mar
'05
8:29 p.m.
woodelf wrote:
The main problem I see with the current
crop of PDP-8 >clones/replica's today is
lack of paper tape and other hardware
at the time -- non volatile memory.
Actually, it's possible to build a reader fairly
easily. I liberated read mechanisms from Remex
"Director" readers I picked up cheap on eBay, and
built a microcontroller stepper controller/serializer
(which was fairly trivial-the power to the stepper is
done with 0-brainer power FETs; it was not difficult
to reverse engineer the connections to the head). The
smallest one I built is not much bigger than the read
mechanism itself, and runs on an old laptop power
supply. It will do only about 60 CPS mostly due to
power limitations, but how fast can you handle tape (I
have reeled units that I built serializers for also).
All of the readers and mechs that I have are
industrial strength, and I beleive that in home use
the mechanisms will last essentially forever
(certainly longer than I will). Older Remex's used
now-unobtainable lightbulbs, which I have replaced as
needed with high-power LEDs (the "Intel" readers used
Remex mechanisms).
The oldest reader I have is an Oktronics (NIB) which
had a very special (and broken) lightbulb. The #$@$#@
lightbulb cost $42-I couldn't bear hacking LED's into
it.
It is possible to build a "slot" type reader with
LED's and photodiodes; I built one but abandoned it
when I built the mechanized units because they are
"neater".
The real problem in a paper tape "ecosystem" is in the
punches, "handling equipment" and the tape itself; I
got lucky and got a lot of Mylar cheap some time ago,
as well as some real DEC fanfold and reel tape. Mylar
tape is now fabulously expensive at retail.
The older DSI punches are very rebuildable; I got a
very beat-up one in, replaced the belts (standard) and
the caps in the power supply, made a replacement chad
box from a $2.00 radio shack project box and some
Velcro and it works great. If the electronics fails (I
think that they use an F8 microprocessor) I'll just
replace it with work-alike modern electronics.
I have a FACIT reader that was severely damaged in
shipping, but it has a good punch head. I intend to
make a smaller case for it and run it with a
microcontroller and FETs; I think I can reduce the
power requirements by only punching one hole at a time
(obviously, it will be slower, but who cares). Most of
the size/weight in the thing is in the colossal linear
power supply. Again, this mechanism in home use could
last a very long time.
Paper tape is a PITA and fun at the same time. I like
to point out to friends that the data on the
metallized Mylar tape will be readable (even
visually!) and usable long, long after their hard
disks are very deep down in landfills!
-Steve Loboyko
Website: http://juliepalooza.8m.com/sl
Nixie Watch (one-tube):http://juliepalooza.8m.com/sl/complwatch.htm
__________________________________________________
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.yahoo.com
11 Mar
11 Mar
10:48 a.m.
On Thu, 10 Mar 2005, Loboyko Steve wrote:
Actually, it's possible to build a reader fairly
easily. [...]
[...]
The real problem in a paper tape "ecosystem"
is in the
punches, "handling equipment" and the tape itself; I
got lucky and got a lot of Mylar cheap some time ago,
as well as some real DEC fanfold and reel tape. Mylar
tape is now fabulously expensive at retail.
I concur with all you wrote. There are a lot of 1970's era tape
reader mechanisms around, and they're really easy to interface to.
I made one from a Plessey read head (they're common) and a PIC,
basically it steps until it sees the sprocket hole, then outputs
the byte. (Sprocket hole is smaller, therefore you get good
overlap).
Punches are indeed the harder part. I've got a DSI in decent
shape, nice serial interface for general use, but punch mechanisms
are harder to find, esp. ones easily interfaced. THere seems to be
10 times the number of readers as punches (makes perfect sense in
context of what paper tape was must used fo rin the computer
world).
Paper tape is a PITA and fun at the same time. I like
to point out to friends that the data on the
metallized Mylar tape will be readable (even
visually!) and usable long, long after their hard
disks are very deep down in landfills!
Yup! Even the non-oiled paper will last 500 years if it's kept
dry. It indeed suffers in the data-density department but it's fun
to work with on a non-production basis!
11:31 a.m.
Tom Jennings wrote:
Punches are indeed the harder part. I've got a DSI in decent
shape, nice serial interface for general use, but punch mechanisms
are harder to find, esp. ones easily interfaced. THere seems to be
10 times the number of readers as punches (makes perfect sense in
context of what paper tape was must used fo rin the computer
world).
Well I could buy a used serial punch for about $800 to $1000 us
but they are not for a song like many people have picked up 8's
with a tty or high speed punch. The lack of a punch has kept me from
doing a PDP/8 in a CPLD as well as not knowing of where to get
ferromagnetic? ram. I want real non-voiltile memory.
Yup! Even the non-oiled paper will last 500 years if
it's kept
dry. It indeed suffers in the data-density department but it's fun
to work with on a non-production basis!
Keeping dry is the hard part unless you got a water tight storage for tape.
Accidents with water like a broken water line, flood or heavry rain just
wait to happen.
Ben alias woodelf
11:55 a.m.
Ben wrote:
as well as not knowing of where to get
ferromagnetic? ram. I want real non-voiltile memory.
Ramtron ferroelectric RAM (nonvolatile) is available from Newark
Electronics. The traditional problem with FRAM is that it had limited
read endurance, as well as limited write endurance. Just reading the
part would wear it out. The newest 3V parts have fixed that.
The FM18L08 is a 32K*8 3V 70ns part with unlimited endurance. IIRC,
the last time I bought some from Newark I paid under $10 each.
There is a newer 128K*8 55ns part, the FM20L08. Newark doesn't have
it yet. All American Direct gives a price of $25.59, but it is not
clear whether they actually have parts in stock.
If you want to use these in a 5V system, it's probably a good idea
to use some 74LVC buffers that can function as level translators.
There are also some 128K*8 nonvolatile parts from Simtek, e.g.,
STK14CA8 and STK16CA8. They are not ferroelectric, but instead combine
SRAM and flash on one part, with automatic transfers between the two at
powerup and powerdown. Since they use SRAM, they are much faster
than the Ramtek parts. They also respond to a series of writes to magic
addresses to allow software controlled transfers, in a manner similar
to the Dallas Semiconductor battery-backed RAM with RTC. Although the
probability of accidental activation of that feature is miniscule,
I still don't much like it.
Eric
12:50 p.m.
Eric Smith wrote:
Ramtron ferroelectric RAM (nonvolatile) is available
from Newark
Electronics. The traditional problem with FRAM is that it had limited
read endurance, as well as limited write endurance. Just reading the
part would wear it out. The newest 3V parts have fixed that.
The FM18L08 is a 32K*8 3V 70ns part with unlimited endurance. IIRC,
the last time I bought some from Newark I paid under $10 each.
But is that DIP packaging? I got the data sheet but interfacing to 5v
logic and power
up is where I have lack of knowlage. I still think a generic FRAM
board needs
to be made with 3V parts to replace CORE memory in older machines
but I have
no idea how to interface 3V logic with 5 volt logic NOT using CMOS parts.
There is a newer 128K*8 55ns part, the FM20L08. Newark
doesn't have
it yet. All American Direct gives a price of $25.59, but it is not
clear whether they actually have parts in stock.
I still am dealing with classic small machines ... 2 FM18L08's replace a
whole rack of
core memory for a PDP/8. @ 2 cents a bit that is $8K for 8 banks of 4K
word memory
or $500 1980? $ for 4K.
If you want to use these in a 5V system, it's
probably a good idea
to use some 74LVC buffers that can function as level translators.
SMT I think.
Eric
Looking at orginal PDP / 8 ( 1965) prices $800 for a paper tape
punch/reader today is
a great deal. $15,000 would buy you a home back then.
Ben
1:29 p.m.
I wrote:
The FM18L08 is a 32K*8 3V 70ns part with unlimited
endurance. IIRC,
the last time I bought some from Newark I paid under $10 each.
Ben wrote:
But is that DIP packaging?
No, SOIC. They used to make DIPs but they are phasing them out.
There may still be some in the distribution channel, but don't
count on future availability.
If all else fails you can solder the SOICs onto carrier boards that
will plug into DIP sockets.
I got the data sheet but interfacing to 5v logic and
power
up is where I have lack of knowlage. I still think a generic FRAM
board needs to be made with 3V parts to replace CORE memory in older
machines but I have no idea how to interface 3V logic with 5 volt logic
NOT using CMOS parts.
What's wrong with CMOS parts for interfacing?
I wrote:
If you want to use these in a 5V system, it's
probably a good idea
to use some 74LVC buffers that can function as level translators.
Ben wrote:
SMT I think.
Available in DIP. 74LVC runs on 3V but has 5V tolerant inputs, so it
can be used to interface 3V CMOS to 5V TTL in both directions.
When powered by 3V, 74LVC outputs will not go above 3V, so they can't
drive pure 5V CMOS inputs. However, many 5V VLSI CMOS parts will work
with TTL-level inputs (min 2.0V Vih), so 3V 74LVC can drive those
just fine.
In general, if you want to build things with state-of-the-art chips,
you're going to have to deal with SMT, whether you want to or not.
Eric
1:44 p.m.
Eric Smith wrote:
In general, if you want to build things with
state-of-the-art chips,
you're going to have to deal with SMT, whether you want to or not.
I would rather not. SMT is something my craftmanship can't handle.
Mostly I still
would like to say with 5 volt logic and DIP/PLCC packaging because they
are still
easy to get prototypes PCB's made for them. I am not quite ready to try
the 'toaster
oven' just yet.
Eric
2:01 p.m.
On Fri, 11 Mar 2005, woodelf wrote:
Eric Smith wrote:
Regular .050" lead pitch SMT parts are very easy to hand solder. Certainly not
a lot harder than DIPs. Heck I've done .8 mm ball pitch BGAs (CSP) with
nothing more than flux and a Grainger hot air gun...
Peter Wallace
In general, if you want to build things with
state-of-the-art chips,
you're going to have to deal with SMT, whether you want to or not.
I would rather not. SMT is something my craftmanship can't handle. Mostly I
still
would like to say with 5 volt logic and DIP/PLCC packaging because they are
still
easy to get prototypes PCB's made for them. I am not quite ready to try the
'toaster
oven' just yet.
Eric
2:29 p.m.
woodelf wrote:
Eric Smith wrote:
Do it the simple, but inelegant way. 6264 CMOS DRAMs and a couple of AA
batteries.
Gordon.
In general, if you want to build things with
state-of-the-art chips,
you're going to have to deal with SMT, whether you want to or not.
I would rather not. SMT is something my craftmanship can't handle.
Mostly I still
would like to say with 5 volt logic and DIP/PLCC packaging because they
are still
easy to get prototypes PCB's made for them. I am not quite ready to try
the 'toaster
oven' just yet.
> Eric
12 Mar
12 Mar
12:41 a.m.
Gordon JC Pearce wrote:
Do it the simple, but inelegant way. 6264 CMOS DRAMs and a couple of
AA batteries.
Hmm 8Kx8 ram ... thats the dreadied BYTE ... back back you cursed thing.
Now a 8Kx9 ram would be very useful for me. 16kx4's useful for PDP 8's.
Well I am using 16kx4's and 16kx1's for a homebrew 18/9 bit computer that
is to give the illusion of a mini-computer in the 1980 to 1985 time
frame that
could have been built had the 8 bit micro not advanced into a 16 bit machine
as quick as it did. Since the computer does not have overflow and stack
operations
are shoe-horned into the CPU it is more a digital controler than a real
computer
like say a PDP 11 but it does have some logic that may make compiler writing
simplier. I'll give more details when I am done but I hope to have the
flavor of
TTL based computer had you been able to get a ALU and Flip/Flops designed
around octal digits.
11 Mar
11 Mar
3:12 p.m.
>>>> "woodelf" == woodelf
<bfranchuk at jetnet.ab.ca> writes:
woodelf> I would rather not. SMT is something my craftmanship can't
woodelf> handle. Mostly I still would like to say with 5 volt logic
woodelf> and DIP/PLCC packaging because they are still easy to get
woodelf> prototypes PCB's made for them. I am not quite ready to try
woodelf> the 'toaster oven' just yet.
You might want to give it a try. With the exception of BGA, which is
for specialists only, SMT parts can be handled perfectly well in the
workshop with nothing more than a good soldering iron, thin solder, a
bright light, and a magnifying glass.
You might want to look for some back issues of QST (or perhaps QEX);
there have been some good articles about SMT for amateurs in there.
Included were examples of people who made boards for SOIC parts with a
Dremel tool, or an Xacto knife. I personally prefer a CAD program and
a board shop, but for small pin counts I'll admit the other is
workable. Yes, if you want to use 240 lead PLCC parts, you'd better
farm out the PCB fabrication to a good board shop, but all the rest
you can do yourself.
It's hard to imagine a PCB shop that can't make boards for those
applications...
paul
3:26 p.m.
Eric Smith wrote:
If all else fails you can solder the SOICs onto carrier
boards that
will plug into DIP sockets.
This is most likely what I will do.
What's wrong with CMOS parts for interfacing?
I have a rug and a cat in my apartment. Meow walk ... zap
I forgot to add , I like things socketed.
Available in DIP. 74LVC runs on 3V but has 5V tolerant
inputs, so it
can be used to interface 3V CMOS to 5V TTL in both directions.
When powered by 3V, 74LVC outputs will not go above 3V, so they can't
drive pure 5V CMOS inputs. However, many 5V VLSI CMOS parts will work
with TTL-level inputs (min 2.0V Vih), so 3V 74LVC can drive those
just fine.
Now what about driving leds?
In general, if you want to build things with
state-of-the-art chips,
you're going to have to deal with SMT, whether you want to or not.
Straight 8's look better and better all the time. :)
3:40 p.m.
I wrote:
What's wrong with CMOS parts for interfacing?
Ben wrote:
I have a rug and a cat in my apartment. Meow walk ...
zap
If your static situation is that bad, you'd better not use any
RAM, EPROM, or Flash chips made in the last 15 years. They're all
CMOS.
I forgot to add , I like things socketed.
I like things reliable. Soldered is more reliable than socketed.
If you *must* use sockets, use high quality machined-pin sockets.
I wrote:
Available in DIP. 74LVC runs on 3V but has 5V
tolerant inputs, so it
can be used to interface 3V CMOS to 5V TTL in both directions.
Ben wrote:
Now what about driving leds?
Why do you need your RAM to drive LEDs? Anyhow, 74LVC should drive
LEDs just fine.
Straight 8's look better and better all the time.
:)
I'd love to get one. Closest I have is an 8/I.
The only nontrivial electronic computing devices I have which use
no ICs are Friden EC130 and EC132 desktop electronic calculators, circa
1964 and 1965, which use lots of transistors, a delay line memory, and a
CRT display.
Eric
3:48 p.m.
Eric Smith wrote:
So would we all. :)
I like things reliable. Soldered is more reliable than
socketed.
If you *must* use sockets, use high quality machined-pin sockets.
I got them ... well on order. With sockets if I zap a chip I can
replace it.
Why do you need your RAM to drive LEDs? Anyhow, 74LVC
should drive
LEDs just fine.
I never thought of that ... I plan to have the tradional front panel on
all my computers
I build.
Straight
8's look better and better all the time. :)
I'd love to get one. Closest I have is an 8/I.
Well anyhow I do plan to use a lot CMOS but not for I/O that is
connected to
people. I just don't want to solder in chips. But then a PDP 8/ non
volitile memory
is my next project in a few years.
Ben.
4:33 p.m.
I have a rug
and a cat in my apartment. Meow walk ... zap
If your static situation is that bad, you'd better not use any RAM, EPROM, or Flash chips made in the last 15 years.
They're all
CMOS.
For prototyping / experimatal work I always use sockets if I can. It
makes debugging a lot easier (you can pull the chip and force pins to
particular levels). I only ever use what we called 'turned pin' sockets
(and I guess are the same as 'machined pin' ones, and I don't think I've
ever had a bad contact on one. I've had plenty of problems with cheaper
sockets, though... My time may not be worth much, but it's worth enough
not to want to deal with that sort of problem...
I forgot to add , I like things socketed.
I like things reliable. Soldered is more reliable than socketed.
If you *must* use sockets, use high quality machined-pin sockets. Straight
8's look better and better all the time. :)
I'd love to get one. Closest I have is an 8/I.
The only nontrivial electronic computing devices I have which use
no ICs are Friden EC130 and EC132 desktop electronic calculators, circa
1964 and 1965, which use lots of transistors, a delay line memory, and a
CRT display.
The only no-IC machine I have is a Casio AL1000 calculator, from around
1967 -- IIRC it was sold under the Commodore name as well. It's all
discrete transsitors and diodes, core memory, and a row of nixie tubes
for the display.
Somebody is going to say 'don't you have an HP9100'. Yes, I have a pair
of 9100Bs, but they have 8 ICs in them. 8 op-amps on the magnetic card
reader PCB (read amplifer and comparator for the 3 data tracks and the
timing track). The logic, of course, is all discrete transistors and
diodes, with a core read/write memory, core-on-a-rope microcode and that
inductively coupled PCB main firmware ROM.
-tony
5:33 p.m.
On Fri, 11 Mar 2005, Eric Smith wrote:
CMOS isn't *THAT* sensitive; easy, benign, reasonable precautions
are more than adequate. Ben, the days of early MOS/CMOS
hypersensitivity to static are long gone. Even 1970's 4000 series
had input-protection diodes.
When I sit down to work on stuff I just make sure to touch chassis
first, or other ground, then tools and board and parts in order,
it's not a big deal, and I haven't static-harmed a part in 20
years, if I ever did ever.
Vintage MOS memories are another story.
What's
wrong with CMOS parts for interfacing?
Ben wrote:
I have a rug and a cat in my apartment. Meow
walk ... zap
If your static situation is that bad, you'd better not use any
RAM, EPROM, or Flash chips made in the last 15 years. They're all
CMOS.
7:31 a.m.
>>>> "William" == William
Donzelli <aw288 at osfn.org> writes:
> I have aluminum foil ... :)
William> A
cheap substitute, but it is actually not perfect. With ESD
William> safe facilities, you want some resistance, not zero. You
William> want charges to dissipate relatively slowly, rather than a
William> quick zap.
Maybe. I think the main reason for the 1 Mohm resistor normally used
in wrist straps is safety: if you touch a live wire, you won't get a
substantial current via the strap.
Another issue with aluminum foil is that it's coated with an excellent
insulator (aluminum oxide) which may or may not be thick enough to
interfere with what you were trying to do.
paul
9:32 a.m.
Maybe. I think the main reason for the 1 Mohm
resistor normally used
in wrist straps is safety: if you touch a live wire, you won't get a
substantial current via the strap.
Certainly an issue, but it is the secondary reason. A lesser value
resistor could be used for this reason and still be reasonably safe,
however a decently sized resistor is needed for those multi-kV zaps.
Somehow I imagine that a comittee was formed, spent oodles of time and
money, and came up with 1 Mohm as a comprimise.
William Donzelli
aw288 at osfn.org
8:06 a.m.
New subject: Proper Grounding Techniques (was: Troubleshooting...
Rumor has it that William Donzelli may have mentioned these words:
But then you don't have "tactile feedback." 'Course, you can always make
sure the kids are nearby... ;-)
Dontcha just need to put a 1MegOhm resistor between the "Grounding Station"
(i.e. foil) and the earth ground connection?
Come to think of it, I should build a grounding station area near my
TV/stereo/TiVo area - getting off the couch and welding your finger to the
TV can't be good...
Laterz,
Roger "Merch" Merchberger
--
Roger "Merch" Merchberger | A new truth in advertising slogan
SysAdmin, Iceberg Computers | for MicroSoft: "We're not the oxy...
zmerch at 30below.com | ...in oxymoron!"
I have
aluminum foil ... :)
A cheap substitute, but it is actually not perfect. With ESD safe
facilities, you want some resistance, not zero. You want charges to
dissipate relatively slowly, rather than a quick zap.
8:29 a.m.
New subject: Proper Grounding Techniques
Roger Merchberger wrote:
Dontcha just need to put a 1MegOhm resistor between the "Grounding
Station" (i.e. foil) and the earth ground connection?
Come to think of it you do. How ever in my case the only time I'll be
using a ground for static
electricity control is when I'll be inserting the chips and that will be
with power off. Now modern
production of PCB's are a whole different matter.
Ben alias woodelf
9:47 a.m.
New subject: Proper Grounding Techniques
Roger Merchberger wrote:
The idea of the 1M resistor (actually, there should be two 1M resistors in
series, in case one fails), is to protect the engineer from shock IN BOTH
DIRECTIONS, not only touching live parts within the equipment, but also if
you come between a "floating" earth (Earth conductor in the UK can be
several 10s of volts - I've measured it as high as 80V above true earth) and
true ground.
I can recomend the 3M products. These are a dissipative type mat, which has
a conductive layer between two insulating ones, and discharges static by a
capacitive method, the usual wrist straps, and commoning blocks. They also
produce a field service kit, consisting of a wrist strap, dissipative mat
(with pockets - it will double as an ESD bag), and a grounding point which
plugs into a standard mains outlet. The grounding unit has a self test
facility for both the ground and the wrist strap.
Dissipative mats are better, as you can place a board containing a backup
battery on them, without worrying about shorting it out!
There are also various mat and starp testing products, but for most of us,
these would be a luxury.
Jim.
Dontcha just need to put a 1MegOhm resistor between the "Grounding
Station" (i.e. foil) and the earth ground connection?
Come to think of it you do. How ever in my case the only time I'll be
using a ground for static
electricity control is when I'll be inserting the chips and that will be
with power off. Now modern 
10:24 a.m.
New subject: Proper Grounding Techniques
The idea of the 1M resistor (actually, there should be two 1M resistors in
series, in case one fails..
Considering that resistors might open, but rarely short...i would expect
this to actually be parallel....
10:36 a.m.
New subject: Proper Grounding Techniques
From: "David V. Corbin" <dvcorbin at optonline.net>
The idea of the 1M resistor (actually, there should be two 1M resistors in
series, in case one fails..
Considering that resistors might open, but rarely short...i would expect
this to actually be parallel....
No, the resistor is there to protect the engineer, not the equipment, hence
series.
Jim.
13 Mar
13 Mar
11:18 a.m.
New subject: Proper Grounding Techniques
David wrote:
Considering that resistors might open, but rarely
short...i would expect
this to actually be parallel....
Jim wrote:
No, the resistor is there to protect the engineer, not
the equipment,
hence series.
If they're in series, but one fails open, it's not going to be protecting
either the engineer or the equipment.
If you have two resistors in parallel, and one fails open, you'll still
have some protection.
If you're really worried about a resistor failing as a short (which I've
only ever seen once), you could use four resistors, with two parallel
pairs in series. Then any one resistor can fail either open or short,
and you'll still have some protection.
Eric
1:40 p.m.
New subject: Proper Grounding Techniques
From: "Eric Smith" <eric at brouhaha.com>
David wrote:
But the engineer is protected by an open resistor - there is no path to
ground for a potentially fatal shock form the mains. The risk of static
damage is always secondary to the risk of a fatality. Remeber, the worst
case shock is one where the current passes through the heart, which is a
huge risk if one hand is connected directly to ground by a short
circuit.........
Jim.
Considering that resistors might open, but rarely
short...i would expect
this to actually be parallel....
Jim wrote:
No, the resistor is there to protect the
engineer, not the equipment,
hence series.
If they're in series, but one fails open, it's not going to be protecting
either the engineer or the equipment.
3:14 p.m.
New subject: Proper Grounding Techniques
Jim Beacon wrote:
But the engineer is protected by an open resistor - there is no path to
ground for a potentially fatal shock form the mains. The risk of static
damage is always secondary to the risk of a fatality. Remeber, the worst
case shock is one where the current passes through the heart, which is a
huge risk if one hand is connected directly to ground by a short
circuit.........
And the static will probably leak away through the body of the resistor
*anyway*.
Gordon.
12 Mar
12 Mar
1:08 p.m.
New subject: Proper Grounding Techniques (was: Troubleshooting...
At 11:06 AM 3/12/05 -0500, Roger "Merch" Merchberger wrote:
Rumor has it that William Donzelli may have mentioned
these words:
But then you don't have "tactile feedback." 'Course, you can always make
sure the kids are nearby... ;-)
Dontcha just need to put a 1MegOhm resistor between the "Grounding Station"
(i.e. foil) and the earth ground connection?
Come to think of it, I should build a grounding station area near my
TV/stereo/TiVo area - getting off the couch and welding your finger to the
TV can't be good...
Hey, all the more reason to by that fancy new TV with the remote control :-)
Joe
I have
aluminum foil ... :)
A cheap substitute, but it is actually not perfect. With ESD safe
facilities, you want some resistance, not zero. You want charges to
dissipate relatively slowly, rather than a quick zap.
11 Mar
11 Mar
12:38 p.m.
>>>> "woodelf" == woodelf
<bfranchuk at jetnet.ab.ca> writes:
woodelf> ... as well as not
woodelf> knowing of where to get ferromagnetic? ram. I want real
woodelf> non-voiltile memory.
Try www.ramtron.com. Ferroelectric, by the way. Nice stuff, you can
pretend it's fairly slow SRAM but the bits stay when the power goes
off.
paul
12 Mar
12 Mar
5:01 a.m.
Use Dallas NVRAM's.
That's what I run in my Imlac to simulate core...
----- Original Message -----
From: "woodelf" <bfranchuk at jetnet.ab.ca>
To: "General Discussion: On-Topic and Off-Topic Posts"
<cctalk at classiccmp.org>
Sent: Friday, March 11, 2005 2:31 PM
Subject: Re: OT Troubleshooting: Old computers with modern parts
Tom Jennings wrote:
Punches are indeed the harder part. I've got a DSI in decent
shape, nice serial interface for general use, but punch mechanisms
are harder to find, esp. ones easily interfaced. THere seems to be
10 times the number of readers as punches (makes perfect sense in
context of what paper tape was must used fo rin the computer
world).
Well I could buy a used serial punch for about $800 to $1000 us
but they are not for a song like many people have picked up 8's
with a tty or high speed punch. The lack of a punch has kept me from
doing a PDP/8 in a CPLD as well as not knowing of where to get
ferromagnetic? ram. I want real non-voiltile memory.
Yup! Even the non-oiled paper will last 500 years
if it's kept
dry. It indeed suffers in the data-density department but it's fun
to work with on a non-production basis!
Keeping dry is the hard part unless you got a water tight storage for
tape.
Accidents with water like a broken water line, flood or heavry rain just
wait to happen.
Ben alias woodelf
7228
days inactive
7230
days old
31 comments
14 participants
participants (14)
-
ard＠p850ug1.demon.co.uk -
aw288＠osfn.org -
bfranchuk＠jetnet.ab.ca -
bshannon＠tiac.net -
dvcorbin＠optonline.net -
eric＠brouhaha.com -
gordon＠gjcp.net -
jim＠g1jbg.co.uk -
pcw＠mesanet.com -
pkoning＠equallogic.com -
rigdonj＠cfl.rr.com -
sloboyko＠yahoo.com -
tomj＠wps.com -
zmerch＠30below.com