30 May
2006
30 May
'06
1:28 p.m.
Tony Duell wrote:
So (to get it on-topic for classiccmp), people with machines that only
support IDE drives are going to have problems in the future finding
working replacements (I doubt very much if any of us could repair any
reasonably-modern IDE drive, alas). I've seen adapters to use IDE drives
on SATA hosts, but not the reverse.
Oh well... Looks like designing an ST506 (host) to IDE (drive) interface
is not going to be a particularly useful thing. Better to work out a way
to keep the old ST506s/412s/etc working.
-tony
------------------------------------------------------
I'm lost in your logic here. ST506 and 412 have been obsolete for 15 years.
They will become harder and harder to find and keep working. If you are
trying to keep the old systems running, you will eventually have to do a
dongle of some sort with newer drives. Why not now?
Keeping the old drives working is going to require parts and technology that
don't exist commercially any more. You can still find the high volume
circuitry in the surplus stores. But it's the heads and platters that wear
out and ferrite heads and oxide media are gone. Occasional bits and pieces
show up less and less often. Why not put the energy into adapting the
latest drive technology? That buys you an extra decade or so.
IDE drives are going to be around in far greater volumes long after the last
ST506 is still working. I just had a report that last year was the biggest
year ever for hard drives: 376 million units, the huge majority IDE. That
gives a pool of drives to use that will outlast most of us on this list.
And there are already IDE to SATA dongles to stretch out another generation.
And there are the side issues to keeping the old systems going. How long
will we be able to find 8" and 5.25" media? Printer ribbons? Rubber parts
for teletypes and typewriters? When do you toss in the towel on trying to
keep the old systems going?
One of my other hobbies is amateur publishing. We used Gestener mimeograph
machines. The machines are easy to keep running. But I haven't been able
to buy any stencils in the last 5 years. A place in India made them until
around 2000. Now it's just the odd cache that shows up on eBay every 4-6
months. The ink we've tried to make ourselves with a little success. And
of course, the twiltone paper is long gone. Modern paper is too slick and
non-absorbent. When we get a mimeo going, it still looks like crap because
of the poor ink and paper.
So I wonder again, when do we give up on keeping the old machines running?
If my CDC 160-A memory goes, there's no more core memories. My Jetex
engines don't have fuel any more. My old typewriters have platens that turn
to powder if I try to type.
Some time, we have to let go and upgrade as much as it is unaesthetic or
displeasing. When?
Billy
30 May
30 May
2:30 p.m.
On 5/30/2006 at 1:28 PM Billy Pettit wrote:
Some time, we have to let go and upgrade as much as it
is unaesthetic or
displeasing. When?
Now that's an interesting question.'
I think if "maintainability" is plotted against time, it often looks like
a bell-shaped curve. Near the peak of the curve, we have devices that use
available commodity components (e.g. 7400-series TTL and/or discretes), so,
for example, it's possible to keep a PC XT mobo going for a long time. Way
off on the back end of the curve are devices constructed from things that
were once commodity items, but which are old enough to be rare
(point-contact transistors, germanium diodes, UV-201As, your 160A core,
etc.). Off the front end of the curve are those devices constructed from
devices that, although only slightly out of date, are no longer available
and whose specs may never have been published outside of an internal
company document (try to replace a house-numbered ASIC on a 1990's PC
mobo).
If one wanted to make an ST-506 or 412 replacement, I'd be sore tempted
just to do it with some battery-backed SRAM and a microcontroller. 100 MB
was once a big hard drive, but it's not even big for SRAM nowadays. The
replacement might extend the useful life of your device by decades, since
there's no moving parts to wear out. I'm sure that you could do likewise
for your 160A core.
Someone once offered me a genuine Linotype hot-type typesetting machine. I
didn't take the guy up on his offer, but I wonder if there are any of the
beasts still in use.
Cheers,
Chuck
3:19 p.m.
On 5/30/2006 at 1:28 PM Billy Pettit wrote:
YEs, but most of those devices are well-understood, and there may well be
workarounds. If you take a 1920's radio using UV201As, you could probably
kludge FETs in place of them (doesn't some compay actually sell little
PCBs to upt over the pins of dead UV201As to do this?). Germanium diodes
can often be replaced with shottky diodes. And so on.
Some of the odder TTL chips are getting hard to obtain now, but at least
data sheets exist (I will _never_ give up my shelf of TI TTL
handbooks...). I can see myself programming some modern CPLD-thing to
replace an obscure TTL counter/decoder/whatever. Wasteful of resources in
the new chip, sure, but it'll keep my PDPs, P800s, HP9800s, etc running.
Some time, we have to let go and upgrade as much
as it is unaesthetic or
displeasing. When?
Now that's an interesting question.'
I think if "maintainability" is plotted against time, it often looks like
a bell-shaped curve. Near the peak of the curve, we have devices that use
available commodity components (e.g. 7400-series TTL and/or discretes), so,
for example, it's possible to keep a PC XT mobo going for a long time. Way
off on the back end of the curve are devices constructed from things that
were once commodity items, but which are old enough to be rare
(point-contact transistors, germanium diodes, UV-201As, your 160A core, etc.). Off the front end of the curve are those
devices constructed from
devices that, although only slightly out of date, are no longer available
and whose specs may never have been published outside of an internal
company document (try to replace a house-numbered ASIC on a 1990's PC
mobo).
That's the big problem. Even though you probably could replace one of
these ASCIs with a modern FPGA suitably programmed, since you don't have
the specs of it, you have a very long day ahaed of you.
-tony
4:11 p.m.
Tony Duell wrote:
That's the big problem. Even though you probably
could replace one of
these ASCIs with a modern FPGA suitably programmed, since you don't have
the specs of it, you have a very long day ahaed of you.
FPGA's and CPLD's seem to be going obsolite even before the datasheets
are printed or even updated on the web. The main problem with
programable logic is that almost all the software is windows based that
is low cost. Allmost all the high priced stuff is unix based and very
closed source in most aspects.
31 May
31 May
3:40 p.m.
Tony Duell wrote:
Tell me about it. About 10 years ago I was looking at the Xilinx XC6200
series -- this being one of the few FPGAs that were 100% documented (the
configuration bit pattern was documented, you didn't have to use the
Xilinx tools). They were discontinued before the UK distributor got any.
That, I am afraid, is one reason I avoid programmable logic chips. By the
time I've finished the design, the darn chip is unavailable.
That's the big problem. Even though you
probably could replace one of
these ASCIs with a modern FPGA suitably programmed, since you don't have
the specs of it, you have a very long day ahaed of you.
FPGA's and CPLD's seem to be going obsolite even before the datasheets
are printed or even updated on the web. The main problem with programable logic is that almost all the software is
windows based that
is low cost. Allmost all the high priced stuff is unix based and very
closed source in most aspects.
And that's the other reason I avoid them. Well, and the fact that it
takes me longer to design, debug, and program a device with those tools,
than to hand-wire a 1-off board of TTL-ish logic chips.
-tony
4:57 p.m.
Tony Duell wrote:
That, I am afraid, is one reason I avoid programmable
logic chips. By the
time I've finished the design, the darn chip is unavailable.
I bought my chips first ... and the programmer/demo board. :)
And that's the other reason I avoid them. Well, and the fact that it
takes me longer to design, debug, and program a device with those tools,
than to hand-wire a 1-off board of TTL-ish logic chips.
But alas I want to do a home brew 18 bit CPU and that don't fit well
with 4 bit wide TTL logic, so I got about 200 14/16 pin TTL for a 18 bit
ALU or two CPLDs. I still favor the TTL rather than CPLD's.
-tony
1 Jun
1 Jun
5:06 p.m.
Tony Duell wrote:
Hmm.. I always feel I am going to make a mistake and ruin the chip, so I
want to be sure of being able to get replacements, at least while I'm
still designing.
-tony
That, I am afraid, is one reason I avoid
programmable logic chips. By the
time I've finished the design, the darn chip is unavailable.
I bought my chips first ... and the programmer/demo board. :)
2 Jun
2 Jun
12:19 a.m.
Tony Duell wrote:
Hmm.. I always feel I am going to make a mistake and
ruin the chip, so I
want to be sure of being able to get replacements, at least while I'm
still designing.
In that case send me all your old DTL stuff, I had planned to
build one
since about 1972 or so... Only now I have the $$$.
-tony
12:28 a.m.
On Friday 02 June 2006 03:19 am, woodelf wrote:
Tony Duell wrote:
One what?
I always wondered what DTL was good for...
--
Member of the toughest, meanest, deadliest, most unrelenting -- and
ablest -- form of life in this section of space, a critter that can
be killed but can't be tamed. --Robert A. Heinlein, "The Puppet Masters"
-
Information is more dangerous than cannon to a society ruled by lies. --James
M Dakin
Hmm.. I always feel I am going to make a mistake
and ruin the chip, so I
want to be sure of being able to get replacements, at least while I'm
still designing.
In that case send me all your old DTL stuff, I had planned to build one
since about 1972 or so... Only now I have the $$$.
12:48 a.m.
On Friday 02 June 2006 03:40 am, woodelf wrote:
Roy J. Tellason wrote:
Yeah, but that ain't DTL, that's TTL, I still have a bunch of it. I forget
what the DTL numbers were but I think they're up in the 9xxx range.
--
Member of the toughest, meanest, deadliest, most unrelenting -- and
ablest -- form of life in this section of space, a critter that can
be killed but can't be tamed. --Robert A. Heinlein, "The Puppet Masters"
-
Information is more dangerous than cannon to a society ruled by lies. --James
M Dakin
On Friday 02 June 2006 03:19 am, woodelf wrote:
One what?
umm could be a computer or am I off topic :)
I always wondered what DTL was good for...
I allways remember the polypack ads ... 74xx for
only 5 cents each or so if I recall right. 
5:17 p.m.
Ethan Dicks wrote:
On 6/2/06, Roy J. Tellason <rtellason at blazenet.net> wrote:
RTL: 7xx, 8xx, 9xx (etc...)
DTL: 8xx, 9xx, 18xx, 19xx (etc...)
At the decade level the grouping makes more sense.
There were actually lots of other series of SSI RTL/DTL/TTL ICs made in the
60s, by both the major manufacturers and smaller players, in addition to the
various 'standard' series.
I received an email request from a fellow in Cuba a year or two ago, asking
for info about a non-standard TTL-class series of ICs (59xx) made by
Mitsubishi in the late 60s (I have info about some ICs of the series).
Apparently the series had been used in the first Cuban minicomputer
which he had worked on the design of, around 1970.
Yeah, but that ain't DTL, that's TTL,
I still have a bunch of it. I forget
what the DTL numbers were but I think they're up in the 9xxx range.
What about the Motorola 7xx and 8xx parts? DTL or RTL?
-ethan 
5:49 p.m.
New subject: MATE-N-LOK 8p flat connector availability
FYI,
I was looking for the old AMP MATE-N-LOK 8p flat connector with the tongue
(now out of production) and found a source.
State Electronics (http://www.state-elec.com) has them available, the
AMP part
number is 1-480460-0. I bought 100pcs at US$0.75/ea (a pretty decent price,
I had been quoted up to $3.00 each by another vendor).
These connectors are used in old DEC equipment for the 20ma interfaces, for
the BA11-K box power module connections, and on some front panels (11/70).
Just thought I'd pass this along, as they are now out of production and
hard to find.
The other matrix (6p, 9p, 12p, 15p) are still in production and widely
available.
Don North
ak6dn
3 Jun
3 Jun
3:09 p.m.
New subject: MATE-N-LOK 8p flat connector availability
The other matrix (6p, 9p, 12p, 15p) are still in
production and widely
available.
I am looking for a somewhat similar connector. I am not sure if it's AMP
or Molex, but I suspect it's one or the other.
It has 5 pins in a straight line. The pins look to be the normal 0.093"
ones. The face of the socket housing looks a bit like this :
/o\
|o|
|o|
|o|
|o|
-
That is, 5 pins in a line, with one end of the housing pointed for
polarisation.
It's used to connect the mains switch (in the keyboard assembly to the
PSU harness in the HP9810 and HP9820 calculators. I want a plug housing
and pins so I can make up a jumper plug to power up the machine with the
keyboard totally removed. Unfortuately the HP service manual is a
boardswapper guide and doesn't include any details of this connector (it
comes as part of the complete keyboard assembly).
Does anyone recognise it and/or know of a source.
-tony
2 Jun
2 Jun
5:49 p.m.
On Friday 02 June 2006 08:17 pm, Brent Hilpert wrote:
Ethan Dicks wrote:
There's much of that in my chart at:
http://www.classiccmp.org/rtellason/by-generic-number.html
but I have bunches more data to get, still.
On 6/2/06, Roy J. Tellason <rtellason at
blazenet.net> wrote:
RTL: 7xx, 8xx, 9xx (etc...)
DTL: 8xx, 9xx, 18xx, 19xx (etc...)
At the decade level the grouping makes more sense. Yeah, but that ain't DTL, that's TTL,
I still have a bunch of it. I
forget what the DTL numbers were but I think they're up in the 9xxx
range.
What about the Motorola 7xx and 8xx parts? DTL or RTL?
-ethan There were actually lots of other series of SSI
RTL/DTL/TTL ICs made in the
60s, by both the major manufacturers and smaller players, in addition to
the various 'standard' series.
I've heard that, in particular that Fairchild had made some early TTL in the
9000 numbers, I think it was? But I seldom run across any of that info at
all, these days, and didn't then. First TTL databook I have goes back to
about 1970, and is a TI book with the standard, H, and L series in it only.
I received an email request from a fellow in Cuba a
year or two ago, asking
for info about a non-standard TTL-class series of ICs (59xx) made by
Mitsubishi in the late 60s (I have info about some ICs of the series).
Apparently the series had been used in the first Cuban minicomputer
which he had worked on the design of, around 1970.
Mitsubishi? I didn't even know they made TTL. Feel free to pass any info
along and I'll include it in my charts...
--
Member of the toughest, meanest, deadliest, most unrelenting -- and
ablest -- form of life in this section of space, a critter that can
be killed but can't be tamed. --Robert A. Heinlein, "The Puppet Masters"
-
Information is more dangerous than cannon to a society ruled by lies. --James
M Dakin
6:11 p.m.
On 6/2/2006 at 8:49 PM Roy J. Tellason wrote:
I've heard that, in particular that Fairchild had
made some early TTL in
the 9000 numbers, I think it was? But I seldom run across any of that
info
at all, these days, and didn't then. First TTL
databook I have goes
back
to about 1970, and is a TI book with the standard, H,
and L series in it
only.
When I'm working something out that needs a dual one-shot, I use a 96LS02.
And let's not forget the Signetics 8Txxx TTL stuff.
Japanese numbers can be a problem if they're not conventional. Mitsubishi
and Fujitsu particularly. I've got lots of "what the heck is this thing"
samples of both.
Cheers,
Chuck
11:15 p.m.
On Friday 02 June 2006 09:11 pm, Chuck Guzis wrote:
On 6/2/2006 at 8:49 PM Roy J. Tellason wrote:
I don't use one-shots much myself, but I do have a couple of those, both the
LS and the straight 9602 versions, maybe one and two of them. I notice
that some designers seem to favor those though I'm not familiar enough with
the details to know just why that is.
Back in '78 when I was messing with that H11 system that at that time had _no_
external storage except for paper tape, the company got a hold of a dual
cassette drive to evaluate. I got the job of seeing if I could come up with
an interface for the '11, and did get as far as wire-wrapping a board but
never got the handshaking right and the machine never recognized it. I
recall a conversation on the phone, though, where this guy at the company
who'd sent out the drive advised *very* strongly against the '123 parts,
saying that they had some serious stability issues, or somesuch. I do know
I hardly ever see them used, with the 96x02s being more common for some
reason.
I've heard that, in particular that Fairchild
had made some early TTL in
the 9000 numbers, I think it was? But I seldom run across any of that
info
at all, these days, and didn't then. First
TTL databook I have goes
back
to about 1970, and is a TI book with the
standard, H, and L series in it
only.
When I'm working something out that needs a dual one-shot, I use a 96LS02. And let's not forget the Signetics 8Txxx TTL
stuff.
I did forget about those. I also remember seeing them in some databook here,
though with things in their current state of disarray I wouldn't begin to
guess which book that was. I also remember those with some other numbers
attached, maybe something like "8T97/8097" perhaps?
Japanese numbers can be a problem if they're not
conventional. Mitsubishi
and Fujitsu particularly. I've got lots of "what the heck is this thing"
samples of both.
I've managed okay with their transistors -- and those numbers seem to make up
the bulk of my data, for sure, as well as a lot of my salvage lately. The
chips, though, are another matter entirely.
And I haven't really even begun to seriously dig. There's both a mass of
stuff in books here that I could just enter some basic info on, and a whole
lot more out there on the 'net that I haven't dug into yet -- most of that
stuff came from one or two sites, which I stopped using when they started
"branding" datasheets, none of which I've uploaded. I've a couple of
others
to hit for replacements, and then all of the manufacturer's sites yet.
One guy had a bunch of stuff on a web page a while back and it looked like a
link to my pages would fit in well, so I dropped him an email. The one word
that stands out in my recollection from his response was "Ambitious!" :-)
Not that I ever actually started out to try and make this project happen,
initially. It was more a matter of getting some data on some parts I
happened to have around (which is why there are part entries with no data at
all, I have 'em and haven't found any yet) and arranging what data I did
find into a form that I found useful. I'm glad to see some other folks
finding it useful as well, and it looks like more folks are doing so lately,
particularly since I added the datasheets to the pile since I changed hosts.
More data in whatever form is welcome, and I'll add it as time permits.
--
Member of the toughest, meanest, deadliest, most unrelenting -- and
ablest -- form of life in this section of space, a critter that can
be killed but can't be tamed. --Robert A. Heinlein, "The Puppet Masters"
-
Information is more dangerous than cannon to a society ruled by lies. --James
M Dakin
1 a.m.
On 6/2/2006 at 1:40 AM woodelf wrote:
I allways remember the polypack ads ... 74xx for
only 5 cents each or so if I recall right.
I remember buying 923 RTL FFs from them for $1 apiece and thinking that it
was a steal. Typical floor sweepings--some packages didn't even have the
leads plated.
I still have a couple of Fairchild 8 bit addressable memories that I bought
from them for way too much. DTL, requiring a 6 volt
supply and a 7 volt
clock. Ran hot as a pistol.
My first 741 op amp came from them. My first LED's were purchased directly
from Montsanto as part of their
"experimenter's program".
Cheers,
Chuck
7:03 a.m.
Hi
I have 9xx DTL in my Nicolet 1080 and it runs on 5 volts, not 6 volts.
DTL can, for the most part be replaced by TTL with the same pinouts.
That is except one case. I don't recall the number but there is a DTL
gate that can be expanded externally with more diodes to create
a wider input.
I built an electronic dice using RTL once. RTL runs at 3.6 volts.
It used a number of jk flops and drove some small LEDs that were
aranged like dice. It ran on a 9 volt battery through a regulator.
It flattened the battery in just a few minutes of operation.
Dwight
From: "Chuck Guzis" <cclist at
sydex.com>
On 6/2/2006 at 1:40 AM woodelf wrote:
I allways remember the polypack ads ... 74xx for
only 5 cents each or so if I recall right.
I remember buying 923 RTL FFs from them for $1 apiece and thinking that it
was a steal. Typical floor sweepings--some packages didn't even have the
leads plated.
I still have a couple of Fairchild 8 bit addressable memories that I bought
from them for way too much. DTL, requiring a 6 volt supply and a 7 volt
clock. Ran hot as a pistol.
My first 741 op amp came from them. My first LED's were purchased directly
from Montsanto as part of their "experimenter's program".
Cheers,
Chuck
9:52 a.m.
On 6/2/2006 at 7:03 AM dwight elvey wrote:
I have 9xx DTL in my Nicolet 1080 and it runs on 5
volts, not 6 volts.
DTL can, for the most part be replaced by TTL with the same pinouts.
That is except one case. I don't recall the number but there is a DTL
gate that can be expanded externally with more diodes to create
a wider input.
Early 7400-series TTL had the 7450 and 7453 AND-OR-INVERT gates that could
be explanded with another 4 inputs via the 7460 dual 4-input expander. I'd
hate to own equipment now that used some of those that required
replacement--they must be very difficult to find.
The bigger problems in replacing DTL with TTL (other than the obvious
impedance mismatch issues) is that DTL could use a relatively wide range of
supply voltages (up to 8 volts) and outputs could be wire-OR-ed with
relative impunity (i.e. you didn't have to consult the databook to see if
you could do it). And yes, expandable gates were more common. In addition
to flat pack and DIP, DTL could also be furnished in TO-100 cans; which led
to the interesting situation that you could have two packages with the same
base part number but with a different number of active pins (e.g. MC833P
vs. MC833G). For Moto, DTL devices were, by and large, the 830/930- 960
families, with some late parts having numbers in the 1800-1900 range.
There was an old DTL 200/250 series, but I've never seen equipment that
used it. There was also a "radiation hardened" line of DTL, the MCE930
series that featured nichrome resistors and dielectric isolation of circuit
elements from the substrate.
RTL also featured expandable gates, but with an interesting twist the
expanders (basically open-collector gates) could be used on just about any
other RTL device--you didn't need the base device to be one of the
"expandable" versions Again, wired-OR operation was very common.
Where I'd be stuck is finding replacements for HTL--15 volt Vcc and the
switching threshold somewhere around 7 volts.
Cheers,
Chuck
10:08 a.m.
On Friday 02 June 2006 12:52 pm, Chuck Guzis wrote:
Where I'd be stuck is finding replacements for
HTL--15 volt Vcc and the
switching threshold somewhere around 7 volts.
Is that stuff pretty much totally gone, then? My last ECG book (not recent)
still listed it, I've asked NTE to send me some literature but it hasn't
arrived yet and I don't know if they support it as well, or if it's not
included in their lineup...
--
Member of the toughest, meanest, deadliest, most unrelenting -- and
ablest -- form of life in this section of space, a critter that can
be killed but can't be tamed. --Robert A. Heinlein, "The Puppet Masters"
-
Information is more dangerous than cannon to a society ruled by lies. --James
M Dakin
2:48 p.m.
On 6/2/2006 at 1:08 PM Roy J. Tellason wrote:
Is that stuff pretty much totally gone, then? My last
ECG book (not
recent)
still listed it, I've asked NTE to send me some literature but it hasn't
arrived yet and I don't know if they support it as well, or if it's not
included in their lineup...
All gone. If you run some MHTL part numbers, say, MC660* by the NTE
cross-reference at:
http://www.nteinc.com/
they all come up with "NTEOBS-NLA"; i.e. NTE obsolete, no longer available.
I wonder if 4000 series CMOS might provide some of the functionality at
slower speeds...
Cheers,
Chuck
2:54 p.m.
On Friday 02 June 2006 05:48 pm, Chuck Guzis wrote:
On 6/2/2006 at 1:08 PM Roy J. Tellason wrote:
Bummer. But then, I don't remember the last time I ran into any of that
stuff.
Is that stuff pretty much totally gone, then? My
last ECG book (not
recent)
still listed it, I've asked NTE to send me some literature but it hasn't
arrived yet and I don't know if they support it as well, or if it's not
included in their lineup...
All gone. If you run some MHTL part numbers, say, MC660* by the NTE
cross-reference at:
http://www.nteinc.com/
they all come up with "NTEOBS-NLA"; i.e. NTE obsolete, no longer available.
I wonder if 4000 series CMOS might provide some of
the functionality at
slower speeds...
I don't get the impression that speed is going to be a major issue there,
it's more likely that drive current capabiilty or some other aspect of it is
going to be a problem.
--
Member of the toughest, meanest, deadliest, most unrelenting -- and
ablest -- form of life in this section of space, a critter that can
be killed but can't be tamed. --Robert A. Heinlein, "The Puppet Masters"
-
Information is more dangerous than cannon to a society ruled by lies. --James
M Dakin
5:12 p.m.
Early 7400-series TTL had the 7450 and 7453
AND-OR-INVERT gates that could
be explanded with another 4 inputs via the 7460 dual 4-input expander. I'd
hate to own equipment now that used some of those that required
replacement--they must be very difficult to find.
There were at least 2 different expanders, I forget the number of the
other one.
The HP98x0 machines use these gates as the input selectors to the
bit-serial ALU (and I think to the T register (memory data register)). I
did have a 74H52 fail in one 9810 (that's an AND-OR gate), fortunately,
not much of it is used in _that machine_ and I could dead-bug in a
74LS51. That wouldn't work in the 9830, or I assume the 9820 (not looked
at the hardware of that machine yet, next project) as those use rather
more input sections of the AND-OR gate.
-tony
5:40 p.m.
On Friday 02 June 2006 08:12 pm, Tony Duell wrote:
Have a look at
http://www.classiccmp.org/rtellason/by-generic-number.html
and search for the word "expander" with your browser, you'll find as many
as
I've been able to find mention of, and not just TTL either.
Early
7400-series TTL had the 7450 and 7453 AND-OR-INVERT gates that
could be explanded with another 4 inputs via the 7460 dual 4-input
expander. I'd hate to own equipment now that used some of those that
required replacement--they must be very difficult to find.
There were at least 2 different expanders, I forget the number of the
other one. The HP98x0 machines use these gates as the input
selectors to the
bit-serial ALU (and I think to the T register (memory data register)). I
did have a 74H52 fail in one 9810 (that's an AND-OR gate), fortunately,
not much of it is used in _that machine_ and I could dead-bug in a
74LS51. That wouldn't work in the 9830, or I assume the 9820 (not looked
at the hardware of that machine yet, next project) as those use rather
more input sections of the AND-OR gate.
Speaking of 74H stuff, I have a fair box of chips here that I pulled out of
the boards of a Burroughs B500 that was scrapped. Almost all of these are
"house-numbered". Does anybody know where there might be a list that crosses
these over? I found only one or two that had 74H numbers on them, and I
think in maybe one case it had both...
--
Member of the toughest, meanest, deadliest, most unrelenting -- and
ablest -- form of life in this section of space, a critter that can
be killed but can't be tamed. --Robert A. Heinlein, "The Puppet Masters"
-
Information is more dangerous than cannon to a society ruled by lies. --James
M Dakin
11:16 p.m.
On 6/3/2006 at 1:12 AM ard at p850ug1.demon.co.uk wrote:
There were at least 2 different expanders, I forget the
number of the
other one.
Well, for the 3000-series TTL, there was the 3130, a dual 4-input job, and
for 2000 series TTL, there was the 2002/2052 3-2-2-3 AND-OR-INVERT
expander, as well as the 2006/2056 dual 4-input. There were similar parts
for the 400/500 series TTL also.
Cheers,
Chuck
11:22 p.m.
On Saturday 03 June 2006 02:16 am, Chuck Guzis wrote:
On 6/3/2006 at 1:12 AM ard at p850ug1.demon.co.uk
wrote:
I've been wondering for a while now what the major use was of those
"AND-OR-Invert" gates, since the early TTL stuff I became familiar with
first had a bunch of them in the databook...
Anybody know?
--
Member of the toughest, meanest, deadliest, most unrelenting -- and
ablest -- form of life in this section of space, a critter that can
be killed but can't be tamed. --Robert A. Heinlein, "The Puppet Masters"
-
Information is more dangerous than cannon to a society ruled by lies. --James
M Dakin
There were at least 2 different expanders, I
forget the number of the
other one.
Well, for the 3000-series TTL, there was the 3130, a dual 4-input job, and
for 2000 series TTL, there was the 2002/2052 3-2-2-3 AND-OR-INVERT
expander, as well as the 2006/2056 dual 4-input. There were similar parts
for the 400/500 series TTL also. 
3 Jun
3 Jun
12:31 a.m.
New subject: AND-OR-Invert, was Re: Semi-OT: IDE & SATA to USB "dongles"
Roy J. Tellason wrote:
I've been wondering for a while now what the major
use was of those
"AND-OR-Invert" gates, since the early TTL stuff I became familiar with
first had a bunch of them in the databook...
Anybody know?
This particular configuration of gates comes up frequently in digital
logic, and combining gates in this way provided more gates with fewer
packages. The 7451, using 2-input AND and OR gates, was often used as
a dual 2-to-1 MUX with unencoded select inputs. The inversion of the
output was not so much a feature as consequence of the fact that an
inverted output was easier to provide, e.g., NAND and NOR are more
commonly used than AND and OR for this reason. NAND and NOR are
also in some sense more universal -- you can compute any boolean
function with sufficient numbers of either NAND or NOR gates alone, but
not with AND and OR, which may require the use of inverters as well.
--Bill
1:07 a.m.
New subject: AND-OR-Invert, was Re: Semi-OT: IDE & SATA to USB "dongles"
William Maddox wrote:
Roy J. Tellason wrote:
Indeed. In the days before PAL's (later renamed PLA's), GAL's,
FPGA's and other ASICs, implementing "junk logic" used up lots
of DIPs really quick. Consider that two device pins are required
to connect an output from one gate to an input of another...
The AOI's facilitate sum-of-product implementations of arbitrary
combinatorial functions. And, they are often faster (end-to-end)
than gluing together discrete gates.
Using AOI's was a bit of an acquired taste -- they are their own
little specialized logic family, of sorts. E.g., "expanders"
presented different signal characteristics than the regular logic
inputs on the same device.
I've been wondering for a while now what the
major use was of those
"AND-OR-Invert" gates, since the early TTL stuff I became familiar
with first had a bunch of them in the databook...
Anybody know?
This particular configuration of gates comes up frequently in digital
logic, and combining gates in this way provided more gates with fewer packages. The 7451, using 2-input AND and OR gates,
was often used as
a dual 2-to-1 MUX with unencoded select inputs. The inversion of the
output was not so much a feature as consequence of the fact that an
inverted output was easier to provide, e.g., NAND and NOR are more
commonly used than AND and OR for this reason. NAND and NOR are
also in some sense more universal -- you can compute any boolean
function with sufficient numbers of either NAND or NOR gates alone, but
not with AND and OR, which may require the use of inverters as well.
4:50 a.m.
New subject: AND-OR-Invert, was Re: Semi-OT: IDE & SATA to USB "dongles"
On Saturday 03 June 2006 03:31 am, William Maddox wrote:
Roy J. Tellason wrote:
Now that you mention it, that early TTL book didn't have any AND or OR gates
in it, either...
--
Member of the toughest, meanest, deadliest, most unrelenting -- and
ablest -- form of life in this section of space, a critter that can
be killed but can't be tamed. --Robert A. Heinlein, "The Puppet Masters"
-
Information is more dangerous than cannon to a society ruled by lies. --James
M Dakin
I've been wondering for a while now what the
major use was of those
"AND-OR-Invert" gates, since the early TTL stuff I became familiar with
first had a bunch of them in the databook...
Anybody know?
This particular configuration of gates comes up frequently in digital
logic, and combining gates in this way provided more gates with fewer
packages. The 7451, using 2-input AND and OR gates, was often used as
a dual 2-to-1 MUX with unencoded select inputs. The inversion of the
output was not so much a feature as consequence of the fact that an
inverted output was easier to provide, e.g., NAND and NOR are more
commonly used than AND and OR for this reason. NAND and NOR are
also in some sense more universal -- you can compute any boolean
function with sufficient numbers of either NAND or NOR gates alone, but
not with AND and OR, which may require the use of inverters as well.
7:45 a.m.
New subject: AND-OR-Invert, was Re: Semi-OT: IDE & SATA to USB "dongles"
Hi
One thing that one should consider is that just tying descrete TTL's
is not the same as how it would be done on silicon. Combining nands
together would create a AND OR structure with no inversion.
When doing it on silicon, it is easier to move the and and or functions
to the inputs but then to drive the output requires the inversion.
As is mentioned this creates the AND OR invert structure that
would be faster than combining simple nand gates together making
the non-inverting AND OR.
These were, as mentioned, used as multiplexers. These were also
better to have inverted outputs for the DTL and RTL type setups.
The outputs could be wired together from several devices to drive
a single bus line. As long as the selects were all zero, on one device,
the output would be high, the undriven state of the output, the
other device could then drive the bus. As I recall, some of the TTL
devices had open collector as well and could be used this way.
Dwight
From: William Maddox <wmaddox at pacbell.net>
Reply-To: "General Discussion: On-Topic and Off-Topic
Posts"<cctalk at classiccmp.org>
To: General Discussion: On-Topic and Off-Topic Posts
<cctalk at classiccmp.org>
Subject: AND-OR-Invert, was Re: Semi-OT: IDE & SATA to USB "dongles"
Date: Sat, 03 Jun 2006 00:31:37 -0700
Roy J. Tellason wrote:
I've been wondering for a while now what the
major use was of those
"AND-OR-Invert" gates, since the early TTL stuff I became familiar with
first had a bunch of them in the databook...
Anybody know?
This particular configuration of gates comes up frequently in digital
logic, and combining gates in this way provided more gates with fewer
packages. The 7451, using 2-input AND and OR gates, was often used as a
dual 2-to-1 MUX with unencoded select inputs. The inversion of the output
was not so much a feature as consequence of the fact that an inverted
output was easier to provide, e.g., NAND and NOR are more commonly used
than AND and OR for this reason. NAND and NOR are
also in some sense more universal -- you can compute any boolean function
with sufficient numbers of either NAND or NOR gates alone, but not with AND
and OR, which may require the use of inverters as well.
--Bill
8:16 a.m.
New subject: AND-OR-Invert, was Re: Semi-OT: IDE & SATA to USB "dongles"
dwight elvey wrote:
Hi
One thing that one should consider is that just tying descrete TTL's
is not the same as how it would be done on silicon. Combining nands
together would create a AND OR structure with no inversion.
Yes, but you can cascade a NOR following an AND. A NOR would be
preferred (over OR) as ORs are slow.
When doing it on silicon, it is easier to move the and
and or functions
to the inputs but then to drive the output requires the inversion.
As is mentioned this creates the AND OR invert structure that
would be faster than combining simple nand gates together making
the non-inverting AND OR.
These were, as mentioned, used as multiplexers. These were also
better to have inverted outputs for the DTL and RTL type setups.
The outputs could be wired together from several devices to drive
a single bus line. As long as the selects were all zero, on one device,
the output would be high, the undriven state of the output, the
other device could then drive the bus. As I recall, some of the TTL
devices had open collector as well and could be used this way.
That depends on the AOI. Some had totem pole outputs.
Others were OC. And, not all were "expandable".
Note that the *expanders* could be used for wired AND *or*
wired OR logic as they typically had uncommitted collectors
*and* emitters on the output stage.
9:20 a.m.
New subject: AND-OR-Invert, was Re: Semi-OT: IDE & SATA to USB "dongles"
One of the things that I found highly instructive when starting with
integrated circuit logic were the schematics of the logic found in the old
databooks. Not done any longer (probably for practical reasons), but they
were sure instructive as to showing how a given logic function was
accomplished. Often it wasn't what you expected.
Cheers,
Chuck
3:35 p.m.
I've been wondering for a while now what the major
use was of those
"AND-OR-Invert" gates, since the early TTL stuff I became familiar with
first had a bunch of them in the databook...
Anybody know?
A couple of thoughts
1) It's fundamentally a sum-of-products circuit (OR of ANDs). This is one
standard way to genrate an arbitrary logic function
2) It's also the basic circuit of a multiplexer. If you want a mux where
you can enable several inputs simulaneously, or where the select lines
can't easily be turned into n binary-encoded lines, then an
And-OR-(invert) gate is esseintally what you're going to use.
3) Rememebr a PAL is fundamentally a set of AND-OR-(invert) gates with
programmable input wiring. Before there were PALs you got to use the AOIs
and wire the inputs by hand :-)
-tony
7:35 p.m.
Anybody know?
As long as were talkin'...
I am not home right now - going thru a pile of stuff I have stored
elsewhere - and came across some logic I am not familiar with. MC5xx from
Motorola ring a bell? This is the oldest logic IC system I have
personally seen, as the chips have 1965 datecodes (!). They are part of a
Doppler processor for the obscuro AN/KPQ-1 radar set.
Being not home, I can not look this up, and it is starting to bug me.
William Donzelli
aw288 at osfn.org
8:05 p.m.
On Saturday 03 June 2006 10:35 pm, William Donzelli wrote:
The oldest Motorola book I have on hand here is their 1971 Master Selection
Guide, and those numbers appear on page 122, under "MTTL I". If you'd
care
to call out specific numbers I can tell you what they are.
--
Member of the toughest, meanest, deadliest, most unrelenting -- and
ablest -- form of life in this section of space, a critter that can
be killed but can't be tamed. --Robert A. Heinlein, "The Puppet Masters"
-
Information is more dangerous than cannon to a society ruled by lies. --James
M Dakin
Anybody know?
As long as were talkin'...
I am not home right now - going thru a pile of stuff I have stored
elsewhere - and came across some logic I am not familiar with. MC5xx from
Motorola ring a bell? This is the oldest logic IC system I have
personally seen, as the chips have 1965 datecodes (!). They are part of a
Doppler processor for the obscuro AN/KPQ-1 radar set.
9:57 p.m.
> Anybody know?
The AND-OR-INVERT is a very useful function. Think about it--you have not
only a good hunk of a multiplexer, but also most of a XOR gate as well as
a half-adder. It took awhile to get integration up to the point where full
adders (much less ALUs) were available on a single chip.
Cheers,
Chuck
4 Jun
4 Jun
8:31 a.m.
On Sat, 3 Jun 2006, Chuck Guzis wrote:
If I'm not mistaken the sort-of-processor-from-TTL-bits the Diablo HyType II
had used a pair of AOI gates for the logical part of its ALU (along with a
pair of 74LS283s)
Peter Wallace
Mesa Electronics
> Anybody
know?
The AND-OR-INVERT is a very useful function. Think about it--you have not
only a good hunk of a multiplexer, but also most of a XOR gate as well as
a half-adder. It took awhile to get integration up to the point where full
adders (much less ALUs) were available on a single chip.
Cheers,
Chuck
8:42 a.m.
From: "Chuck Guzis" <cclist at
sydex.com>
Hi
Of course, I think is is interesting that in CMOS, one often uses
analog transmission gates to create XOR functions. Analog transmission
gates are usually used for MUX's.
Dwight
> Anybody know?
The AND-OR-INVERT is a very useful function. Think about it--you have not
only a good hunk of a multiplexer, but also most of a XOR gate as well as
a half-adder. It took awhile to get integration up to the point where full
adders (much less ALUs) were available on a single chip.
Cheers,
Chuck
2:57 a.m.
On 6/2/06, Roy J. Tellason <rtellason at blazenet.net> wrote:
I always wondered what DTL was good for...
Well.. for one thing, you can make some of those electronics toys and
games from "Radio Electronics", et al., c. 1973. Some of the projects
have been scanned and are floating around on the 'net, but I don't
think they'll work with TTL parts (without a redesign).
-ethan
9:41 a.m.
On Friday 02 June 2006 05:57 am, Ethan Dicks wrote:
On 6/2/06, Roy J. Tellason <rtellason at
blazenet.net> wrote:
I thought some of those used RTL, which _is_ a very different beast. I
remember building some of that stuff...
--
Member of the toughest, meanest, deadliest, most unrelenting -- and
ablest -- form of life in this section of space, a critter that can
be killed but can't be tamed. --Robert A. Heinlein, "The Puppet Masters"
-
Information is more dangerous than cannon to a society ruled by lies. --James
M Dakin
I always wondered what DTL was good for...
Well.. for one thing, you can make some of those electronics toys and
games from "Radio Electronics", et al., c. 1973. Some of the projects
have been scanned and are floating around on the 'net, but I don't
think they'll work with TTL parts (without a redesign). 
30 May
30 May
5:34 p.m.
"Chuck Guzis" wrote:
If one wanted to make an ST-506 or 412 replacement, I'd be sore tempted
just to do it with some battery-backed SRAM and a microcontroller.
You know, some friends of mine once did that for the original macintosh.
I just remembered. They figured out the signals coming out of the
floppy port in the back and faked the signals in software. With a z-80
if I recall correctly. I think it had a big (for the time) battery
backed up sram bank. It was probably all of 1-2mb. I'm guessing this
was essentially talking to the IWM (and doing group code)
at the time I was working on the Hyperdrive and we had *10mb*, woo hoo.
hmmm... I wonder if either of them saved any of that. I'll ask.
those two guys later founded Shiva
(and I got was a t-shirt! :-)
-brad
2:49 p.m.
On 5/30/06, Billy Pettit <Billy.Pettit at wdc.com> wrote:
Some time, we have to let go and upgrade as much as it
is unaesthetic or
displeasing. When?
There comes a time when it's no longer viable to keep old machines
running. Computers aren't anything unique in this regard.
The oldest human weapon that wasn't simply a thrown object is probably
a sling. There's at least one nearly 3,000yr old one that was found in
Egypt, but it's far too fragile to use or test - but for trying it
out, it can be reconstructed:
http://www.petrie.ucl.ac.uk/detail/details/index_no_login.php?objectid=UC69…
After a while, when parts and consumables can no longer be found or
fabricated, old computers, like, say, old motorbikes, are going to
become purely museum pieces, that cannot actually be *run*. They're
already no use - you wouldn't go to work on a 1905 Douglas and you
wouldn't run a PDP-8 to read your email. Doesn't mean they're
valueless. Better to preserve them in their original state in a museum
than to bodge together some hybrid of ancient and modern kit.
To give people the experience of using them, write and emulator & make
it perfect & faithful. That way,m the OS, the software, the data
files, the *feel* of the machine can be kept alive indefinitely.
One day, future computers won't run our current "modern" software. But
we can run the emulator on an emulator when that happens!
--
Liam Proven ? Blog, homepage &c: http://lproven.livejournal.com/profile
Email: lproven at cix.co.uk ? GMail/Google Talk/Orkut: lproven at gmail.com
AOL/AIM/iChat: liamproven at aol.com ? MSN/Messenger: lproven at hotmail.com
Yahoo: liamproven at yahoo.co.uk ? Skype: liamproven ? ICQ: 73187508
3:10 p.m.
After a while, when parts and consumables can no
longer be found or
fabricated, old computers, like, say, old motorbikes, are going to
become purely museum pieces, that cannot actually be *run*. They're
already no use - you wouldn't go to work on a 1905 Douglas and you
wouldn't run a PDP-8 to read your email. Doesn't mean they're
valueless. Better to preserve them in their original state in a museum
than to bodge together some hybrid of ancient and modern kit.
To give people the experience of using them, write and emulator & make
it perfect & faithful. That way,m the OS, the software, the data
files, the *feel* of the machine can be kept alive indefinitely.
I was giving this some thought just this morning. At least some of my
systems will be going into the "Historical Resource Center" that I mentioned
in the Library thread. I've always wanted to have these systems where they
can be seen and used by the public, but then I got to thinking. Do I really
want to maintain something like a PDP-11 that is on 24/7 so the public can
play with it? It makes far more sense to have such systems on display, and
just emulate them. Sure people like us will know the difference, and mind
it, but realistically to the general public, does it really matter? At
least to start with, I'll be the only technically oriented person involved,
and I simply do not have the time to maintain old systems for anyone other
than myself (don't really have time to do it for myself).
At the same time, I'm looking into a setup such that the general public can
go into the "Historical Resource Center" and transfer their old media to
modern media. I invision this including 5 1/4" floppies, and 3 1/2"
floppies at a minimum. (I figure at least Commodore, Apple and PC). While
this can be done with a combination of old and new systems, it might be better
to do it with a modern PC and a Catweasel card.
Zane
3:12 p.m.
On Tuesday 30 May 2006 17:49, Liam Proven wrote:
On 5/30/06, Billy Pettit <Billy.Pettit at
wdc.com> wrote:
Or, we can keep the source around[0], and write things in fairly
portable languages, and modify them as necessary to run on each
successive new technology[1], where necessary.
0. That implies that we have the source by one way or another, of
course.
1. One might conclude then that open-source software has a longer
potential life-span than close-source things do.
Pat
--
Purdue University Research Computing --- http://www.rcac.purdue.edu/
The Computer Refuge --- http://computer-refuge.org
Some time, we have to let go and upgrade as much
as it is
unaesthetic or displeasing. When?
[...]
One day, future computers won't run our current "modern" software.
But we can run the emulator on an emulator when that happens!
3:28 p.m.
After a while, when parts and consumables can no
longer be found or
fabricated, old computers, like, say, old motorbikes, are going to
become purely museum pieces, that cannot actually be *run*. They're
already no use - you wouldn't go to work on a 1905 Douglas and you
wouldn't run a PDP-8 to read your email. Doesn't mean they're
Why not? I would think a PDP8 would be a lot easier to keep running than
a modern PC. You are speaking to somebody who is reading/writing mail on
an old PC mainly because it is easier to maintain than a more modern machine.
valueless. Better to preserve them in their original
state in a museum
than to bodge together some hybrid of ancient and modern kit.
No, I disagree totally. I have no problem at all with making reverseable
modificuations to old machines (be they computers or anything else). I
certainly see no problem in plugging in a modern peripheral.
Are you trying to convince me that I shouldn't use 74Fxxx parts to repair
a PDP11 (they are much easier to get than 74Sxxx and work in most
circuits). You'll really hate what I've done to a lot of machines here, then
To give people the experience of using them, write and emulator & make
it perfect & faithful. That way,m the OS, the software, the data
So far, no emulator has come anywhere near the experience of running the
real machine. And I don't think it ever will for me. I want to be able to
connect my 'scope (a real Tekky 555, please) to any pin of any component
in the machine. And have all the feelings associated with so doing.
files, the *feel* of the machine can be kept alive
indefinitely.
-tony
4:43 p.m.
There comes a time when it's no longer viable to
keep
old machines
running.
Like my pre-1902 Pratt & Whitney precision bench
lathe? I advise you NOT to utter that sort of language
around a fanatical gear-head. He/she just might have a
triangular hand scraper in their back pocket. Needless
to say it's it has a keen edge!
Computers aren't anything unique in this regard.
The oldest human weapon that wasn't simply a
thrown
object is probably
a sling. There's at least one nearly 3,000yr old one
that was found in
Egypt, but it's far too fragile to use or test - but
for trying it
out, it can be reconstructed:
Point taken, but computers don't fall into the same
category as rare ancient paraphernalia. Granted some
of the real old stuph should be preserved in as
original condition as possible, but half (or a much
higher ration!) of the thrill of having these things
around is the ability to fire them up. If there comes
a point where this can't be done other then by
substituting modern material, then I'm all for it.
After a while, when parts and consumables can no
longer be found or
fabricated, old computers, like, say, old motorbikes,
are going to
become purely museum pieces, that cannot actually be
*run*. They're
already no use - you wouldn't go to work on a 1905
Douglas and you
wouldn't run a PDP-8 to read your email. Doesn't mean
they're
valueless. Better to preserve them in their original
state in a museum
than to bodge together some hybrid of ancient and
modern kit.
Ehh I don't know. I would have to say it's a rare
instance where parts for machinery and whatnot can't
be reproduced. Regarding low-technologies, some have
gone as far as to build backyard cupola furnaces for
melting cast iron. Or natural draft units for melting
bronze/brass, ala Theophilus. Microchips are more
challenging granted. But I doubt the day is very far
off when fabricating something as intricate as a
microchip would be possible on your kitchen table.
To give people the experience of using them, write
and emulator & make
it perfect & faithful. That way,m the OS, the
software, the data
files, the *feel* of the machine can be kept alive
indefinitely.
One day, future computers won't run our >current
"modern" software. But
we can run the emulator on an emulator when that
happens!
I have nothing against emulators. But they're only
one side of this vintage groove - the software side.
And my hat goes off to those who have written
emulators for classic stuph. But the investigation,
maintenance, and when practical the reproduction of
classic hardware is engaging persuit in itself.
__________________________________________________
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.yahoo.com
31 May
31 May
3:46 p.m.
There comes a time when it's no longer viable
to keep
old machines
running.
Like my pre-1902 Pratt & Whitney precision bench
lathe? I advise you NOT to utter that sort of language Point taken, but computers don't fall into the
same
category as rare ancient paraphernalia. Granted some
In an earlier message, old computers were compared to classic
cars/motorbikes. I can think of one very important difference here. A
motor vehicle, could, alas, be involved in a collision, which would
totally ruin it, both for use and as a museum exhibit. But it's rare for
a component failure in a computer to make it useless as a static exhibit.
An example. One day the BPC hybrid processor module in my 9825 is going
to fail. It's a custom part that I can't easily recreate, and can't
repair. When that happens, if I can't find a replacement, the machine is
no longer going to be useful as a computer. But it certainly can be used
as a static exhibit. And while it's still useable, I see no harm in
using it. If I never used it, the processor module could still fail, just
sitting on the shelf.
-tony
6 a.m.
On 5/30/06, Tony Duell <ard at p850ug1.demon.co.uk> wrote:
Why not? I would think a PDP8 would be a lot easier to
keep running than
a modern PC.
Only to somebody with what is rapidly becoming a /very/ specialist skillset.
You are speaking to somebody who is reading/writing
mail on
an old PC mainly because it is easier to maintain than a more modern machine.
Fine. But for most people, modern PCs are a damn sight /easier./ All
in one integrated motherboards, simple mix&match DIMMs,
interchangeable commodity disks and screens and keyboards, simple USB
peripherals. The day of the ?100 PC is just around the corner. Soon
they'll be as disposable as transistor radios.
Q.v. http://www.indybay.org/news/2006/05/1824094.php
No, I disagree totally. I have no problem at all with
making reverseable
modificuations to old machines (be they computers or anything else). I
certainly see no problem in plugging in a modern peripheral.
Are you trying to convince me that I shouldn't use 74Fxxx parts to repair
a PDP11 (they are much easier to get than 74Sxxx and work in most
circuits). You'll really hate what I've done to a lot of machines here, then
I don't even know what a 74Fxxx is. I've been repairing and working on
PCs and Macs for 20y now, I've built dozens and dozens of computers,
and I don't own a multimeter or an oscilloscope and never have. I have
a soldering iron; I use it occasionally to fix the lights on my
bicycle.
This is great stuff if you know it, but it's no more relevant to C21
computing than the correct bloodletting and the correct administration
of laudanum is to C21 medicine. It's an admirable skill and I salute
you, but it's not mainstream any more. I'm not saying it's not
worthwhile; I'm just saying that things have moved on. If a circuit
board fails in a modern machine, recycle it & bung in a new one.
For many people, it's already a case of "if the PC fails, or just gets
a bit slow, bin it & get a new one." My front-room Linux box came out
of a skip in Brighton; a perfectly-working Cyrix 550MHz PC with 128M
RAM, a 40G HD and a CD burner. Now it's a 1.2GHz Duron with 256M RAM,
2 x 10G HDs and Ubuntu, and it was all built out of scrap. The 40G
disk is part of the RAID in my fileserver, a dual PII 400, all from
free junk kit.
So far, no emulator has come anywhere near the
experience of running the
real machine. And I don't think it ever will for me. I want to be able to
connect my 'scope (a real Tekky 555, please) to any pin of any component
in the machine. And have all the feelings associated with so doing.
Yeah, it's not quite the same, but you do realise, we're talking of
something of the level of a glass case in a museum, with a few shiny
buttons and lights for people to play with for 5min? They will neither
know nor care.
Secondly, as far as preserving old kit for posterity, your
hacked-about with machines will confuse other techies, who would
probably have more chance of getting the original unmodified kit they
know going than something with home-made one-off upgrades. In the
longer run, after you are no longer in a position to maintain it, the
stock kit has better prospects.
--
Liam Proven ? Blog, homepage &c: http://lproven.livejournal.com/profile
Email: lproven at cix.co.uk ? GMail/Google Talk/Orkut: lproven at gmail.com
AOL/AIM/iChat: liamproven at aol.com ? MSN/Messenger: lproven at hotmail.com
Yahoo: liamproven at yahoo.co.uk ? Skype: liamproven ? ICQ: 73187508
8:44 a.m.
On 5/31/2006 at 1:00 PM Liam Proven wrote:
Yeah, it's not quite the same, but you do realise,
we're talking of
something of the level of a glass case in a museum, with a few shiny
buttons and lights for people to play with for 5min? They will neither
know nor care.
I suspect that part of what's going on here is the personal association
with an old piece of hardware and wanting to keep the memory of the
experience alive. (If that makes any sense).
My father once wanted to acquire an old receiver because he remembered
tinkering with A and B batteries, variocouiplers, book variable capacitors
and other bits of early electronics when he was young. He eventually found
an old Atwater-Kent, but soon tired of it when he realized that what wasn't
there was the old radio programming. Old receivers can make interesting
decorations, but modern gear does the job better and cheaper.
I still have a couple of old systems that I was part of the development
effort on back in the 1970's. They work, but, after all, they're pretty
limited in capability and I've tried to give them away. I'll probably hang
onto them for a couple of more years and then, if there are no takers,
break them down and recycle their parts. After all, what's the difference
between machines that run CP/M 2.2? (These run a proprietary operationg
system, but compared to NetBSD, it's painfully primitive). Besides, I have
an emulator that seamlessly integrates most of the useful CP/M programs
into Windows 2000, if I have such a need.
While I have some pleasant memories of the experiences, I have no desire to
revisit the days of big mainframes. Compared to today's hardware, they
were big, slow, noisy and terribly unreliable.
While I think that it's worthwhile keeping samples of the old stuff around
for museums as well as preserving the ephemera and media content for
historical purposes, keeping old hardware alive for personal collections
has a purpose that eludes my understanding.
I have a small collection of old musical instruments, but one can still
play a 100 year old cornet or tuba and it compares favorably with items of
modern manufacture.
Cheers,
Chuck
11:16 a.m.
Chuck Guzis wrote:
On 5/31/2006 at 1:00 PM Liam Proven wrote:
There is also a segment that is interested in playing with the software
running on real hardware. This, and an interest in preserving the systems
Yeah, it's not quite the same, but you do
realise, we're talking of
something of the level of a glass case in a museum, with a few shiny
buttons and lights for people to play with for 5min? They will neither
know nor care.
I suspect that part of what's going on here is the personal association
with an old piece of hardware and wanting to keep the memory of the
experience alive. (If that makes any sense). from a historical point of view, would describe me.
I have fond memories of my VIC-20 and Kaypro 2000 (an early laptop), and
wish I still had both systems (espiecially the kit built pieces of the
VIC-20), but realisically, I have no desire to touch them. Either on a
hardware or software level. As much as I hated the C-64 at the time
(because it obsoleted my VIC-20 a few months after I got it, the C-64 is a
much more interesting system).
I am interested in the C-64 and Apple ][ computers, and in keeping them
running, I had neither when they were current, and only got to touch one a
couple times back then. I'm actually looking at the devices that let you
use MMC cards with interest (I know they exist for the C-64, and assume they
exist for the Apple ][). The same goes for the PDP-11 to a certain extent,
however, I've moved to SCSI drives on both my Unibus system, and my Q-Bus
systems. In the case of the PDP-11's though I'm more inclined to use an
emulator, and have been leaning towards prefering the emulator for several
years.
While I have some pleasant memories of the
experiences, I have no desire to
revisit the days of big mainframes. Compared to today's hardware, they
were big, slow, noisy and terribly unreliable.
I don't care to revisit the days of the big mainframes, however, I would
like to be able to tinker around on an emulator with GCOS-8, and I do enjoy
playing with TOPS-10 and TOPS-20. I neither have the room, or desire to try
to keep the hardware needed to run these systems going, nor can I afford the
electricity.
While I think that it's worthwhile keeping samples
of the old stuff around
for museums as well as preserving the ephemera and media content for
historical purposes,
This is where I'm starting to become actively involved. In part because
I've spent a lot of my own money preserving and storing this stuff, that
would be better off in a museum. Parts of my collection have started to
move to where they can be retained, and as well as possible preserved for
future generations. While this isn't saving any money, it has already
started to allow us to have more room in our apartment.
keeping old hardware alive for personal collections
has a purpose that eludes my understanding.
Basically it's the I/O devices. Seriously. Other than the I/O devices
you're better off emulating where possible on modern hardware.
Zane
11:40 a.m.
And thusly were the wise words spake by Zane H. Healy
I am interested in the C-64 and Apple ][ computers, and in keeping them
running, I had neither when they were current, and only got to touch one a
couple times back then. I'm actually looking at the devices that let you
use MMC cards with interest (I know they exist for the C-64, and assume they
exist for the Apple ][). The same goes for the PDP-11 to a certain extent,
however, I've moved to SCSI drives on both my Unibus system, and my Q-Bus
systems. In the case of the PDP-11's though I'm more inclined to use an
emulator, and have been leaning towards prefering the emulator for several
years.
Zane,
There is the IDE64 which allows you to hookup IDE drive,
CD/DVD-ROMS and Compact Flash cards. Plus the creator is working
on an ethernet adapter that would attach to it. Right now the IDE64
board is in its fourth revision.
There is also the Retro Replay cartridge (a program
freezer) which has an addon called the RR-Net that gives you a
10Mbps (yes I know the C64 can't move that much data!) ;) ethernet port.
This is supported by the Contiki OS which is set to move to V2.0.
Cheers,
Bryan
5:28 p.m.
New subject: C64 upgrades (was Re: Semi-OT: IDE & SATA to USB "dongles")
There is the IDE64 which allows you to hookup IDE
drive,
CD/DVD-ROMS and Compact Flash cards. Plus the creator is working
on an ethernet adapter that would attach to it. Right now the IDE64
board is in its fourth revision.
Now that is interesting, is this a board you can buy somehow, or do you have
to build it yourself?
While trying to figure that one out, I ran across a *VERY* interesting
website: http://www.64hdd.com/64hdd.html This is for a piece of software
that turns an MS-DOS PC into a Commodore Disk drive, plus it can handle D64
files. I think this is the route I might end up going.
There is also the Retro Replay cartridge (a program
freezer) which has an addon called the RR-Net that gives you a
10Mbps (yes I know the C64 can't move that much data!) ;) ethernet port.
This is supported by the Contiki OS which is set to move to V2.0.
Cheers,
Bryan
I must confess I'm not totally clear on why I would want this. The Retro
Replay cartridge looks cool, but what does the RR-Net addon get me? Can I
use it with Quantum Link Reloaded? It looks like it can probably be used to
make disk images of existing floppies, or write disk images back out to
floppies, but I'm honestly not sure on that one.
Zane
6:06 p.m.
New subject: C64 upgrades (was Re: Semi-OT: IDE & SATA to USB "dongles")
And thusly were the wise words spake by Zane H. Healy
No, it comes pre-built. :) GOTO http://www.ide64.org for the main
web page. http://wings.webhop.org (Mordernity Group) is a distributor
of the IDE64, but they currently do not have any as they are waiting
for V4 to be completed. I bought my IDE64 from Greg Nacu of MG at
the 2004 LUCKI Expo.
I don't think so... I believe it only works with Contiki at this
time.
Cheers,
Bryan
There is the IDE64 which allows you to hookup
IDE drive,
CD/DVD-ROMS and Compact Flash cards. Plus the creator is working
on an ethernet adapter that would attach to it. Right now the IDE64
board is in its fourth revision.
Now that is interesting, is this a board you can buy somehow, or do you have
to build it yourself?
While trying to figure that one out, I ran across a
*VERY* interesting
website: http://www.64hdd.com/64hdd.html This is for a piece of software
that turns an MS-DOS PC into a Commodore Disk drive, plus it can handle D64
files. I think this is the route I might end up going.
This requires the XE1541 cable which goes from the parallel port on
your PC to the IEC connector on the C64. But you need an older PC
as the newer PCs parallel port will not work properly. There is
also the new XM1541 ( http://sta.c64.org/xm1541.html ) which is
supposed to work with most parallel ports and PC boards. The is also
a link there for The Star Commander which handles quite a few different
file formats (and this cable!) :)
There is also
the Retro Replay cartridge (a program
freezer) which has an addon called the RR-Net that gives you a
10Mbps (yes I know the C64 can't move that much data!) ;) ethernet port.
This is supported by the Contiki OS which is set to move to V2.0.
Cheers,
Bryan
I must confess I'm not totally clear on why I would want this. The Retro
Replay cartridge looks cool, but what does the RR-Net addon get me? Can I
use it with Quantum Link Reloaded? It looks like it can probably be used to
make disk images of existing floppies, or write disk images back out to
floppies, but I'm honestly not sure on that one.
8 p.m.
New subject: C64 upgrades (was Re: Semi-OT: IDE & SATA to USB "dongles")
Bryan Pope wrote:
And thusly were the wise words spake by Zane H. Healy
This requires the XE1541 cable which goes from the parallel port on
your PC to the IEC connector on the C64. But you need an older PC
as the newer PCs parallel port will not work properly. There is
also the new XM1541 ( http://sta.c64.org/xm1541.html ) which is
supposed to work with most parallel ports and PC boards. The is also
a link there for The Star Commander which handles quite a few different
file formats (and this cable!) :)
I've fiddled with a homebrew XM1541 a bit. It does OK with the
multi-mode parallel ports, but I found that it's most reliable on
sub-200MHz PCs. I remember that one of the software packages worked
really well and one just pissed me off a lot, but I don't remember
which. :)
Doc
8:29 p.m.
New subject: C64 upgrades (was Re: Semi-OT: IDE & SATA to USB "dongles")
I've had pretty good luck with Star Commander on a 200 MHz Pendium.
Instead of making up a cable, I just nibbled the a appropriate hole in the
back of my 1571 and installed a female DB-25 connector with all of the
applicable parts on it. I use a straight male-male DB25 to hook the 1571
to the parallel port.
FWIW, I used some generic junkbox Schottky diodes that I had kicking
around--they worked fine.
Cheers,
Chuck
10:37 p.m.
New subject: C64 upgrades (was Re: Semi-OT: IDE & SATA to USB "dongles")
At 9:06 PM -0400 5/31/06, Bryan Pope wrote:
And thusly were the wise words spake by Zane H. Healy
I'm wondering if one of my Dauphin DTR-1 handhelds can be used
(basically a 486 running DOS). I've got the external floppy, and
they have a parallel port. It might very well work with the X1541 I
built for my 486 years ago. If not I might just order an XE1541
cable from c64reloaded.
While trying
to figure that one out, I ran across a *VERY* interesting
website: http://www.64hdd.com/64hdd.html This is for a piece of software
that turns an MS-DOS PC into a Commodore Disk drive, plus it can handle D64
files. I think this is the route I might end up going.
This requires the XE1541 cable which goes from the parallel port on
your PC to the IEC connector on the C64. But you need an older PC
as the newer PCs parallel port will not work properly. There is
also the new XM1541 ( http://sta.c64.org/xm1541.html ) which is
supposed to work with most parallel ports and PC boards. The is also
a link there for The Star Commander which handles quite a few different
file formats (and this cable!) :) > I must confess I'm not totally clear on why
I would want this. The Retro
> Replay cartridge looks cool, but what does the RR-Net addon get me? Can I
I don't think so... I believe it only works with
Contiki at this
time.
I hate to say it, but it gets back to the, why? If that's all it is
good for, I honestly can't see a purpose.
Zane
--
| Zane H. Healy | UNIX Systems Administrator |
| healyzh at aracnet.com (primary) | OpenVMS Enthusiast |
| MONK::HEALYZH (DECnet) | Classic Computer Collector |
+----------------------------------+----------------------------+
| Empire of the Petal Throne and Traveller Role Playing, |
| PDP-10 Emulation and Zane's Computer Museum. |
| http://www.aracnet.com/~healyzh/ |
11:39 p.m.
New subject: C64 upgrades (was Re: Semi-OT: IDE & SATA to USB "dongles")
On 5/31/2006 at 10:37 PM Zane H. Healy wrote:
I'm wondering if one of my Dauphin DTR-1 handhelds
can be used
(basically a 486 running DOS). I've got the external floppy, and
they have a parallel port. It might very well work with the X1541 I
built for my 486 years ago. If not I might just order an XE1541
cable from c64reloaded.
The X1541, while it'll probably work with most older systems (including
your DTR-1), is just lousy design. The 1541 serial interface is
bi-directional, and the X1541 relies on the 1541 to pull down an
output-only parallel port pin. While this works for LSTTL-driven parallel
ports, it's pretty hostile to later CMOS-driven ones.
The XE1541 uses some diodes to steer the signals to the proper pins whether
the 1541 serial interface is in input or output mode. It's pretty easy to
modify a X1541 to be a XE1541 and you can find "any idiot can do it"
instructions on the web.
Cheers,
Chuck
(been there)
9:45 p.m.
New subject: C64 upgrades (was Re: Semi-OT: IDE & SATA to USB "dongles")
On 6/1/06, Zane H. Healy <healyzh at aracnet.com> wrote:
I built a rev2 board as a kit years ago, but I think the more recent versions
are tending towards pre-assembled. I hung my 1.3" HP Kittyhawk off of
mine - works great. I'm contemplating buying a rev4 board when I get home.
http://www.ide64.org/
-ethan
There is
the IDE64 which allows you to hookup IDE drive,
CD/DVD-ROMS and Compact Flash cards... Right now the IDE64
board is in its fourth revision.
Now that is interesting, is this a board you can buy somehow, or do you have
to build it yourself?
1 Jun
1 Jun
4:10 p.m.
On 5/30/06, Tony Duell <ard at p850ug1.demon.co.uk> wrote:
And alas skills that nobody seems to want ;-(
Why not? I would think a PDP8 would be a lot
easier to keep running tha=
n
a modern PC.
Only to somebody with what is rapidly becoming a /very/ specialist skills=
et. You are speaking to somebody who is
reading/writing mail on
an old PC mainly because it is easier to maintain than a more modern ma=
chine.
Fine. But for most people, modern PCs are a damn sight /easier./ All in one integrated motherboards, simple mix&match
DIMMs,
interchangeable commodity disks and screens and keyboards, simple USB
peripherals. The day of the =A3100 PC is just around the corner. Soon
they'll be as disposable as transistor radios.
You are speaking to the person who _repairs_ transistor radios.
No, I disagree
totally. I have no problem at all with making reverseabl=
e
modificuations to old machines (be they computers
or anything else). I
certainly see no problem in plugging in a modern peripheral.
Are you trying to convince me that I shouldn't use 74Fxxx parts to repa=
ir
a PDP11 (they are much easier to get than 74Sxxx
and work in most
circuits). You'll really hate what I've done to a lot of machines here,=
then
I don't even know what a 74Fxxx is. I've been repairing and working on PCs and Macs for 20y now, I've built dozens and
dozens of computers,
No, you've _assembled_ dozens of computers. To me, making a computer
involves starting from chips, transistors, etc. And it's a bonus if you
don't use a ready-built microprocessor, so you can define your own
instruction set.
and I don't own a multimeter or an oscilloscope
and never have. I have
a soldering iron; I use it occasionally to fix the lights on my
bicycle.
This is great stuff if you know it, but it's no more relevant to C21
computing than the correct bloodletting and the correct administration
That maybe why I don't get invloved in C21 computing...
of laudanum is to C21 medicine. It's an admirable
skill and I salute
you, but it's not mainstream any more. I'm not saying it's not
worthwhile; I'm just saying that things have moved on. If a circuit
board fails in a modern machine, recycle it & bung in a new one.
It's quicker for me to repair many faults to component level than to
track down a replacement and hope there's not some nasty incompatibility
with some software I'm using.
So far, no emulator has come anywhere near the
experience of running th=
e
real machine. And I don't think it ever will
for me. I want to be able =
to
connect my 'scope (a real Tekky 555, please)
to any pin of any componen=
t
in the machine. And have all the feelings
associated with so doing.
Yeah, it's not quite the same, but you do realise, we're talking of
something of the level of a glass case in a museum, with a few shiny
buttons and lights for people to play with for 5min? They will neither
know nor care.
Secondly, as far as preserving old kit for posterity, your
hacked-about with machines will confuse other techies, who would
Only if said techies are somewhat brain-dead. For one thing I keep
records of all significant modifications (and the person who gets my
collection when I snuff it is clueful enough to look at all my manuals
too). And for another, ICs have date codes on them. If you came across a
machine, clearly from the 1970's, where all the ICs but one had 1976 date
codes, and that one had a 1990 date code, the only sensible conclusion is
that it's been replaced.
-tony
31 May
31 May
4:48 a.m.
There comes a time when it's no longer viable to
keep old machines
running. Computers aren't anything unique in this regard.
To be a wiseguy somewhat, I'd say it's summer.
Keep your more precious beasts off when ambient is
above 80 or certainly 90 degrees farenheit.
Personally, I don't think continuous duty at all for anything >10y-o.
Use commodity PC's for work not itself of historical significance.
John A.
1 Jun
1 Jun
5:44 p.m.
John Allain wrote:
Personally, I don't think continuous duty at all
for anything >10y-o.
I do my best work on my IBM 5160; it has gotten continuous use for
nearly two decades.
--
Jim Leonard (trixter at oldskool.org) http://www.oldskool.org/
Help our electronic games project: http://www.mobygames.com/
Or check out some trippy MindCandy at http://www.mindcandydvd.com/
A child borne of the home computer wars: http://trixter.wordpress.com/
6:41 p.m.
At 7:44 PM -0500 6/1/06, Jim Leonard wrote:
John Allain wrote:
My DEC PWS 433au is right at about 10 years and gets continuous use.
It runs 24x365, and has for the last 6 years. A lot of people have
VAXen that they been using for 10+ years (if not 20+), same with
PDP-11's. In the DEC world 10+ years isn't so much unusual, as the
norm. Some DEC systems have been in continuous use for 30+ years.
Zane
--
| Zane H. Healy | UNIX Systems Administrator |
| healyzh at aracnet.com (primary) | OpenVMS Enthusiast |
| MONK::HEALYZH (DECnet) | Classic Computer Collector |
+----------------------------------+----------------------------+
| Empire of the Petal Throne and Traveller Role Playing, |
| PDP-10 Emulation and Zane's Computer Museum. |
| http://www.aracnet.com/~healyzh/ |
Personally, I don't think continuous duty at
all for anything >10y-o.
I do my best work on my IBM 5160; it has gotten continuous use for
nearly two decades.
2 Jun
2 Jun
6:35 a.m.
>John Allain wrote:
>>Personally, I don't think continuous duty at all for anything >10y-o.
> I do my best work on my IBM 5160; it has gotten
continuous use for
> nearly two decades.
My DEC PWS 433au is right at about 10 years and gets
continuous use.
The idea is if it dies, and nobody can get another,
then you are the one who killed it.
There are counterarguments I guess, like that CRTs and electrolytics need at
least some periodic duty to keep them up, but for me that's actually about 1
week a year, hardly continuous.
John A.
30 May
30 May
3:13 p.m.
They will
become harder and harder to find and keep working. If you are
trying to keep the
old systems running, you will eventually have to do a
dongle of some sort with newer drives. Why not now?
Keeping the old drives working is going to require parts and technology =
that
don't exist commercially any more. You can still find the high volume circuitry in the surplus stores. But it's the
heads and platters that =
wear
out and ferrite heads and oxide media are gone. Occasional bits and
True. But other problems inside the HDA would be much easier to repair on
an older drive with a much larger head flying height. I think it would be
possible to make a 'clean box' to open up ST412s at home and repair them
and expect the repaired HDA to be reliable enough to use. That is not the
case with modern drives.
=
pieces
show up less and less often. Why not put the energy into adapting the
latest drive technology? That buys you an extra decade or so.=20
Becasue by the time I've designed and debugged the interface, the darn
drive I've chosen will be discontinued!
IDE drives are going to be around in far greater
volumes long after the =
last
ST506 is still working. I just had a report that last year was the =
I am not convinced of that, actually....
biggest
year ever for hard drives: 376 million units, the huge majority IDE. =
That
gives a pool of drives to use that will outlast most of us on this list.
And there are already IDE to SATA dongles to stretch out another =
generation.
And there are the side issues to keeping the old systems going. How =
long
will we be able to find 8" and 5.25" media? Printer ribbons? Rubber =
parts
for teletypes and typewriters? When do you toss in the towel on trying =
Rubber parts may not be a problem. There are 2-pack synthetic 'rubbers'
made, e.g. by Devcon. The data sheet I've seen actually says they can be
used to make rubber rollers.
OK, you'd have to make a mould, but that's not beyond a good home
workshop. And once you've done that you can turn out as many parts as you
need. I might actually be doing this in the next year or so to make a
roller for an old calculator printer.
-tony
31 May
31 May
3:50 p.m.
If my CDC 160-A memory goes, there's no more
core memories.
Core memory can be repaired, but it isn't fun.
6547
days inactive
6552
days old
68 comments
22 participants
participants (22)
-
ak6dn＠mindspring.com -
allain＠panix.com -
ard＠p850ug1.demon.co.uk -
aw288＠osfn.org -
bfranchuk＠jetnet.ab.ca -
Billy.Pettit＠wdc.com -
bpope＠wordstock.com -
brad＠heeltoe.com -
cclist＠sydex.com -
chrism3667＠yahoo.com -
dgy＠DakotaCom.Net -
dkelvey＠hotmail.com -
doc＠mdrconsult.com -
ethan.dicks＠gmail.com -
healyzh＠aracnet.com -
hilpert＠cs.ubc.ca -
lproven＠gmail.com -
pat＠computer-refuge.org -
pcw＠mesanet.com -
rtellason＠blazenet.net -
trixter＠oldskool.org -
wmaddox＠pacbell.net