1 Jan
2022
1 Jan
'22
10:12 a.m.
This:
https://www.ebay.com/itm/275084268137
The starting price is expensive, but probably not utterly unreasonable,
given that:
- the 780 was the first VAX, and thus historically important
- 780's are incredibly rare; this is the first one I recall seeing for sale
in the classic computer era (versus several -11/70's, /40s, etc)
- this one appears to be reasonably complete; no idea if all the key CPU
boards are included, but it's things like the backplane, etc (all of which
seem to be there) which would be completely impossible to find now - if any
boards _are_ missing, there's at least the _hope_ that they can be located
(780 boards seem to come by every so often on eBait), since people seem to
keep boards, not realizing that without the other bits they are useless
Anyway I fully expect it to go (because of those, especially i and ii) for a
_lot_ more than the opening price.
I've sent the seller info on the complete 780 board set, and a suggestion
that it's in their own best interest (maximize bidding) to check to see if
it's complete.
Noel
1 Jan
1 Jan
10:40 a.m.
On Jan 1, 2022, at 1:12 PM, Noel Chiappa via cctalk
<cctalk at classiccmp.org> wrote:
This:
https://www.ebay.com/itm/275084268137
The starting price is expensive, but probably not utterly unreasonable,
given that:
- the 780 was the first VAX, and thus historically important
- 780's are incredibly rare; this is the first one I recall seeing for sale
in the classic computer era (versus several -11/70's, /40s, etc)
- this one appears to be reasonably complete; no idea if all the key CPU
boards are included, but it's things like the backplane, etc (all of which
seem to be there) which would be completely impossible to find now - if any
boards _are_ missing, there's at least the _hope_ that they can be located
(780 boards seem to come by every so often on eBait), since people seem to
keep boards, not realizing that without the other bits they are useless
Interesting, but the argument for why it's not tested is implausible which makes me
very suspicious. I suppose there might be a few American homes that have only 110 volt
power, but I'm hard pressed to think of any I have ever seen, and that includes really
old houses.
paul
10:46 a.m.
On 1/1/22 1:40 PM, Paul Koning via cctalk wrote:
Having 240 in your house does not necessarily mean you have
240 outlets anywhere and not everyone is capable of doing
their own house wiring.
bill
On Jan 1, 2022, at 1:12 PM, Noel Chiappa via
cctalk <cctalk at classiccmp.org> wrote:
This:
https://www.ebay.com/itm/275084268137
The starting price is expensive, but probably not utterly unreasonable,
given that:
- the 780 was the first VAX, and thus historically important
- 780's are incredibly rare; this is the first one I recall seeing for sale
in the classic computer era (versus several -11/70's, /40s, etc)
- this one appears to be reasonably complete; no idea if all the key CPU
boards are included, but it's things like the backplane, etc (all of which
seem to be there) which would be completely impossible to find now - if any
boards _are_ missing, there's at least the _hope_ that they can be located
(780 boards seem to come by every so often on eBait), since people seem to
keep boards, not realizing that without the other bits they are useless
Interesting, but the argument for why it's not tested is implausible which makes me
very suspicious. I suppose there might be a few American homes that have only 110 volt
power, but I'm hard pressed to think of any I have ever seen, and that includes really
old houses.
10:51 a.m.
On Jan 1, 2022, at 1:46 PM, Bill Gunshannon via cctalk
<cctalk at classiccmp.org> wrote:
On 1/1/22 1:40 PM, Paul Koning via cctalk wrote:
Having 240 in your house does not necessarily mean you have
240 outlets anywhere and not everyone is capable of doing
their own house wiring.
True. But if you're trying to get > $5000 for something, it doesn't seem
unreasonable to suggest that investing a bit in getting an extension cord run to the
location of the machine would be a good idea. The absence of that effort makes me wonder
if the owner knows what the outcome of such a test would be and doesn't want to have
to report it.
paul
On Jan 1,
2022, at 1:12 PM, Noel Chiappa via cctalk <cctalk at classiccmp.org> wrote:
This:
https://www.ebay.com/itm/275084268137
The starting price is expensive, but probably not utterly unreasonable,
given that:
- the 780 was the first VAX, and thus historically important
- 780's are incredibly rare; this is the first one I recall seeing for sale
in the classic computer era (versus several -11/70's, /40s, etc)
- this one appears to be reasonably complete; no idea if all the key CPU
boards are included, but it's things like the backplane, etc (all of which
seem to be there) which would be completely impossible to find now - if any
boards _are_ missing, there's at least the _hope_ that they can be located
(780 boards seem to come by every so often on eBait), since people seem to
keep boards, not realizing that without the other bits they are useless
Interesting, but the argument for why it's not tested is implausible which makes me
very suspicious. I suppose there might be a few American homes that have only 110 volt
power, but I'm hard pressed to think of any I have ever seen, and that includes really
old houses. 
10:55 a.m.
On 1/1/22 11:51 AM, Paul Koning via cctalk wrote:
True. But if you're trying to get > $5000 for
something, it doesn't
seem unreasonable to suggest that investing a bit in getting an
extension cord run to the location of the machine would be a good idea.
The absence of that effort makes me wonder if the owner knows what the
outcome of such a test would be and doesn't want to have to report it.
Maybe.
But how many "one more step" tests can the seller do vs being honest and
up front.
Maybe the seller had heard about power supplies (caps) failing
spectacularly ~> catastrophically and doesn't know how nor want to do
the work to test.
What's after the "smoke test"?
Or, just call it where it is. ;-)
--
Grant. . . .
unix || die
11:19 a.m.
On 1/1/22 12:51 PM, Paul Koning via cctalk wrote:
True. But if you're trying to get
> $5000 for something, it doesn't seem unreasonable to suggest that investing a bit
in getting an extension cord run to the location of the machine would be a good idea. The
absence of that effort makes me wonder if the owner knows what the outcome of such a test
would be and doesn't want to have to report it.
The /780 came standard with a 5-pin Hubbell twist-lock
connector for 3-phase 208/240 V WYE power.
While it could be rewired for single-phase power, it would
take fairly massive rewiring of the power control.? The 3
blowers all ran off line-line power, I think, and the H7100
power supplies ran off 120V line-neutral from different phases.
Jon
11:50 a.m.
12:19 p.m.
On Sat, Jan 1, 2022 at 11:55 AM W2HX via cctalk <cctalk at classiccmp.org>
wrote:
True.
But if you're trying to get > $5000 for something, it doesn't seem
unreasonable to suggest that investing a bit in getting an extension cord
run to the location of the machine would be a good idea. The absence of
that effort makes me wonder if the owner knows what the outcome of such a
test would be and doesn't want to have to report it.
But what would that accomplish? I think testing something like this
requires a lot more effort than plugging it in and hitting the circuit
breaker. To test this to see if some ODT comes up probably requires quite a
lot of effort (locate a terminal/pc, wire it up, figure out where to plug
it into the 780, etc. If this guy is a bulk dealer I would be surprised if
he has the knowledge to do anything more than a power test which, again,
would not be very useful and could even be detrimental.
Right. I'll pay significantly less for an 11/780 that some random scrapper
has powered up.
- Josh
73 Eugene W2HX
Subscribe to my Youtube Channel:
https://www.youtube.com/c/w2hx-channel/videos
12:27 p.m.
On 1/1/22 1:50 PM, W2HX via cctalk wrote:
True. But if you're trying to get > $5000 for
something, it doesn't seem unreasonable to suggest that investing a bit in getting an
extension cord run to the location of the machine would be a good idea. The absence of
that effort makes me wonder if the owner knows what the outcome of such a test would be
and doesn't want to have to report it.
But what would that accomplish? I think testing something like this requires a lot more
effort than plugging it in and hitting the circuit breaker. To test this to see if some
ODT comes up probably requires quite a lot of effort (locate a terminal/pc, wire it up,
figure out where to plug it into the 780, etc. If this guy is a bulk dealer I would be
surprised if he has the knowledge to do anything more than a power test which, again,
would not be very useful and could even be detrimental.
The /780 needs a readable "console floppy" and working
drive.? It loads microcode from the floppy and runs some
power-on self-tests before even starting the VAX CPU.? Not
real likely the floppy is going to be in good condition or
the drive, either.? Also, the cooling blowers below the
circuit boards are likely to be seized and need new
bearings.? Not to mention the cap banks in the power
supplies at least need to be re-formed.? Likely, this
machine has not been turned on since 1990 or so, (just
guessing).
Jon
10:53 a.m.
On 1/1/22 11:46 AM, Bill Gunshannon via cctalk wrote:
Having 240 in your house does not necessarily mean you
have 240 outlets
anywhere and not everyone is capable of doing their own house wiring.
There may even be 240 V outlets but not available when / where needed.
E.g. in use (stove, dryer, water heater) or too far away to be able to
plug the VAX in for testing. Then there's also the chance that the
plugs don't match.
I do 120 V wiring semi-regular. I've done 240 V wiring before. I'm
sure that I'll do it again. But I'm always afraid that failure mode on
the 240 V is going to fail spectacularly. My only saving grace is that
the breaker will almost certainly trip in short order to mitigate my
failure. Thankfully no breaker trips thusfar.
--
Grant. . . .
unix || die
11:11 a.m.
On 1/1/22 1:53 PM, Grant Taylor via cctalk wrote:
On 1/1/22 11:46 AM, Bill Gunshannon via cctalk wrote:
In my current house I have done 240V/50A wiring, 240V/50A Sub Panel,
lots of 240V/30A outlets. None of which I would advise the usual
amateur to do. :-)
Just out of curiosity, how much current is needed for an 11/780?
bill
Having 240 in your house does not necessarily
mean you have 240
outlets anywhere and not everyone is capable of doing their own house
wiring.
There may even be 240 V outlets but not available when / where needed.
E.g. in use (stove, dryer, water heater) or too far away to be able to
plug the VAX in for testing.? Then there's also the chance that the
plugs don't match.
I do 120 V wiring semi-regular.? I've done 240 V wiring before.? I'm
sure that I'll do it again.? But I'm always afraid that failure mode on
the 240 V is going to fail spectacularly.? My only saving grace is that
the breaker will almost certainly trip in short order to mitigate my
failure.? Thankfully no breaker trips thusfar.
11:21 a.m.
On 1/1/22 12:17 PM, Dennis Boone via cctalk wrote:
IIRC nothing in the VAX cpu actually requires 3 phase.
The PDU(s)
just eat that from the wall.
I think that's the same with all the /computer/ equipment that I've seen
which connects to 3? power. A single phase could be used for all of it.
But I'm told that it's simpler / more economical (from a wiring /
cable plant point of view) to deliver more power to the equipment as 3?
than it is 1?.
--
Grant. . . .
unix || die
11:33 a.m.
On Jan 1, 2022, at 2:21 PM, Grant Taylor via cctalk
<cctalk at classiccmp.org> wrote:
On 1/1/22 12:17 PM, Dennis Boone via cctalk wrote:
Largely true, but some disk drives (RP06? RP04?) use 3-phase spindle motors. And
mainframes may use 3-phase power supplies; CDC 6000 mainframes do, though with those
it's 3 phase 400 Hz produced by motor-generators (which in turn run off 3-phase mains
power).
paul
IIRC nothing in the VAX cpu actually requires 3
phase. The PDU(s) just eat that from the wall.
I think that's the same with all the /computer/ equipment that I've seen which
connects to 3? power. A single phase could be used for all of it. But I'm told that
it's simpler / more economical (from a wiring / cable plant point of view) to deliver
more power to the equipment as 3? than it is 1?.
11:38 a.m.
On 1/1/22 12:33 PM, Paul Koning via cctalk wrote:
And mainframes may use 3-phase power supplies; CDC
6000 mainframes do,
though with those it's 3 phase 400 Hz produced by motor-generators
(which in turn run off 3-phase mains power).
I know that some IBM mainframes prefer 3? power but I know of multiple
people that have re-wired their CECs to use 1? power. They too are
largely independent 1? supplies on different phases to jointly power a
common internal bus. So re-wiring the necessary supplies to the same
phase is mostly not a big deal.
Apparently even IBM sold and supported some systems in this
configuration. The only down side is that it put an upper cap on the
amount of power that you could draw and thus things that you could run
in this configuration. But for smaller installations, it's perfectly fine.
--
Grant. . . .
unix || die
11:43 a.m.
I know that some IBM mainframes prefer 3? power but I
know of multiple
people that have re-wired their CECs to use 1? power.
My Multiprise 2003/205 claims to want a pair of 3-phase inputs. I've
run it on two "phases" on a single side. It bitches, of course, and one
loses the intended redundancy, but it runs.
De
11:57 a.m.
On 1/1/22 12:43 PM, Dennis Boone via cctalk wrote:
My Multiprise 2003/205 claims to want a pair of
3-phase inputs.
Is it assumed that those 2 x 3? inputs are feed from different paths,
thus path redundancy? I'd think that the phases would be the same.
I've run it on two "phases" on a single
side.
I have questions....
Where are you getting two /different/ phases? -- Remember, the
different legs on residential 120/240 wiring are really the same single
phase.
How do you get *two* /different/ phases without access to a *third*
phase? There are only a few places in the U.S.A. (and I'm not aware of
anywhere else in the world) that actually have 2? power (where the ? are
90? out of phase with each other).
What is a phase vs what is a line comes into context here.
It bitches, of course,
I could see how wiring two legs of a 120/240 1? as if it were 2 phases
of a 3? supply would make something mad.
and one loses the intended redundancy, but it runs.
Indeed.
But if one is doing this, chances are fairly good that one is most
interested in the fact that /it/ /runs/ and cares much less about
/redundancy/. ;-)
--
Grant. . . .
unix || die
11:59 a.m.
I have questions....
Where are you getting two /different/ phases? -- Remember, the
different legs on residential 120/240 wiring are really the same single
phase.
They're 180 degrees out. Sufficiently different for a number of applications, but
certainly not all!
Thanks,
Jonathan
12:11 p.m.
On 1/1/22 12:59 PM, Jonathan Chapman via cctalk wrote:
They're 180 degrees out. Sufficiently different
for a number of
applications, but certainly not all!
They are 180? out of phase with respect to each other. But they are the
same single phase when viewed by anything outside of them.
--
Grant. . . .
unix || die
11:44 a.m.
On Jan 1, 2022, at 2:38 PM, Grant Taylor via cctalk
<cctalk at classiccmp.org> wrote:
On 1/1/22 12:33 PM, Paul Koning via cctalk wrote:
Could be. For the CDC case, I see actual 3-phase DC supplies, i.e., 3 phase bridge
rectifiers fed by 3-phase transformers. If you run those off one phase the ripple would
be a lot larger. And of course 400 Hz gave them an additional advantage in reduced
transformer size and reduced ripple (for a given filter capacitor).
They'd do this even in devices that don't draw large amounts of power, for example
the DD60 console display gets its power (other than fans which are single phase mains
power) from the 400 Hz 3-phase supply feeding the mainframe. I'd be surprised if it
uses even a kW, so this was probably a case of "why not since it's there to be
used".
paul
And mainframes may use 3-phase power supplies;
CDC 6000 mainframes do, though with those it's 3 phase 400 Hz produced by
motor-generators (which in turn run off 3-phase mains power).
I know that some IBM mainframes prefer 3? power but I know of multiple people that have
re-wired their CECs to use 1? power. They too are largely independent 1? supplies on
different phases to jointly power a common internal bus. So re-wiring the necessary
supplies to the same phase is mostly not a big deal.
Apparently even IBM sold and supported some systems in this configuration. The only down
side is that it put an upper cap on the amount of power that you could draw and thus
things that you could run in this configuration. But for smaller installations, it's
perfectly fine.
12:08 p.m.
On 1/1/22 12:44 PM, Paul Koning via cctalk wrote:
Could be. For the CDC case, I see actual 3-phase DC
supplies, i.e.,
3 phase bridge rectifiers fed by 3-phase transformers.
Interesting. That brings to mind 6? DC supplies. -- Two typical 3?
rectifier networks (making DC pulses be 60? apart) separated by a ? / Y
or Y / ? transformer inducing a 30? phase shift between them. Thereby
causing a net effect of 12 x DC pulses 30? apart.
If you run those off one phase the ripple would be a
lot larger.
Indeed.
And of course 400 Hz gave them an additional advantage
in reduced
transformer size and reduced ripple (for a given filter capacitor).
I was going to ask how the frequency altered the size of the ripple, but
"for a given filter capacitor" makes sense. You are messing with the
ratio and having a fixed value "for a given filter capacitor". -- If
I'm even remotely understanding correctly.
They'd do this even in devices that don't draw
large amounts of power,
for example the DD60 console display gets its power (other than fans
which are single phase mains power) from the 400 Hz 3-phase supply
feeding the mainframe. I'd be surprised if it uses even a kW, so
this was probably a case of "why not since it's there to be used".
My understanding is that some of the special power requirements had more
to do with isolation and cleanliness / quality of the power. Meaning
don't try to power the equipment off of the same circuit(s) that motors
and the likes are on which will induce lots of noise on the line, some
of which might make it into the computer and adversely effect things.
The motor / generator provides a quite strong isolation for things like
that. Probably more so than just a normal transformer.
The frequency change means that a simple isolation transformer won't
suffice. The higher frequency also has benefits of reducing the size of
other down stream components.
I find this to be interesting to learn, even if I'll almost certainly
never actually work near anything that uses it.
--
Grant. . . .
unix || die
12:27 p.m.
On Jan 1, 2022, at 3:08 PM, Grant Taylor via cctalk
<cctalk at classiccmp.org> wrote:
On 1/1/22 12:44 PM, Paul Koning via cctalk wrote:
The impedance of a capacitor is inversely proportional to the frequency. Given a desired
ripple for a given load, you get the needed capacitor impedance, so if you increase the
freqency you'd end up with smaller capacitors. You see this, big time, in modern
switching supplies, where the frequency may into the many kHz or sometimes even a MHz or
so, resulting in tiny filter capacitors.
...
And of course 400 Hz gave them an additional advantage in reduced transformer size and
reduced ripple (for a given filter capacitor).
I was going to ask how the frequency altered the size of the ripple, but "for a
given filter capacitor" makes sense. You are messing with the ratio and having a
fixed value "for a given filter capacitor". -- If I'm even remotely
understanding correctly. They'd do
this even in devices that don't draw large amounts of power, for example the DD60
console display gets its power (other than fans which are single phase mains power) from
the 400 Hz 3-phase supply feeding the mainframe. I'd be surprised if it uses even a
kW, so this was probably a case of "why not since it's there to be used".
My understanding is that some of the special power requirements had more to do with
isolation and cleanliness / quality of the power. Meaning don't try to power the
equipment off of the same circuit(s) that motors and the likes are on which will induce
lots of noise on the line, some of which might make it into the computer and adversely
effect things. The motor / generator provides a quite strong
isolation for things like that. Probably more so than just a normal transformer.
Way more because of inertia. Not quite a UPS, but short power glitches enough to blink
the computer room lights are not going to be seen by the mainframe.
paul
2 Jan
2 Jan
6:19 p.m.
On Sat, Jan 1, 2022, 12:33 Paul Koning via cctalk <cctalk at classiccmp.org>
wrote:
Largely true, but some disk drives (RP06? RP04?) use
3-phase spindle
motors.
The RP06 has 3-phase power input (and output, phase-rotated, for a second
drive), but uses a single phase spindle motor. The US version runs the
spindle between two phases, nominally 208V, but the motor is rated for
operation on a range that spans 208V and 240V.
1 Jan
1 Jan
11:35 a.m.
I think that's the same with all the /computer/
equipment that I've seen
which connects to 3? power. A single phase could be used for all of it.
Computer exist which require three phase at a unit power supply level. It's often also
used for large blowers if the cabinet(s) don't strictly require airflow from e.g. a
raised floor.
But I'm told that it's simpler / more
economical (from a wiring /
cable plant point of view) to deliver more power to the equipment as 3?
than it is 1?.
It also saves you on current-carrying conductors for raceway fill calculations as long as
your power factor isn't too bad.
Thanks,
Jonathan
11:56 a.m.
The other benefit to 3phase power to something like a computer is that the filter caps
required can be much smaller. Once you rectify three phases 60* you get MUCH less ripple
because every 60 degrees you have a new peak arriving. When you have a real gas (electron)
guzzler like one of these, DC filtering becomes an issue.
73 Eugene W2HX
Subscribe to my Youtube Channel: https://www.youtube.com/c/w2hx-channel/videos
-----Original Message-----
From: cctalk <cctalk-bounces at classiccmp.org> On Behalf Of Jonathan Chapman via
cctalk
Sent: Saturday, January 1, 2022 2:35 PM
To: Grant Taylor <cctalk at gtaylor.tnetconsulting.net>; General Discussion:
On-Topic and Off-Topic Posts <cctalk at classiccmp.org>
Subject: Re: VAX 780 on eBay
I think that's the same with all the /computer/
equipment that I've
seen which connects to 3? power. A single phase could be used for all of it.
Computer exist which require three phase at a unit power supply level. It's often also
used for large blowers if the cabinet(s) don't strictly require airflow from e.g. a
raised floor.
But I'm told that it's simpler / more
economical (from a wiring /
cable plant point of view) to deliver more power to the equipment as
3? than it is 1?.
It also saves you on current-carrying conductors for raceway fill calculations as long as
your power factor isn't too bad.
Thanks,
Jonathan
12:32 p.m.
On 1/1/22 1:56 PM, W2HX via cctalk wrote:
The other benefit to 3phase power to something like a
computer is that the filter caps required can be much smaller. Once you rectify three
phases 60* you get MUCH less ripple because every 60 degrees you have a new peak arriving.
When you have a real gas (electron) guzzler like one of these, DC filtering becomes an
issue.
But, they DIDN'T!? All power supplies on the /780 were
single phase.
On the IBM 370/145, they used a motor/generator set to
convert 208/3 phase 60 Hz power to 415 Hz 3-phase 120 V
regulated power, then fed that to 3-phase
transformer/rectifiers, and finally to low-drop regulators.?
They also used "electronic capacitors", which was a scheme
to pull extra current through the smoothing inductor during
the voltage peaks, to even out the ripple.? The 370/145 had
2 power supplies, one for +1.25 V 390A and the other for -3
V 390A.
Jon
11:36 a.m.
On 1/1/22 1:11 PM, Bill Gunshannon via cctalk wrote:
Just out of curiosity, how much current is needed for an
11/780?
Depending on options, the CPU cabinet would draw about 10-12
A per line, 3-Phase 208 V.? And, that is on the upper end of
the range.? Starting surge would be a lot more, of course.
Jon
12:35 p.m.
On 2022-01-01 12:36 p.m., Jon Elson via cctalk wrote:
On 1/1/22 1:11 PM, Bill Gunshannon via cctalk wrote:
How much power is used for memory
and hard drives?
Ben.
Just out of curiosity, how much current is needed for an 11/780?
Depending on options, the CPU cabinet would draw about 10-12 A per line,
3-Phase 208 V.? And, that is on the upper end of the range.? Starting
surge would be a lot more, of course.
Jon
2:24 p.m.
On 1/1/22 2:35 PM, ben via cctalk wrote:
On 2022-01-01 12:36 p.m., Jon Elson via cctalk wrote:
memory power was provided by the row of power supplies below
the CPU cards as seen in the pics.
Memory power on the /780 was pretty insignificant, as that
system didn't have large memories.
The original /780 had 256 KB memory cards, I think the one
in the current sale has the newer memory system with 1 MB/card.
Hard drive power was QUITE substantial, the bigger drives
had pretty big 3 phase motors to spin the big disk packs/HDA's.
Jon
On 1/1/22 1:11 PM, Bill Gunshannon via cctalk
wrote:
How much power is used for memory
and hard drives?
Just out of curiosity, how much current is needed for an
11/780?
Depending on options, the CPU cabinet would draw about
10-12 A per line, 3-Phase 208 V.? And, that is on the
upper end of the range.? Starting surge would be a lot
more, of course.
Jon
12:16 p.m.
It splits the input 3 phase 120/208 30A input to separate 120V circuits.
3*30A 120V circuits if you want. Mine used about 3KVA, so it'll actually
(just barely) run off of a single 30A/120V or multiple 15/20A 120V circuits
if you rewire the input a bit.
Patrick Finnegan
On Sat, Jan 1, 2022, 14:11 Bill Gunshannon via cctalk <cctalk at classiccmp.org>
wrote:
On 1/1/22 1:53 PM, Grant Taylor via cctalk wrote:
On 1/1/22 11:46 AM, Bill Gunshannon via cctalk
wrote:
In my current house I have done 240V/50A wiring, 240V/50A Sub Panel,
lots of 240V/30A outlets. None of which I would advise the usual
amateur to do. :-)
Just out of curiosity, how much current is needed for an 11/780?
bill
Having 240 in your house does not necessarily
mean you have 240
outlets anywhere and not everyone is capable of doing their own house
wiring.
There may even be 240 V outlets but not available when / where needed.
E.g. in use (stove, dryer, water heater) or too far away to be able to
plug the VAX in for testing. Then there's also the chance that the
plugs don't match.
I do 120 V wiring semi-regular. I've done 240 V wiring before. I'm
sure that I'll do it again. But I'm always afraid that failure mode on
the 240 V is going to fail spectacularly. My only saving grace is that
the breaker will almost certainly trip in short order to mitigate my
failure. Thankfully no breaker trips thusfar. 
1 p.m.
In my current house I have done 240V/50A wiring,
240V/50A Sub Panel,
lots of 240V/30A outlets.? None of which I would advise the usual
amateur to do.? :-)
This is wise, but 240v is quite nice for computer equipment. I had an
electrician run a 240v 30a (10g wire) circuit out to my work shed, where
a subpanel is installed with 120v 15a and 240v 15a circuits for the
equipment.
240 just allows you to get twice the power to the device using the same
size wire (14g for 15a for example). But modern switching computer
supplies are more efficient at 240v than 120v so you waste less power.
120 is good for like light bulbs and stuff. Drawing over say 10a
continuous at 120 will pretty much burn up sockets and such in short order.
10:20 p.m.
On 1/1/22 2:00 PM, Chris Zach via cctalk wrote:
This is wise, but 240v is quite nice for computer
equipment. I had an
electrician run a 240v 30a (10g wire) circuit out to my work shed, where
a subpanel is installed with 120v 15a and 240v 15a circuits for the
equipment.
I had two 120v 20A circuits, one on each leg of the 120v/240v
residential feed, run to my computer equipment in the house I sold a few
years ago. I'd like to have the same again once I finally set things up
the way that I want to.
I'd really like 3? at home for a number of different things. That would
allow me to run computer equipment at 208v which, as I understand it, is
more efficient than 120v. I don't think anything I do would actually
/need/ 240v. I'd also like 3? for other things in the house, namely
wood working tools.
240 just allows you to get twice the power to the
device using the same
size wire (14g for 15a for example).
Ya. One might wonder why higher voltage means that you can get twice
the power (wattage) to something. But that's where the fact that most
wire used for 120v circuits is rated for up to 250v which means that the
same wire can safely carry 240v. Hence how you can get twice the power
by doubling the voltage.
But modern switching computer supplies are more
efficient at 240v
than 120v so you waste less power.
Do you have something that you can point to as
reference?
Where does 208v compare to 240v / 120v?
120 is good for like light bulbs and stuff. Drawing
over say 10a
continuous at 120 will pretty much burn up sockets and such in short order.
It depends on the quality of the socket. My understanding is that a
NEMA 5-15 is rated to sustain 15A at 120V indefinitely. Then there are
NEMA 5-20s which are rated for 20A. So ... over 10A continuous seems
quite possible if not expected with some installations.
--
Grant. . . .
unix || die
2 Jan
2 Jan
11:16 a.m.
I'd really like 3? at home for a number of
different things.
In North America? Good luck! Part of what got me to buy a smallish industrial building was
needing three phase. It's usually cheaper/easier to either use a VFD if it's for
motor equipment, or a rotary converter, than trying to get the power company to give you
three phase. Even if it's literally on the pole behind your house, they want $LOL to
make the connection.
It depends on the quality of the socket. My
understanding is that a
NEMA 5-15 is rated to sustain 15A at 120V indefinitely. Then there are
NEMA 5-20s which are rated for 20A.
Yes, these "melted the socket, never tripped the breaker" failures are a result
of crap receptacles. On basically all good brands of receptacles, the 15A and 20A have the
same internal parts, anyway. It's my opinion that anything in a shop area, or really
even the kitchen, ought to be spec-grade receptacles. They're not that expensive when
you consider they'll last a lifetime. We always require Hubble spec-grade, and
that's what I've installed at the house, too.
So ... over 10A continuous seems
quite possible if not expected with some installations.
Indeed, you can have a continuous load up to 12A with no special NEC rules on a regular
old branch circuit.
Thanks,
Jonathan
12:37 p.m.
On 1/2/22 11:16 AM, Jonathan Chapman via cctalk wrote:
Availability depends on your utility and location. It's not uncommon
for a farm to have 3-phase service--it's needed for things like hay dryers.
For example, here's my rural electric co-op's rate sheet for 3-phase
service:
https://laneelectric.com/wp-content/uploads/GS-3-General-Service-Three-Phas…
By way of comparison, here's their single-phase general-service rate sheet:
https://laneelectric.com/wp-content/uploads/GS-1-General-Service-Single-Pha…
The real sting is for 3-phase electric vehicle charging stations:
https://laneelectric.com/wp-content/uploads/EV-1-Electric-Vehicle-Charging-…
In my rural location, 3-phase 11KV buried is available at a disconnect
box at the street--single phase is buried along the driveway and
terminates at a transformer on a pad in my front yard. Buried 240/120V
single-phase 200A service then goes to the house.
--Chuck
I'd really
like 3? at home for a number of different things.
In North America? Good luck! Part of what got me to buy a smallish industrial building
was needing three phase. It's usually cheaper/easier to either use a VFD if it's
for motor equipment, or a rotary converter, than trying to get the power company to give
you three phase. Even if it's literally on the pole behind your house, they want $LOL
to make the connection. 
3:21 p.m.
The poles supplying my rural property just have one phase of 12kV on them, so even if SCE
was willing to provide 3? service to me, I bet it would be *expensive* to get the whole
line upgraded to 3? service all the way up to the last pole with all three phases present.
I think that's probably within a mile or so, but that's still a bunch of poles to
be upgraded. I use a rotary phase converter to feed my milling machine and lathe. If all
three phases were on the pole already, I probably would have asked about getting 3?
service when I built on my property. But with just one phase on the poles, I didn't
bother asking.
They did just replace those overhead wires and some poles for fire prevention, and in the
process they put in a whole great big pole-mounted switch that can be flipped from ground
level just for my service disconnect, and another one for a neighbor's disconnect. It
looks like they installed a regular 3? disconnect switch, but only two of the wires are
present and one switch pole is not connected. My service transitions from overhead to
buried at that pole, and is buried for the last 100' or so to my ground-mounted
transformer.
Two of the poles they replaced were in the easement just on the other side of my property
line. The contractors were happy with my cooperation in giving them access to my property
so they could put trucks on both sides of the fence line. I also handed out 15 60mm ammo
cans from my stash, since I like contractors on my property to be happy. So, they
conveniently forgot to take the two old 45' poles they replaced with them at the end
of the day, and I won't have to buy a new ham radio tower after all. ;)
I am seriously lusting after that VAX-11/780 system on eBay, but I just don't have the
available indoor space for it, or a big enough pile of cash to buy it and then spend
enough to give it a good home. I'm glad that it's not in southern California,
because if it was within an hour or two of driving I don't think I'd be able to
resist the temptation. I do have my nice little 11/730 already. I'm eying the UNIBUS
expansion bus from that seller to add to my 730, but I think should leave it available for
the buyer of that 780.
--
Mark J. Blair <nf6x at nf6x.net>
Blog: https://www.nf6x.net
Git: https://gitlab.com/users/NF6X/groups
HECnet: DOGPAK::MBLAIR
Twitter: @nf6x
3:28 p.m.
In the unlikely event that I win the lottery before that 11/780 batch sells, I'm going
to buy it, buy a box truck, go pick it up, and then leave the box truck parked outside
while I build a new building to put it all in. :D
--
Mark J. Blair <nf6x at nf6x.net>
Blog: https://www.nf6x.net
Git: https://gitlab.com/users/NF6X/groups
HECnet: DOGPAK::MBLAIR
Twitter: @nf6x
4:20 p.m.
On 1/2/2022 6:21 PM, Mark J. Blair via cctalk wrote:
I am seriously lusting after that VAX-11/780 system on
eBay
I keep thinking about it, but no. The 780 was neat from a historical
perspective, and the 785 at UMBC (UMBC5) ran Ultrix (and was on the
ARPANET, not the crappy Bitnet the 8600's were on) but the 780 was
really slow. I have a Microvax II here that would equal it CPU
speed-wise, and the running of the main memory on the SBI was cool for
the 782 option but was pretty slow.
Now a Vax 8600 or 8650.... That would be interesting. Partially due to
the Jupiter angle, partially because it was the fastest MASSBUS system
with true pdp11 compatibility.
5:04 p.m.
On Jan 2, 2022, at 4:20 PM, Chris Zach via cctalk
<cctalk at classiccmp.org> wrote:
On 1/2/2022 6:21 PM, Mark J. Blair via cctalk wrote:
The VAX-11/780 may be very slow, but it is still still quite a bit faster than my 11/730.
That being said, the 11/730 captures the look and feel of that era of VAX in a much
smaller volume and with a lot lower energy cost.
I am seriously lusting after that VAX-11/780
system on eBay
I keep thinking about it, but no. The 780 was neat from a historical
perspective, and the 785 at UMBC (UMBC5) ran Ultrix (and was on the ARPANET, not the
crappy Bitnet the 8600's were on) but the 780 was really slow. I have a Microvax II
here that would equal it CPU speed-wise, and the running of the main memory on the SBI was
cool for the 782 option but was pretty slow. Now a Vax 8600 or 8650.... That would be interesting.
Partially due to the Jupiter angle, partially because it was the fastest MASSBUS system
with true pdp11 compatibility.
Oh, heck yeah! I'd love to have an 8600 or 8650, and I would see them as the logical
conclusion of the VAX-11/7xx line despite the model number change.
--
Mark J. Blair <nf6x at nf6x.net>
Blog: https://www.nf6x.net
Git: https://gitlab.com/users/NF6X/groups
HECnet: DOGPAK::MBLAIR
Twitter: @nf6x
8:31 p.m.
The VAX-11/780 may be very slow, but it is still still
quite a bit faster than my 11/730. That being said, the 11/730 captures the look and feel
of that era of VAX in a much smaller volume and with a lot lower energy cost.
Oh yes, the 730 is probably the neatest little "pocket Vax". Especially
if you have the R80 drive as well as the RL02. The R80 did not use the
Unibus, correct?
9:10 p.m.
On Sun, Jan 2, 2022 at 11:31 PM Chris Zach via cctalk
<cctalk at classiccmp.org> wrote:
Oh yes, the 730 is probably the neatest little
"pocket Vax". Especially
if you have the R80 drive as well as the RL02. The R80 did not use the
Unibus, correct?
Correct. The R80 connects to the RB730 controller which has its own
slot at the top of the backplane.
My first UNIX install was on an 11/730. I ran a Usenet node (k-panda)
for a couple of years on it.
-ethan
5:11 p.m.
On 1/2/22 5:20 PM, Chris Zach via cctalk wrote:
I keep thinking about it, but no. The 780 was neat
from a historical
perspective, and the 785 at UMBC (UMBC5) ran Ultrix (and was on the
ARPANET, not the crappy Bitnet the 8600's were on)
Is that Bitnet as in the
Because it was there network that many IBM
mainframes were on?
but the 780 was really slow. I have a Microvax II here
that would
equal it CPU speed-wise, and the running of the main memory on the
SBI was cool for the 782 option but was pretty slow.
Can I ask for a rough translation in to comparative VAX Units of
Performance (VUPs)? I /think/ that the VAX 11/780 was 1 VUP.
--
Grant. . . .
unix || die
5:18 p.m.
On Jan 2, 2022, at 5:11 PM, Grant Taylor via cctalk
<cctalk at classiccmp.org> wrote:
?On 1/2/22 5:20 PM, Chris Zach via cctalk wrote:
VAX-11/780 = 1 VUP
VAX-11/782 = 1.8 VUP
MicroVAX 1 = 0.3 VUP
MicroVAX 2 = 0.7 VUP
Zane
I keep thinking about it, but no. The 780 was
neat from a historical perspective, and the 785 at UMBC (UMBC5) ran Ultrix (and was on the
ARPANET, not the crappy Bitnet the 8600's were on)
Is that Bitnet as in the
Because it was there network that many IBM mainframes were on?
but the 780 was really slow. I have a Microvax II
here that would equal it CPU speed-wise, and the running of the main memory on the SBI was
cool for the 782 option but was pretty slow.
Can I ask for a rough translation in to comparative VAX Units of Performance (VUPs)? I
/think/ that the VAX 11/780 was 1 VUP. 
5:20 p.m.
I've always felt that in terms of performance the proper measure would
have been VUPS/kW :-)
That way my little MicroVAX 2 would be able to hold its head high!
Nigel Johnson, MSc., MIEEE, MCSE VE3ID/G4AJQ/VA3MCU
Amateur Radio, the origin of the open-source concept!
Skype: TILBURY2591nw.johnson at ieee.org
On 2022-01-02 20:18, Zane Healy via cctalk wrote:
On Jan 2,
2022, at 5:11 PM, Grant Taylor via cctalk<cctalk at classiccmp.org> wrote:
?On 1/2/22 5:20 PM, Chris Zach via cctalk wrote:
VAX-11/780 = 1 VUP
VAX-11/782 = 1.8 VUP
MicroVAX 1 = 0.3 VUP
MicroVAX 2 = 0.7 VUP
Zane
I keep thinking about it, but no. The 780 was
neat from a historical perspective, and the 785 at UMBC (UMBC5) ran Ultrix (and was on the
ARPANET, not the crappy Bitnet the 8600's were on)
Is that Bitnet as in the
Because it was there network that many IBM mainframes were on?
but the 780 was really slow. I have a Microvax II
here that would equal it CPU speed-wise, and the running of the main memory on the SBI was
cool for the 782 option but was pretty slow.
Can I ask for a rough translation in
to comparative VAX Units of Performance (VUPs)? I /think/ that the VAX 11/780 was 1 VUP.
5:28 p.m.
On Jan 2, 2022, at 5:20 PM, Nigel Johnson Ham via cctalk <cctalk at classiccmp.org>
wrote:
?I've always felt that in terms of performance the proper measure would have been
VUPS/kW :-)
That way my little MicroVAX 2 would be able to hold its head high!
On that note a Raspberry Pi 2b running SIMH/VAX is about 1.6 VUPS.
Zane
5:59 p.m.
On 2022-01-02 6:28 p.m., Zane Healy via cctalk wrote:
On Jan 2, 2022, at 5:20 PM, Nigel Johnson Ham via
cctalk <cctalk at classiccmp.org> wrote:
But can the Pi handle a gazillion students all time sharing
at once @ 2400? How long was the VAX timesharing era as I suspect
networked PC's come out soon after that.
Ben.
?I've always felt that in terms of performance the proper measure would have been
VUPS/kW :-)
That way my little MicroVAX 2 would be able to hold its head high!
On that note a Raspberry Pi 2b running SIMH/VAX is about 1.6 VUPS.
Zane
6:14 p.m.
On 1/2/22 6:59 PM, ben via cctalk wrote:
But can the Pi handle a gazillion students all time
sharing at once
@ 2400?
I think that will depend on how you connect the serial terminals.
I know that it's possible to establish network connectivity to serial
terminal servers. I don't know how many of those terminal servers can
be connected to SimH. It may be better to use LAT between terminal
servers and VMS running in the emulated VAX.
How long was the VAX timesharing era as I suspect
networked PC's come
out soon after that.
Good question.
--
Grant. . . .
unix || die
3 Jan
3 Jan
8:13 a.m.
On Jan 2, 2022, at 9:14 PM, Grant Taylor via cctalk
<cctalk at classiccmp.org> wrote:
On 1/2/22 6:59 PM, ben via cctalk wrote:
That was the whole point of LAT. At the time almost all serial terminals were connected by
DZ-style interfaces, so 1 interrupt when you typed a character and a second interrupt
after the character was echoed. LAT created a virtual circuit between the terminal server
and a specific host. The slot layer ran over the virtual circuit between a logical
terminal on the host and a physical terminal on the server. Data messages at the virtual
circuit layer were limited to no more frequent than a terminal server parameter ( default
80mS). With appropriate optimizations and data availability it was possible to shrink the
interrupt load on the host to 1-2 interrupts/ethernet packet (say around 1400 characters).
There were a lot of internal religious wars around the use of CTERM/Telnet/LAT.
After finishing DECnet-Ultrix I had a couple of weeks of spare time and implemented a
prototype Ultrix-LAT - one of the good things that came out of having a MicroVAX I in my
office!
John.
But can the Pi handle a gazillion students all
time sharing at once @ 2400?
I think that will depend on how you connect the serial terminals.
I know that it's possible to establish network connectivity to serial terminal
servers. I don't know how many of those terminal servers can be connected to SimH.
It may be better to use LAT between terminal servers and VMS running in the emulated VAX.
How long was
the VAX timesharing era as I suspect networked PC's come out soon after that.
Good question.
--
Grant. . . .
unix || die
10:58 a.m.
The discussion about 3-phase made me check some documentation.
The drive I remembered that uses 3-phase power is the RP04. I remember a warning to the
installer to verify the phase order; if that's wrong the drive will try to spin up in
the wrong direction, which might partially unscrew the pack from the spindle -- that would
be bad if the heads tried to load.
The RP07 is also 3-phase. Other drives I looked at are 1 phase.
I also looked at some CDC 6000 series docs. Those used 400 Hz 3 phase for supplying the
CPU and peripheral logic. The CPU cabinets also take 3 phase mains power for the
compressors (for the Freon cooling system). Curiously, the 1964-era 6603 disk drive uses
only 400 Hz power. I wonder if that feeds the spindle motor, or if that was a DC motor?
And the 626 tape drive wants 3-phase mains power.
paul
12:36 p.m.
On 1/3/22 10:58 AM, Paul Koning via cctalk wrote:
The discussion about 3-phase made me check some
documentation.
I also looked at some CDC 6000 series docs. Those
used 400 Hz 3 phase for supplying the CPU and peripheral logic. The CPU cabinets also
take 3 phase mains power for the compressors (for the Freon cooling system). Curiously,
the 1964-era 6603 disk drive uses only 400 Hz power. I wonder if that feeds the spindle
motor, or if that was a DC motor? And the 626 tape drive wants 3-phase mains power.
Are you certain about that? The '66 document for the Bryant series I
drives/6603
http://bitsavers.org/pdf/bryant60127000B_Bryant4000_ServiceHbk_Oct66.pdf
on page 1-3 mentions 280/440 3-phase just for the spindle motor (5 or 10
hp, depending on number of disks). The hydraulic pump had its own
3-phase motor. I think the 6603 electronics were 400Hz. Serious
current draw on those motor lines too--the Series II shows power
consumption draw at 280VAC--somewhere around 50A.
I remember the leak jugs--and the hydraulic fluid that occasionally made
it onto the floor. An operator at SVLOPS made a dash to take care of a
501 printer and slipped and fell in a puddle of the stuff. It was funny
at the time...
The old 808/6638 drives, which displaced the 6603, similarly used 280V
3-phase for the 5 hp spindle motor and hydraulic pump and 440Hz for the
electronics.
The 821 was sort of a double-capacity version of the 808. Other than
for a few units, it never made it to general distribution, being
rendered obsolete by the 844 disk pack drives. I recall that CDC
qualified for a lucrative contract by dint of Mike Miller noticing that
one of the 821s had gone offline and doing a 100 yard sprint to punch
the button. CDC qualified by a matter of seconds.
--Chuck
1:23 p.m.
On Jan 3, 2022, at 3:36 PM, Chuck Guzis via cctalk
<cctalk at classiccmp.org> wrote:
On 1/3/22 10:58 AM, Paul Koning via cctalk wrote:
I forgot that drive was Bryant. The 400 Hz only information comes from the 6000 Site Prep
manual, so I guess they got it wrong.
The discussion about 3-phase made me check some
documentation.
I also looked at some CDC 6000 series docs.
Those used 400 Hz 3 phase for supplying the CPU and peripheral logic. The CPU cabinets
also take 3 phase mains power for the compressors (for the Freon cooling system).
Curiously, the 1964-era 6603 disk drive uses only 400 Hz power. I wonder if that feeds
the spindle motor, or if that was a DC motor? And the 626 tape drive wants 3-phase mains
power.
Are you certain about that? The '66 document for the Bryant series I
drives/6603
http://bitsavers.org/pdf/bryant60127000B_Bryant4000_ServiceHbk_Oct66.pdf
on page 1-3 mentions 280/440 3-phase just for the spindle motor (5 or 10
hp, depending on number of disks). The hydraulic pump had its own
3-phase motor. I think the 6603 electronics were 400Hz. Serious
current draw on those motor lines too--the Series II shows power
consumption draw at 280VAC--somewhere around 50A. I remember the leak jugs--and the hydraulic fluid that
occasionally made
it onto the floor. An operator at SVLOPS made a dash to take care of a
501 printer and slipped and fell in a puddle of the stuff. It was funny
at the time...
The old 808/6638 drives, which displaced the 6603, similarly used 280V
3-phase for the 5 hp spindle motor and hydraulic pump and 440Hz for the
electronics.
The only hydraulic actuator drive I ever saw in person was an IBM 1311. Found out when it
sprang a leak and sprayed hydraulic oil all over the boot pack. FE guy cleaned everything
(heads, pack, etc.), fixed the seal and bled the system, and it all worked afterwards. No
damage to the pack.
The 6603 is highly unusual because it has 12 bit parallel data flow, rather than
bit-serial as everyone else did (until Cray went back to parallel with the Cray 1, if I
remember right). That made the 6603 very much faster, as far as data throughput goes,
than any other drive for quite a number of years. It also has variable sector counts
depending on cylinder number, which came back a long time later. And it has a rotating
head actuator rather than linear motion, just as recent hard drives do.
paul
3:01 p.m.
On 1/3/22 1:23 PM, Paul Koning wrote:
The 6603 is highly unusual because it has 12 bit
parallel data flow, rather than bit-serial as everyone else did (until Cray went back to
parallel with the Cray 1, if I remember right). That made the 6603 very much faster, as
far as data throughput goes, than any other drive for quite a number of years. It also
has variable sector counts depending on cylinder number, which came back a long time
later. And it has a rotating head actuator rather than linear motion, just as recent hard
drives do.
The 808/6638 is parallel also (I even have one of the heads in my desk
drawer memorabilia collection (6 channel).
A real Rube Goldberg setup--4 spindles, 2 motors, 2 positioners, with
each positioner having only 32 possible positions. 12 heads per "track"
on 32 surfaces gives you 384 track access without moving.
There's a photo on twitter:
https://twitter.com/DonaldM38768041/status/1215804561333473280/photo/2
showing a guy standing before an open one at Fermilab.
--Chuck
6:49 p.m.
On 1/3/22 4:01 PM, Chuck Guzis via cctalk wrote:
Interesting series of tweets.
I'm surprised by the tweet about Seymour Cray founding Control Data
Corporation. I know that Cray /worked/ for CDC, and left to found Cray
Research, but I was not aware that Seymour had anything to do with
founding CDC.
--
Grant. . . .
unix || die
11:14 p.m.
On 1/3/22 6:49 PM, Grant Taylor via cctalk wrote:
On 1/3/22 4:01 PM, Chuck Guzis via cctalk wrote:
Interesting series of tweets.
I'm surprised by the tweet about Seymour Cray founding Control Data
Corporation.? I know that Cray /worked/ for CDC, and left to found Cray
Research, but I was not aware that Seymour had anything to do with
founding CDC.
Seymour Cray, along with Bill Norris and Jim Thornton and others left
Remington Rand/UNIVAC after Rand bought the near-bankrupt ERA.
Apparently, the work environment at Rand was felt to be stifling.
Norris had all of the Navy connections and was a great marketer, so
bringing some of Rand's engineering talent along was a natural.
--Chuck
4 Jan
4 Jan
7:40 a.m.
On 1/4/22 12:14 AM, Chuck Guzis via cctalk wrote:
Seymour Cray, along with Bill Norris and Jim Thornton
and others
left Remington Rand/UNIVAC after Rand bought the near-bankrupt ERA.
Apparently, the work environment at Rand was felt to be stifling.
Norris had all of the Navy connections and was a great marketer,
so bringing some of Rand's engineering talent along was a natural.
Interesting.
I guess I thought that since Seymour left CDC to form Cray Research,
that meant that he was more of an employee at CDC and had less influence
on how it operated as a company. I would have assumed that someone that
was a founder would have had more influence and tried to improve things
before splitting off and forming yet another new company.
--
Grant. . . .
unix || die
8:42 a.m.
On Jan 4, 2022, at 10:40 AM, Grant Taylor via cctalk
<cctalk at classiccmp.org> wrote:
On 1/4/22 12:14 AM, Chuck Guzis via cctalk wrote:
Being a founder doesn't necessarily help much if the company gets big and chairwarmers
take control. It's one thing if you're a chairwarmer yourself and rise to CEO,
but if you're a top engineer you may be in a non-control position -- especially back
then when managers were managers and engineers just did what they were told.
For another example, consider Steve Jobs, who didn't even leave Apple voluntarily.
paul
Seymour Cray, along with Bill Norris and Jim
Thornton and others left Remington Rand/UNIVAC after Rand bought the near-bankrupt ERA.
Apparently, the work environment at Rand was felt to be stifling. Norris had all of the
Navy connections and was a great marketer, so bringing some of Rand's engineering
talent along was a natural.
Interesting.
I guess I thought that since Seymour left CDC to form Cray Research, that meant that he
was more of an employee at CDC and had less influence on how it operated as a company. I
would have assumed that someone that was a founder would have had more influence and tried
to improve things before splitting off and forming yet another new company.
8:44 a.m.
On 1/4/22 9:40 AM, Grant Taylor via cctalk wrote:
On 1/4/22 12:14 AM, Chuck Guzis via cctalk wrote:
There are PLENTY of examples of the sole founder being
either forced out by the board or shunted aside.
See James Ryder or Alan Shugart, who both were forced out,
sued to prevent them from using their own names, but then
went into competition with their old companies and ran them
out of business.
Jon
Seymour Cray, along with Bill Norris and Jim
Thornton and
others left Remington Rand/UNIVAC after Rand bought the
near-bankrupt ERA. Apparently, the work environment at
Rand was felt to be stifling. Norris had all of the Navy
connections and was a great marketer, so bringing some of
Rand's engineering talent along was a natural.
Interesting.
I guess I thought that since Seymour left CDC to form Cray
Research, that meant that he was more of an employee at
CDC and had less influence on how it operated as a
company.? I would have assumed that someone that was a
founder would have had more influence and tried to improve
things before splitting off and forming yet another new
company.
10:26 a.m.
On 1/4/22 7:40 AM, Grant Taylor via cctalk wrote:
I guess I thought that since Seymour left CDC to form
Cray Research,
that meant that he was more of an employee at CDC and had less influence
on how it operated as a company.? I would have assumed that someone that
was a founder would have had more influence and tried to improve things
before splitting off and forming yet another new company.
My recollection was over the CDC 8600 project--there is a preliminary
document over at bitsavers:
http://bitsavers.org/pdf/cdc/8600/Preliminary8600RefMan_Aug72.pdf
Sort of a 4-processor 64-bit 7600, but with some important differences,
such as discarding the notion of A and B registers and relying on 16
64-bit registers for everything. P-relative branches and calls and
out-of-user-space "library" routine use.
Apparently, the sharp pencil people were unwilling to fund the thing
fully and Seymour left (with the blessings of CDC which owned stock in
Cray Research) to do his own thing, along with selection of a few key
CDC personnel to help staffing. Afterwards, Cray was forbidden to
recruit from CDC ranks for a number of years.
At the time, the super-duper-computer funding was being funneled into
Jim Thornton's STAR-100 project. You could see a definite difference of
philosophy--the STAR was a virtual-memory vector machine (eventually
became the ETA-10) with scalar subset facilities, whereas the 8600 was a
real-memory fast scalar architecture with no vector facilities. In the
end, the Cray-1 wound up with some vector instructions and registers.
Funding a system like the Cray I or STAR (or any other supercomputer of
the day) required a huge cash commitment and CDC was unwilling to fund
*two* such projects.
--Chuck
5 Jan
5 Jan
2:25 a.m.
tis 2022-01-04 klockan 08:40 -0700 skrev Grant Taylor via cctalk:
On 1/4/22 12:14 AM, Chuck Guzis via cctalk wrote:
CISCO ?
Bosack and Lerner ?
And after getting kicked out by the venture capitalists sold of all
their remaining shares in cisco.
Seymour Cray, along with Bill Norris and Jim
Thornton and others
left Remington Rand/UNIVAC after Rand bought the near-bankrupt ERA.
Apparently, the work environment at Rand was felt to be stifling.
Norris had all of the Navy connections and was a great marketer,
so bringing some of Rand's engineering talent along was a natural.
Interesting.
I guess I thought that since Seymour left CDC to form Cray Research,
that meant that he was more of an employee at CDC and had less
influence
on how it operated as a company.? I would have assumed that someone
that
was a founder would have had more influence and tried to improve
things
before splitting off and forming yet another new company.
4 Jan
4 Jan
4:05 a.m.
There's a photo on twitter:
https://twitter.com/DonaldM38768041/status/1215804561333473280/photo/2
showing a guy standing before an open one at Fermilab.
In of the pictures are shown some very handy tape and disk pack holders on wheels. I never
saw such fancy holders. Would be great to come across one of those - or better so some
myself.
Cheers,
Pierre
-----------------------------------------------------------------------------
http://www.digitalheritage.de
9:43 a.m.
On Jan 4, 2022, at 4:05 AM, P Gebhardt via cctalk <cctalk at classiccmp.org> wrote:
The tape cart is just a standard cart for taking tapes between the tape library room and
where the tape drives are located. I?ve seen an updated version made for DLT/SDLT/LTO
tapes.
The disk pack holder on the other hand is something I?ve never seen, I have to wonder if
it was custom, and how well it handled an uneven load.
Zane
There's a photo on twitter:
https://twitter.com/DonaldM38768041/status/1215804561333473280/photo/2
showing a guy standing before an open one at Fermilab.
In of the pictures are shown some very handy tape and disk pack holders on wheels. I
never saw such fancy holders. Would be great to come across one of those - or better so
some myself.
Cheers,
Pierre 
3 Jan
3 Jan
2:44 a.m.
-----Original Message-----
From: cctalk <cctalk-bounces at classiccmp.org> On Behalf Of ben via cctalk
Sent: 03 January 2022 02:00
To: cctalk at classiccmp.org
Subject: Re: VAX 780 on eBay
On 2022-01-02 6:28 p.m., Zane Healy via cctalk wrote:
But can the Pi handle a gazillion students all time sharing at once @ 2400?
How long was the VAX timesharing era as I suspect networked PC's come out
soon after that.
Ben.
Ben,
Well if I remember properly the VAX couldn't at the time. I arranged for some folks
from the place I worked to go on a VMS course at Liverpool University.
They came back cursing the VAX and VMS and the sluggish terminal response times and the
dreaded "type behind".
With a few students on the VAX was great but get a decent load on and it crawled.
The trouble was that VMS was usually used with the VAX handling character editing, so the
scenario went something like this....
The user types an "a"...
The line driver says "what on earth do I do with this "a"? I know I'll
send it to the user....
.. oh dear the users code is paged out...
... oh dear there are no free pages so I have to page out another user....
.. and page in the users.....
.. and send it the "a" ....
.. the user code says "oh and "a" has arrived, what do I do , I know I will
echo it to the terminal and go to sleep....
.. in the mean time some one on the next terminal has typed a "B"....
The staff came back and said "how do we get a proper mainframe; we need an IBM with
VM" where the editing is handled in the screen controller.....
... so we asked for a large amount of disk storage, allocated a small room and they got an
IBM 4381 as at the time the VAX didn't have disks as big as the 3380-3 .....
Dave
On Jan 2, 2022, at 5:20 PM, Nigel Johnson Ham via
cctalk
<cctalk at classiccmp.org> wrote:
?I've always felt that in terms of performance the proper measure
would have been VUPS/kW :-)
That way my little MicroVAX 2 would be able to hold its head high!
On that note a Raspberry Pi 2b running SIMH/VAX is about 1.6 VUPS.
Zane
3:44 a.m.
On Sun, Jan 02, 2022 at 06:59:47PM -0700, ben via cctalk wrote:
Probably about as well as the actual VAXen back when I was failing my
degree. I remember trying (IIRC) "SET PRI" once to lower my session priority
one notch, and while I did eventually get a DCL prompt back, it was
basically unusable. It was hardly a speed demon even at default priority,
and I returned to my Amiga to get anything useful done.
Raspberry Pis have between zero and six serial ports, depending on model,
amount of faffing about, and whether you consider three-wire LVTTL to be a
serial port or demand a full seven-wire RS232 port. Beyond that, you're
going to have to multiplex connections over some other port, with Ethernet
being by far the most obvious choice.
Meanwhile, back at the university in the mists of time, they had rooms full
of dumb terminals, mostly VT220s but some VT420s, connected to an X.25 PAD
which then winged its way to the VAX cluster, and indeed onwards to JANET,
so it's not "cheating" to multiplex over a packed-switched WAN connection
on
the Pi. I expect the VAX's directly-connected serial port(s) would have been
reserved for the operator console.
The Pi 4 is between three and ten times faster than the Pi 2, depending on
task. I'd expect SIMH to fall towards the bottom of that range, but 5 VUPs
isn't to be sniffed at, especially as that is presumably a per-core
limitation and you could run a cluster of four of them. On a device which
draws ~7W at full tilt.
On 2022-01-02 6:28 p.m., Zane Healy via cctalk wrote:
[...]
On that note a
Raspberry Pi 2b running SIMH/VAX is about 1.6 VUPS.
But can the Pi handle a
gazillion students all time sharing at once @
2400? How long was the VAX timesharing era as I suspect networked PC's
come out soon after that.
6:33 p.m.
Microvax 2 is .9 vup.
C
On January 2, 2022 9:10:39 PM EST, Grant Taylor via cctalk <cctalk at
classiccmp.org> wrote:
On 1/2/22 6:18 PM, Zane Healy via cctalk wrote:
--
Sent from my Android device with K-9 Mail. Please excuse my brevity.
VAX-11/780 = 1 VUP
VAX-11/782 = 1.8 VUP
MicroVAX 1 = 0.3 VUP
MicroVAX 2 = 0.7 VUP
Thank you Zane.
--
Grant. . . .
unix || die 
6:36 p.m.
On 2022-01-02 9:33 p.m., Chris Zach via cctalk wrote:
Microvax 2 is .9 vup.
Oops, Zane vupped up.
C
On January 2, 2022 9:10:39 PM EST, Grant Taylor via cctalk <cctalk at
classiccmp.org> wrote:
On 1/2/22 6:18 PM, Zane Healy via cctalk wrote:
VAX-11/780 = 1 VUP
VAX-11/782 = 1.8 VUP
MicroVAX 1 = 0.3 VUP
MicroVAX 2 = 0.7 VUP
Thank you Zane.
--
Grant. . . .
unix || die
8:35 p.m.
The worst part is I know it?s 0.9 VUPS.
Zane
On Jan 2, 2022, at 6:36 PM, Toby Thain via cctalk
<cctalk at classiccmp.org> wrote:
?On 2022-01-02 9:33 p.m., Chris Zach via cctalk wrote:
Microvax 2 is .9 vup.
Oops, Zane vupped up.
C
> On January 2, 2022 9:10:39 PM EST, Grant Taylor via cctalk <cctalk at
classiccmp.org> wrote:
> On 1/2/22 6:18 PM, Zane Healy via cctalk wrote:
>> VAX-11/780 = 1 VUP
>> VAX-11/782 = 1.8 VUP
>> MicroVAX 1 = 0.3 VUP
>> MicroVAX 2 = 0.7 VUP
>
> Thank you Zane.
>
>
>
> --
> Grant. . . .
> unix || die
9:10 p.m.
Maybe you were thinking the 11/73? I think the 83 was quicker, and the
93/94 hit the 1 VUP speed...
C
On 1/2/2022 11:35 PM, Zane Healy wrote:
The worst part is I know it?s 0.9 VUPS.
Zane
On Jan 2, 2022, at 6:36 PM, Toby Thain via cctalk
<cctalk at classiccmp.org> wrote:
?On 2022-01-02 9:33 p.m., Chris Zach via cctalk wrote:
Microvax 2 is .9 vup.
Oops, Zane vupped up.
C
> On January 2, 2022 9:10:39 PM EST, Grant Taylor via cctalk <cctalk at
classiccmp.org> wrote:
> On 1/2/22 6:18 PM, Zane Healy via cctalk wrote:
>> VAX-11/780 = 1 VUP
>> VAX-11/782 = 1.8 VUP
>> MicroVAX 1 = 0.3 VUP
>> MicroVAX 2 = 0.7 VUP
>
> Thank you Zane.
>
>
>
> --
> Grant. . . .
> unix || die
9:18 p.m.
I had accounts on a MicroVAX 2 and a VAX 11/750. The microvax was faster
for most compute jobs, but the 750 with 1/4 the memory handled more users
mostly in text editors with the occasional compile or nroff/troff jobs.
IIRC, the 750 had faster disks...
Warner
On Sun, Jan 2, 2022, 9:35 PM Zane Healy via cctalk <cctalk at classiccmp.org>
wrote:
The worst part is I know it?s 0.9 VUPS.
Zane
On Jan 2, 2022, at 6:36 PM, Toby Thain via cctalk
<cctalk at classiccmp.org>
wrote:
?On 2022-01-02 9:33 p.m., Chris Zach via cctalk wrote:
cctalk at
classiccmp.org> wrote:
Microvax 2 is .9 vup.
Oops, Zane vupped up.
> C
>> On January 2, 2022 9:10:39 PM EST, Grant Taylor via cctalk < > On
1/2/22 6:18 PM, Zane Healy via cctalk wrote:
>> VAX-11/780 = 1 VUP
>> VAX-11/782 = 1.8 VUP
>> MicroVAX 1 = 0.3 VUP
>> MicroVAX 2 = 0.7 VUP
>
> Thank you Zane.
>
>
>
> --
> Grant. . . .
> unix || die
3 Jan
3 Jan
8:49 a.m.
On Jan 2, 2022, at 9:18 PM, Warner Losh <imp at bsdimp.com> wrote:
I had accounts on a MicroVAX 2 and a VAX 11/750. The microvax was faster for most compute
jobs, but the 750 with 1/4 the memory handled more users mostly in text editors with the
occasional compile or nroff/troff jobs. IIRC, the 750 had faster disks...
Warner
I picked up a copy of ?VAX I/O Subsystems: Optimizing Performance?, it gives I/O speeds
for SDI and DSSI disks, I don?t doubt that the VAX-11/750 had faster disks if it had RA81
or especially RA82 disks.
Here we go, from the HP "OpenVMS I/O Users Reference Manual? for 7.3-1.
"On VAX systems, the RD53 and RD54 are 5.25-inch, full-height, Winchester-type drives
with average access time of 38 ms and a data transfer rate of 0.625 MB per second.?
Even the RA80 can beat that, basically any SDI disk on a VAX-11/750 would be noticeably
faster. There is a reason why MicroVAX I and II systems with RD53/54 drives were
considered to be good for Performance Tuning training. It?s really easy to notice
improvements.
Zane
8:50 a.m.
On Mon, Jan 3, 2022 at 12:18 AM Warner Losh via cctalk
<cctalk at classiccmp.org> wrote:
I had accounts on a MicroVAX 2 and a VAX 11/750. The
microvax was faster
for most compute jobs, but the 750 with 1/4 the memory handled more users
mostly in text editors with the occasional compile or nroff/troff jobs.
IIRC, the 750 had faster disks...
I'll agree with that. We used to run 40-50 users on our 8MB 11/750
(with both CMI and Unibus disk) but it did do some swapping over 8-10
users.
The MicroVAX II was a pretty snappy single-user machine, but was a big
sluggish over 4 users. Our big compute jobs were running compilers.
Most of our user load was VMS MAIL, EDT, MASS-11...
-ethan
9:29 a.m.
On 1/3/22 11:50 AM, Ethan Dicks via cctalk wrote:
On Mon, Jan 3, 2022 at 12:18 AM Warner Losh via
cctalk
<cctalk at classiccmp.org> wrote:
You obviously didn't use the Ada compiler. :-)
Or Eunice, for that matter.
bill
I had accounts on a MicroVAX 2 and a VAX 11/750.
The microvax was faster
for most compute jobs, but the 750 with 1/4 the memory handled more users
mostly in text editors with the occasional compile or nroff/troff jobs.
IIRC, the 750 had faster disks...
I'll agree with that. We used to run 40-50 users on our 8MB 11/750
(with both CMI and Unibus disk) but it did do some swapping over 8-10
users.
10:19 a.m.
On Mon, Jan 3, 2022 at 12:29 PM Bill Gunshannon via cctalk
<cctalk at classiccmp.org> wrote:
On 1/3/22 11:50 AM, Ethan Dicks via cctalk wrote:
Hell no!
90% of our work was in C, the rest in m68k assembler (our own), VAX
MACRO, or FORTRAN. We started off with Whitesmith's C then by the
late 80s also used DEC's VAX C.
Compiling all the code for our product took the 11/750 6 hours to
compile and link plus an additional 2 hours for an 11/730 to link
under a different version of VMS. 8 hours total to rebuild totally
I'll agree with that. We used to run 40-50
users on our 8MB 11/750
(with both CMI and Unibus disk) but it did do some swapping over 8-10
users.
You obviously didn't use the Ada compiler. :-)
Or Eunice, for that matter. from source. Some things about the good old days
weren't so good.
-ethan
10:40 a.m.
On Jan 3, 2022, at 10:19 AM, Ethan Dicks via cctalk
<cctalk at classiccmp.org> wrote:
On Mon, Jan 3, 2022 at 12:29 PM Bill Gunshannon via cctalk
<cctalk at classiccmp.org> wrote:
At my current $WORK it takes a 40 core linux box with 200GB of RAM
22 hours to rebuild our source tree from scratch.
So for some of us it hasn?t gotten any better over the years.
David
On 1/3/22 11:50 AM, Ethan Dicks via cctalk
wrote:
Hell no!
90% of our work was in C, the rest in m68k assembler (our own), VAX
MACRO, or FORTRAN. We started off with Whitesmith's C then by the
late 80s also used DEC's VAX C.
Compiling all the code for our product took the 11/750 6 hours to
compile and link plus an additional 2 hours for an 11/730 to link
under a different version of VMS. 8 hours total to rebuild totally
from source. Some things about the good old days weren't so good.
-ethan I'll agree with that. We used to run 40-50
users on our 8MB 11/750
(with both CMI and Unibus disk) but it did do some swapping over 8-10
users.
You obviously didn't use the Ada compiler. :-)
Or Eunice, for that matter. 
10:47 a.m.
On 2022-01-03 13:19, Ethan Dicks via cctalk wrote:
Compiling all the code for our product took the 11/750
6 hours to
compile and link plus an additional 2 hours for an 11/730 to link
under a different version of VMS. 8 hours total to rebuild totally
from source. Some things about the good old days weren't so good.
Disagree here ;-)
Before you hit compile, you switched your brain on, and thought for a
while what you did.
When I watch this kids in the office now pressing compile after each
character they typed, ....
11:37 a.m.
Hit Compile?
In my first job, it was triggered by the EOJ card!
Nigel Johnson, MSc., MIEEE, MCSE VE3ID/G4AJQ/VA3MCU
Amateur Radio, the origin of the open-source concept!
Skype: TILBURY2591nw.johnson at ieee.org
On 2022-01-03 13:47, emanuel stiebler via cctalk wrote:
On 2022-01-03 13:19, Ethan Dicks via cctalk wrote:
Compiling all the code for our product took the
11/750 6 hours to
compile and link plus an additional 2 hours for an 11/730 to link
under a different version of VMS.? 8 hours total to rebuild totally
from source.? Some things about the good old days weren't so good.
Disagree here ;-)
Before you hit compile, you switched your brain on, and thought for a
while what you did.
When I watch this kids in the office now pressing compile after each
character they typed, ....
11:41 a.m.
On 2022-01-03 2:37 p.m., Nigel Johnson Ham via cctalk wrote:
Hit Compile?
In my first job, it was triggered by the EOJ card!
You had a COMPILER?
As far as I can see, these days the state of the art is to crash at
runtime for as many different stupid reasons as possible.
--T
(get off my lawn!)
Nigel Johnson, MSc., MIEEE, MCSE VE3ID/G4AJQ/VA3MCU
Amateur Radio, the origin of the open-source concept!
Skype:? TILBURY2591nw.johnson at ieee.org
On 2022-01-03 13:47, emanuel stiebler via cctalk wrote:
> On 2022-01-03 13:19, Ethan Dicks via cctalk wrote:
>
>> Compiling all the code for our product took the 11/750 6 hours to
>> compile and link plus an additional 2 hours for an 11/730 to link
>> under a different version of VMS.? 8 hours total to rebuild totally
>> from source.? Some things about the good old days weren't so good.
>
> Disagree here ;-)
> Before you hit compile, you switched your brain on, and thought for a
> while what you did.
>
> When I watch this kids in the office now pressing compile after each
> character they typed, ....
>
11:46 a.m.
Only F*tran, which we used for predicting race horse results on the
midnight shift. The enterprise app was written in ART418 assembler.
I'm having lots of fun with the ESDI drives, got one running RT11 on the
11/73, and one running NetBSD on the MicroVAX, but then tried the two I
got from Jeremy and it recognised them and wouldn't write to them. Then
the controller got a total internal failure and so now I only have one
machine working!
cheers,
Nigel
Nigel Johnson, MSc., MIEEE, MCSE VE3ID/G4AJQ/VA3MCU
Amateur Radio, the origin of the open-source concept!
Skype: TILBURY2591nw.johnson at ieee.org
On 2022-01-03 14:41, Toby Thain wrote:
On 2022-01-03 2:37 p.m., Nigel Johnson Ham via cctalk
wrote:
Hit Compile?
In my first job, it was triggered by the EOJ card!
You had a COMPILER?
As far as I can see, these days the state of the art is to crash at
runtime for as many different stupid reasons as possible.
--T
(get off my lawn!)
Nigel Johnson, MSc., MIEEE, MCSE
VE3ID/G4AJQ/VA3MCU
Amateur Radio, the origin of the open-source concept!
Skype: TILBURY2591nw.johnson at ieee.org
On 2022-01-03 13:47, emanuel stiebler via cctalk wrote:
> On 2022-01-03 13:19, Ethan Dicks via cctalk wrote:
>
>> Compiling all the code for our product took the 11/750 6 hours to
>> compile and link plus an additional 2 hours for an 11/730 to link
>> under a different version of VMS.? 8 hours total to rebuild totally
>> from source.? Some things about the good old days weren't so good.
>
> Disagree here ;-)
> Before you hit compile, you switched your brain on, and thought for
> a while what you did.
>
> When I watch this kids in the office now pressing compile after each
> character they typed, ....
>
11:56 a.m.
On Mon, Jan 3, 2022 at 11:47 AM emanuel stiebler via cctalk <
cctalk at classiccmp.org> wrote:
On 2022-01-03 13:19, Ethan Dicks via cctalk wrote:
Compilers today tend to still suffer from cascading error problems... I
usually fix as many as I can before that starts to set in and then rebuild.
But I must say that computers are sometimes much better at finding
bugs and problems than I am, even when I'm thinking. Better to let
automation fix the easy stuff so I can reserve the thinking time for
really hard bugs...
Warner
Compiling all the code for our product took the
11/750 6 hours to
compile and link plus an additional 2 hours for an 11/730 to link
under a different version of VMS. 8 hours total to rebuild totally
from source. Some things about the good old days weren't so good.
Disagree here ;-)
Before you hit compile, you switched your brain on, and thought for a
while what you did.
When I watch this kids in the office now pressing compile after each
character they typed, ....
2 Jan
2 Jan
9:11 p.m.
Yup. The 8600's (UMBC1 and 2) were connected to Bitnet, and I was a
little Bitnaught until I found AI. It might have connected through the
Gandalf network or something else, I'd have to check old printouts.
Can I ask for a rough translation in to comparative
VAX Units of
Performance (VUPs)?? I /think/ that the VAX 11/780 was 1 VUP.
Kind of like MIPS, but I think they were going for throughput. The 780
may have had the same CPU speed as the Microvax II, but anytime you hit
disk on the MV you were slugging down to the Q-Bus speed, going through
a 4mb memory map window, and the RQDX3 wasn't either fast or very good IMO.
The 780 would plow through the MASSBUS channel interfaces directly
attached to memory (well, on a 780 everything went SBI) and
RM03/RM80/RP07 drives were way faster than RD54's. And you could go to
CI bus to talk to an HSC50 which could do more interesting stuff with
RA81's and 82's.
3 Jan
3 Jan
9:01 a.m.
On 1/2/22 7:11 PM, Grant Taylor via cctalk wrote:
By definition, the /780 was 1.0 VUP.? The KA630 (MicroVAX
II) was supposed to be 0.8 VUP.? It beat the /780 on
floating point, supposedly.
Jon
but the 780 was really slow. I have a Microvax II
here
that would equal it CPU speed-wise, and the running of
the main memory on the SBI was cool for the 782 option
but was pretty slow.
Can I ask for a rough translation in to comparative VAX
Units of Performance (VUPs)?? I /think/ that the VAX
11/780 was 1 VUP.
2 Jan
2 Jan
1:46 p.m.
On 1/2/22 12:16 PM, Jonathan Chapman via cctalk wrote:
In North America? Good luck! Part of what got me to
buy a
smallish industrial building was needing three phase. It's usually
cheaper/easier to either use a VFD if it's for motor equipment, or a
rotary converter, than trying to get the power company to give you
three phase. Even if it's literally on the pole behind your house,
they want $LOL to make the connection.
So I've gathered. Commercial, industrial, and rural farm tend to be the
places where 3? is a viable option. Beyond that, ya rotary converter.
The more that I learn about VFDs, the less that I like them. Especially
for anything electronic.
Yes, these "melted the socket, never tripped the
breaker" failures
are a result of crap receptacles. On basically all good brands of
receptacles, the 15A and 20A have the same internal parts, anyway. It's
my opinion that anything in a shop area, or really even the kitchen,
ought to be spec-grade receptacles. They're not that expensive when you
consider they'll last a lifetime. We always require Hubble spec-grade,
and that's what I've installed at the house, too.
I'd much rather spend the additional money and have what you are
referring to as Hubble spec-grade outlets throughout the house.
I've gotten to the point that I don't want to put up with / tolerate
sub-par things that are going to cause me to have more work in the future.
Indeed, you can have a continuous load up to 12A with
no special NEC
rules on a regular old branch circuit.
:-)
--
Grant. . . .
unix || die
1 Jan
1 Jan
10:49 a.m.
this is the first one I recall seeing for sale
in the classic computer era (versus several -11/70's, /40s, etc)
Last one that went auction-style on eBay went for $1,178.00, local pick-up. I know the
buyer.
I still wouldn't think $5K is totally unreasonable, given the clean condition and that
they're willing to freight it out.
Interesting, but the argument for why it's not
tested is implausible which makes me very suspicious.
Indeed, Tuscon Buying Group deals in a lot of industrial stuff -- we buy from them at
$day_job sometimes. I'd be really surprised if they don't have 240V available for
testing somewhere in their shop.
Thanks,
Jonathan
10:59 a.m.
On 1/1/22 11:49 AM, Jonathan Chapman via cctalk wrote:
Indeed, Tuscon Buying Group deals in a lot of
industrial stuff --
we buy from them at $day_job sometimes. I'd be really surprised if
they don't have 240V available for testing somewhere in their shop.
There's a difference in having something /in/ /shop/ vs the shop selling
it for someone when it's located elsewhere.
Where we see it, the background indicates that the VAX is located in a
residential neighborhood. Or at least was when the pictures were taken.
--
Grant. . . .
unix || die
10:49 a.m.
On 1/1/2022 12:40 PM, Paul Koning via cctalk wrote:
Possibly part of the same installation they have:
TU77: https://www.ebay.com/itm/275083502085 (The 11/780 is visible in the background of
the first picture)
RM05: https://www.ebay.com/itm/284587865252
LA120: https://www.ebay.com/itm/275084461044
Prices seem on the high end but not totally unreasonable.? There don't seem to be any
RM05 packs included with the drive.
--
John H. Reinhardt
On Jan 1, 2022, at 1:12 PM, Noel Chiappa via
cctalk <cctalk at classiccmp.org> wrote:
This:
https://www.ebay.com/itm/275084268137
The starting price is expensive, but probably not utterly unreasonable,
given that:
- the 780 was the first VAX, and thus historically important
- 780's are incredibly rare; this is the first one I recall seeing for sale
in the classic computer era (versus several -11/70's, /40s, etc)
- this one appears to be reasonably complete; no idea if all the key CPU
boards are included, but it's things like the backplane, etc (all of which
seem to be there) which would be completely impossible to find now - if any
boards _are_ missing, there's at least the _hope_ that they can be located
(780 boards seem to come by every so often on eBait), since people seem to
keep boards, not realizing that without the other bits they are useless
Interesting, but the argument for why it's not tested is implausible which makes me
very suspicious. I suppose there might be a few American homes that have only 110 volt
power, but I'm hard pressed to think of any I have ever seen, and that includes really
old houses.
paul
11:04 a.m.
On 1/1/22 11:49 AM, John H. Reinhardt via cctalk wrote:
TU77: https://www.ebay.com/itm/275083502085 (The
11/780 is visible in
the background of the first picture)
The TU77 has a 3? power plug on it. That probably explains why testing
it would be problematic in a residential settings.
1?240V is vastly different than 3?120/208V.
The picture after the plug also looks remarkably like the back plane of
the VAX.
--
Grant. . . .
unix || die
12:49 p.m.
The TU77 has a 3? power plug on it.? That probably
explains why testing
it would be problematic in a residential settings.
A TU77 can run on household 240v. If I recall it used the 240 for the
big blower motor and the rest of the stuff ran on 110.
1?240V is vastly different than 3?120/208V.
1 phase 240 is just 2 120 volts out of phase. Looking at the 780 it has
3-5 power supplies in it, each one running on 120v. My guess is the main
fan blower is 3 phase 208, and each of the power supplies is run off a
208-common leg (giving 120v).
Big ole monster, but if you can put a 240v motor on the blower you
should be able to rewire it to run on a couple of household 240 circuits.
C
(Ran a KS10, TU77, and 2 RM03's in a spare bedroom once. It has...
upgraded wiring)
12:46 p.m.
TU77: https://www.ebay.com/itm/275083502085 (The
11/780 is visible in
That looks complete with the TM03 formatter.
That
looks like it's missing the MassBus adapter. I think two RM05's
shared a cabinet between them that contained the MBA logic. Not sure if
a RM03's MBA will work, quite possible (I have an RM80 that is on top of
a RM03 controller MBA)
CZ
1:19 p.m.
On 1/1/22 10:40 AM, Paul Koning via cctalk wrote:
Without replacing the power controller in the 11/780, you need 208v
3-phase to run it.? It's not impossible...nothing in the CPU actually
*needs* 3-phase as the individual power supplies are 120v but the
overall maximum load is greater than a 30A 120v circuit.
TTFN - Guy
On Jan 1, 2022, at 1:12 PM, Noel Chiappa via
cctalk <cctalk at classiccmp.org> wrote:
This:
https://www.ebay.com/itm/275084268137
The starting price is expensive, but probably not utterly unreasonable,
given that:
- the 780 was the first VAX, and thus historically important
- 780's are incredibly rare; this is the first one I recall seeing for sale
in the classic computer era (versus several -11/70's, /40s, etc)
- this one appears to be reasonably complete; no idea if all the key CPU
boards are included, but it's things like the backplane, etc (all of which
seem to be there) which would be completely impossible to find now - if any
boards _are_ missing, there's at least the _hope_ that they can be located
(780 boards seem to come by every so often on eBait), since people seem to
keep boards, not realizing that without the other bits they are useless
Interesting, but the argument for why it's not tested is implausible which makes me
very suspicious. I suppose there might be a few American homes that have only 110 volt
power, but I'm hard pressed to think of any I have ever seen, and that includes really
old houses. 
4 Jan
4 Jan
2:17 p.m.
New subject: 3phase power for VAXen [was Re: VAX 780 on eBay]
Date: Sat, 1 Jan 2022 13:19:51 -0800
From: Guy Sotomayor via cctalk <cctalk at classiccmp.org>
On 1/1/22 10:40 AM, Paul Koning via cctalk wrote:
>> On Jan 1, 2022, at 1:12 PM, Noel Chiappa via
cctalk <cctalk at classiccmp.org> wrote:
>> This:
>> The starting price is expensive, but probably
not utterly unreasonable,
>> given that:
>> - the 780 was the first VAX, and thus
historically important
>> - 780's are incredibly rare; this is the
first one I recall seeing for sale
>> in the classic computer era (versus several -11/70's, /40s, etc)
>> - this one appears to be reasonably complete;
no idea if all the key CPU
>> boards are included, but it's things like the backplane, etc (all of which
>> seem to be there) which would be completely impossible to find now - if any
>> boards _are_ missing, there's at least the _hope_ that they can be located
>> (780 boards seem to come by every so often on eBait), since people seem to
>> keep boards, not realizing that without the other bits they are useless
> Interesting, but the argument for why it's not
tested is implausible which
> makes me very suspicious. I suppose there might be a few American homes
> that have only 110 volt power, but I'm hard pressed to think of any I have
> ever seen, and that includes really old houses.
Without replacing the power controller in the 11/780,
you need 208v 3-phase
to run it.? It's not impossible...nothing in the CPU actually *needs* 3-phase
as the individual power supplies are 120v but the overall maximum load is
greater than a 30A 120v circuit.
TTFN - Guy
I've been reading this thread for the last few days, without the time to reply.
All the statutes of limitations have run out, so I can tell the story; it will
be clear shortly why I'm piggybacking on Guy's post.
Back in the mists of time, Paul Allen wanted me to acquire a VAX-11/780 for his
collection. Shortly after that request landed in my inbox, the DECUS DFWLUG
announced that they would not be opening their VAX museum due to the untimely
passing of the gentleman who was driving the effort, and that they would be
disposing of the collection. I contacted the person who was handling the
deaccession, but he would not discuss it with me because someone else had
already arranged to take the whole collection.
That person was Guy Sotomayor.
Guy sold Paul two 11/785 systems (one an upgrade, with the 780-5 replacement
label!) in chassis, with a third full set of boards as spares. I flew down to
the Bay Area and had lunch with Guy, saw his DEC-10 and all that, and arranged
for the shipping.
Shortly after that, Paul floated the idea of turning the collection into an
actual museum. (At the time, the project consisted of me and an electrical
engineer named Keith Perez, who devoted his spare time to helping me keep
Paul's big iron running. Keith was building the digital control system for
Paul's submarine at the time.) The project, an online "museum" called
PDPplanet, changed its name to Living Computer Museum; we hired a third
engineer to help, Ian King, who was eminently qualified to get the VAXen up and
running based on his own private collection of VAXen and -11s in his basement.
We did not have a 3phase outlet on the second floor of the building where the
collection was housed, and there was no room for the first VAX in the small
computer room on the third floor, so Ian and Keith came up with an alternative:
They tested all the outlets on the second floor and determined that there were
three within reach of the room in which Ian was going to work on the 785 which
were fed from 3 different phases off the big honking breaker panel (200A
service, IIRC). Keith put together a box with the appropriate NEMA socket and
three heavy duty cords feeding into it, which in turn were plugged into three
outlets on the walls around what eventually became the vintage exhibition hall
at Living Computers: Museum + Labs (the eventual name of the place after the
modern exhibit space on the first floor was built).
So it's possible to power a 780 or 785 without a power supply rebuild if you
simply have the right (industrial) breaker panel in your building...
Happy New Year, everybody!
Rich
2:44 p.m.
New subject: 3phase power for VAXen [was Re: VAX 780 on eBay]
On Tue, 2022-01-04 at 17:17 -0500, Rich Alderson via cctalk wrote:
Paul Allen wanted me to acquire a VAX-11/780 for his
collection
John Zabolitzky has an operating VAX -- I don't know the vintage --
eleven cabinets, in his collection in Neubiberg, a southeastern suburb
of Munich.
He also has a Cyber 180, Cyber 960, Cray Y-ML EL, Cray T3E, NEC SX-6,
IBM 705, ....
Photos
at?https://photos.google.com/share/AF1QipPGeVphL95gtsGIO96iU4APdtSRnHhcm1y4…
3:44 p.m.
New subject: 3phase power for VAXen [was Re: VAX 780 on eBay]
Date: Tue, 04 Jan 2022 14:44:00 -0800
From: Van Snyder via cctalk <cctalk at classiccmp.org>
On Tue, 2022-01-04 at 17:17 -0500, Rich Alderson via
cctalk wrote:
> Paul Allen wanted me to acquire a VAX-11/780 for his
> collection
John Zabolitzky has an operating VAX -- I don't
know the vintage --
eleven cabinets, in his collection in Neubiberg, a southeastern suburb
of Munich.
He also has a Cyber 180, Cyber 960, Cray Y-ML EL, Cray
T3E, NEC SX-6,
IBM 705, ....
Yes, I know about John's collection. Thanks.
I suppose I should have pointed out that the story posted took place about 15
years ago, when we were all younger and healthier and living...
Rich
3:18 p.m.
New subject: 3phase power for VAXen [was Re: VAX 780 on eBay]
As the owner of a VAX-11/780 I can confirm what Guy says about the 866
power controller. This is the only component that needs 3-phase power.
There is a small 3-phase transformer inside that provides 24V DC to the
control circuitry. The power controller not only interfaces to the DEC
power control bus to switch on other cabinets (with a delayed output)
but also connects to airflow sensors above each blower. If the airflow
fails then it cuts power to the system. The blowers themselves are all
single phase.
I converted my power controller to run on single phase by replacing the
transformer (18VAC, 20VA). It's also important to note that the neutral
wire from the line filter to the output rail is undersized for running
the system on single phase. If the system were running on 3-phase then
the neutral current would be minimal but running from single phase means
ALL the current is going through neutral wire. I replaced this wire with
one of a suitable rating.
I am in the UK so the figures I give are for running a system which
would normally be on 415V 3-phase from a 240V single phase supply. For
110V you can probably just double the current figures. I have the FP780
option as well as 2x DW780 and RH780 controllers. To power everything in
the main cabinet takes about 16.5A. I run this all off a 20A circuit
breaker (B-Curve RCBO). I have to turn on the H7100 power supplies one
at a time. If I try to turn them all on at once the inrush will trip the
breaker.
There are two different versions on the H7100 in this system. The older
version uses a 3 ohm resistor, which is shorted by a relay after power
up, to limit the inrush. These have quite a high inrush current. The
later version uses a thermistor in place of the resistor which greatly
improves the situation.
Once you have the main cabinet powered then to get a usable system you
then need about another 4A to run the two BA11K Unibus boxes and
whatever power is required for a disk drive. In my case that's usually
an RA70 or RA72 so fairly minimal. I don't have any Massbus disks but I
do have a TU77 connected to the system.
The heating effect is noticeable after the system has been running for a
while so you may also have to consider air conditioning power requirements.
Matt
3:53 p.m.
New subject: 3phase power for VAXen [was Re: VAX 780 on eBay]
On 1/4/2022 5:17 PM, Rich Alderson via cctalk wrote:
I've been reading this thread for the last few
days, without the time to reply.
All the statutes of limitations have run out, so I can tell the story; it will
be clear shortly why I'm piggybacking on Guy's post.
Someday we will all gather around the bonfire, ringed by our black
helicopters and will share the tale of Doug's KA10 that ran in his
apartment unit.....
We did not have a 3phase outlet on the second floor of
the building where the
collection was housed, and there was no room for the first VAX in the small
computer room on the third floor, so Ian and Keith came up with an alternative:
They tested all the outlets on the second floor and determined that there were
three within reach of the room in which Ian was going to work on the 785 which
were fed from 3 different phases off the big honking breaker panel (200A
service, IIRC). Keith put together a box with the appropriate NEMA socket and
three heavy duty cords feeding into it, which in turn were plugged into three
outlets on the walls around what eventually became the vintage exhibition hall
at Living Computers: Museum + Labs (the eventual name of the place after the
modern exhibit space on the first floor was built).
It was far worse than that....
CZ
5:47 p.m.
New subject: 3phase power for VAXen [was Re: VAX 780 on eBay]
On Tue, Jan 4, 2022 at 5:17 PM Rich Alderson via cctalk <
cctalk at classiccmp.org> wrote:
We did not have a 3phase outlet on the second floor of the building where
the
collection was housed, and there was no room for the first VAX in the small
computer room on the third floor, so Ian and Keith came up with an
alternative:
They tested all the outlets on the second floor and determined that there
were
three within reach of the room in which Ian was going to work on the 785
which
were fed from 3 different phases off the big honking breaker panel (200A
service, IIRC). Keith put together a box with the appropriate NEMA socket
and
three heavy duty cords feeding into it, which in turn were plugged into
three
outlets on the walls around what eventually became the vintage exhibition
hall
at Living Computers: Museum + Labs (the eventual name of the place after
the
modern exhibit space on the first floor was built).
So it's possible to power a 780 or 785 without a power supply rebuild if
you
simply have the right (industrial) breaker panel in your building...
Happy New Year, everybody!
Rich
After looking at the 869-D diagrams and the 11/780 power diagram, this
makes sense. The power diagrams show 3 120VAC circuits. There is no phase
relationship required between the phases, EXCEPT in the 869-D itself. The 4
pole 30A circuit breaker will trip if any (L1, L2, L3, N) is overloaded. If
L1, L2, L3 are out of phase then the neutral doesn't carry much current,
but if they are all the same phase then the total limit will be 30A.
There are 4 connections L1, L2, L3 and N.
In the US you would connect the highest current draw leg to one side of the
120VAC and connect the other two legs to the other 120VAC phase and then
HOPE that it works.
The 869-D outlet are:
unswitched L1
switched L1
switched L2
switched L3
For me, if I can buy a Shelby Cobra, I buy it first and figure out where to
park it later.
Chuck D.
1 Jan
1 Jan
11:10 a.m.
On 1/1/22 12:12 PM, Noel Chiappa via cctalk wrote:
This:
https://www.ebay.com/itm/275084268137
The starting price is expensive, but probably not utterly unreasonable,
given that:
They don't have a front picture, but there is a rear picture
of a TU-77/TU-78 tape drive interior.
As for testing, the 780 came wired for 208/240 3 phase
power, with a big 5-pin plug.
The CPU cabinet seems to have 2 Massbus controllers, I
think. So, maybe there's a Massbus disk included.
Jon
11:31 a.m.
On 1/1/2022 1:10 PM, Jon Elson via cctalk wrote:
On 1/1/22 12:12 PM, Noel Chiappa via cctalk wrote:
In separate auctions they have a RM05 and a TU77
TU77: https://www.ebay.com/itm/275083502085 (The 11/780 is visible in the background of
the first picture)
RM05: https://www.ebay.com/itm/284587865252
--
John H. Reinhardt
This:
???? https://www.ebay.com/itm/275084268137
The starting price is expensive, but probably not utterly unreasonable,
given that:
They don't have a front picture, but there is a rear picture of a TU-77/TU-78
tape drive interior.
As for testing, the 780 came wired for 208/240 3 phase power, with a big 5-pin plug.
The CPU cabinet seems to have 2 Massbus controllers, I think. So, maybe there's a
Massbus disk included.
Jon
853
days inactive
857
days old
97 comments
33 participants
participants (33)
-
abuse＠cabal.org.uk -
bfranchuk＠jetnet.ab.ca -
bill.gunshannon＠hotmail.com -
cc＠alderson.users.panix.com -
cclist＠sydex.com -
cctalk＠gtaylor.tnetconsulting.net -
chd＠chdickman.com -
cz＠alembic.crystel.com -
dave.g4ugm＠gmail.com -
david＠kdbarto.org -
derschjo＠gmail.com -
drb＠msu.edu -
elson＠pico-systems.com -
emu＠e-bbes.com -
ethan.dicks＠gmail.com -
g4ajq1＠gmail.com -
ggs＠shiresoft.com -
healyzh＠avanthar.com -
imp＠bsdimp.com -
jnc＠mercury.lcs.mit.edu -
john＠forecast.name -
johnhreinhardt＠thereinhardts.org -
lists＠glitchwrks.com -
matt＠9track.net -
nf6x＠nf6x.net -
p.gebhardt＠ymail.com -
pat＠vax11.net -
paulkoning＠comcast.net -
spacewar＠gmail.com -
stefan.skoglund＠agj.net -
toby＠telegraphics.com.au -
van.snyder＠sbcglobal.net -
w2hx＠w2hx.com