13 Mar
2015
13 Mar
'15
11:32 a.m.
CHM was able to obtain volumes 18-20 of the IBM 2050 drawings, which are
the microcode charts and ROS dump. I got them scanned and uploaded yesterday
to http://bitsavers.org/pdf/ibm/360/fe/2050
This was one of the things that I had been trying to locate for a while now.
7 May
7 May
10:16 a.m.
On 03/13/2015 01:32 PM, Al Kossow wrote:
CHM was able to obtain volumes 18-20 of the IBM 2050
drawings, which are
the microcode charts and ROS dump. I got them scanned and
uploaded yesterday
to http://bitsavers.org/pdf/ibm/360/fe/2050
This was one of the things that I had been trying to
locate for a while now.
Just a curious thing I've noticed. Given the historic
significance, and the large number of machines produced, I'm
kind of amazed at the incredibly small number of 360's that
apparently exist.
Yes, I know, any would-be collector could drag home a PDP-8
and put it in his garage, even a whole rack mount system
with an RK02 (or 3), dectape and paper tape reader, and
still get his car in the garage. And, the system could be
run off normal mains power.
You can't do that with a real 360 (some 360/20's were pretty
small), even a 360/30 was a pretty big box. And, you can't
run a 360 off normal residential power, either. Many of the
peripherals used 3-phase motors, and hacking the
converter/inverter to run off single phase would not be a
task for any but the most experienced EE.
But, it sure is a shame that there appear to be a tiny
number of machines in existence. One list shows 15 or 16
machines, excluding the model 20. Probably there are a
couple more hidden somewhere, like the B1900 that came to
light so recently.
As fas as I can tell, NONE of these systems is complete
enough to ever run, with the possible exception of the
360/30 at the CHM, which does seem to have a complement of
peripherals, and maybe control units, too.
Given the number of DEC 10's that are actually up and
running, this seems a bit of a surprise. there might be
some emotional attachments that are behind this disparity.
Anybody have some comments?
Jon
10:33 a.m.
1. IBM would very often lease rather than sell, and take back machines when
customers upgraded; it was obviously in their interests to control or
eliminate where possible the market in used machines.
2. Gold. A lot of gold in old IBM kit. I knew a scrap dealer in Chelmsford
UK that did nothing but break old IBM mainframes (only ever big boring grey
boxes when I was there in mid to late 1990s though!)
Mike
On 7 May 2015 13:16, "Jon Elson" <elson at pico-systems.com> wrote:
On 03/13/2015 01:32 PM, Al Kossow wrote:
CHM was able to obtain volumes 18-20 of the IBM
2050 drawings, which are
the microcode charts and ROS dump. I got them scanned and uploaded
yesterday
to http://bitsavers.org/pdf/ibm/360/fe/2050
This was one of the things that I had been trying to locate for a while
now.
Just a curious thing I've noticed. Given the historic significance, and
the
large number of machines produced, I'm kind of amazed at the incredibly
small number of 360's that apparently exist.
Yes, I know, any would-be collector could drag home a PDP-8 and put it in
his garage, even a whole rack mount system with an RK02 (or 3), dectape and
paper tape reader, and still get his car in the garage. And, the system
could be run off normal mains power.
You can't do that with a real 360 (some 360/20's were pretty small), even
a 360/30 was a pretty big box. And, you can't run a 360 off normal
residential power, either. Many of the peripherals used 3-phase motors,
and hacking the converter/inverter to run off single phase would not be a
task for any but the most experienced EE.
But, it sure is a shame that there appear to be a tiny number of machines
in existence. One list shows 15 or 16 machines, excluding the model 20.
Probably there are a couple more hidden somewhere, like the B1900 that came
to light so recently.
As fas as I can tell, NONE of these systems is complete enough to ever
run, with the possible exception of the 360/30 at the CHM, which does seem
to have a complement of peripherals, and maybe control units, too.
Given the number of DEC 10's that are actually up and running, this seems
a bit of a surprise. there might be some emotional attachments that are
behind this disparity.
Anybody have some comments?
Jon
6:48 p.m.
On 05/07/2015 12:33 PM, Mike Ross wrote:
1. IBM would very often lease rather than sell, and
take back machines when
customers upgraded; it was obviously in their interests to control or
eliminate where possible the market in used machines.
In the early days, that was
VERY true, very few machines
were owned. In fact, IBM didn't sell a SINGLE machine until
they were forced to by the US government. In later years,
there were a lot of owned machines, many owned by the
computer brokers, who made a BIG business out of 3rd party
leasing in the 1970's.
2. Gold. A lot of gold in old IBM kit. I knew a scrap
dealer in Chelmsford
UK that did nothing but break old IBM mainframes (only ever big boring grey
boxes when I was there in mid to late 1990s though!)
A lot less gold in the 360's than in earlier machines. They
started using selective plating much earlier than anybody
else to cut the gold content. But, there is truth in what
you say. But, SMS cards had WAY more gold than the SLT systems.
Jon
10:41 a.m.
On May 7, 2015, at 1:16 PM, Jon Elson <elson at
pico-systems.com> wrote:
...
You can't do that with a real 360 (some 360/20's were pretty small), even a
360/30 was a pretty big box. And, you can't run a 360 off normal residential power,
either. Many of the peripherals used 3-phase motors, and hacking the converter/inverter
to run off single phase would not be a task for any but the most experienced EE.
That isn?t really true; hasn?t been for a long time, and even less so in recent decades.
Home machinery operators have used ?phase converters? for ages to drive machinery like
lathes that use three-phase motors. A rotary converter is basically just a three-phase
motor modified to run from one phase power across one of its windings ? it delivers the
three phases at the three winding terminals. And nowadays you can get a ?variable
frequency motor drive? which is in effect a three-phase AC power generator, frequency of
your choice within reason. The smaller ones run off one-phase power. Either option
starts at just a few hundred dollars, and rotary converters can be built from an old three
phase motor plus a capacitor or two for much less than that.
Rotary converters would easily feed any three phase powered computer with the exception of
CDC mainframes. A VFD would take care of those (since they want 400 Hz). All that
assumes you can get enough current at home, of course. At least in the USA, enough
current is often pretty easy, while more than one phase is not likely to be doable.
paul
6:53 p.m.
On 05/07/2015 12:41 PM, Paul Koning wrote:
Well, first, rotary converters draw a LOT of imaginary power
(in other words, they have an awful power factor) and so the
line current can become MUCH higher than you would expect.
Also, the 360 series converter inverters were REALLY
sensitive to line disturbances. Yes, the MOTOR load in the
peripherals could be run off a phase converter or a VFD.
I'm not completely convinced the CPU and controllers, etc.
would like phase converters. And, running them off a VFD
would likely be disastrous, as these put out 400 V square waves.
We tried to rig up a phase converter scheme to run the
motor-generator set on a 370/145 in a guy's house, and it
did NOT go well. he only had a 60 A 240 V service, and the
imaginary current was over 60 A!
Jon
On May 7,
2015, at 1:16 PM, Jon Elson <elson at pico-systems.com> wrote:
...
You can't do that with a real 360 (some 360/20's were pretty small), even a
360/30 was a pretty big box. And, you can't run a 360 off normal residential power,
either. Many of the peripherals used 3-phase motors, and hacking the converter/inverter
to run off single phase would not be a task for any but the most experienced EE.
That isn?t really true; hasn?t been for a long time, and even less so in recent decades.
Home machinery operators have used ?phase converters? for ages to drive machinery like
lathes that use three-phase motors. A rotary converter is basically just a three-phase
motor modified to run from one phase power across one of its windings ? it delivers the
three phases at the three winding terminals. And nowadays you can get a ?variable
frequency motor drive? which is in effect a three-phase AC power generator, frequency of
your choice within reason. The smaller ones run off one-phase power. Either option
starts at just a few hundred dollars, and rotary converters can be built from an old three
phase motor plus a capacitor or two for much less than that.
10:42 a.m.
On 05/07/2015 10:16 AM, Jon Elson wrote:
Given the number of DEC 10's that are actually up
and running, this
seems a bit of a surprise. there might be some emotional attachments
that are behind this disparity.
Anybody have some comments?
Could this perhaps be due to S/360 systems being leased more often than
purchased outright? After all, the leasing system was one of the
cornerstones of IBM's financial success. Some hold the opinion that it
was more important than any IBM technological innovation.
And, as far as I know, obsolete off-lease equipment was most often
simply junked.
--Chuck
11:02 a.m.
On Thu, 7 May 2015, Jon Elson wrote:
Just a curious thing I've noticed. Given the
historic significance, and the
large number of machines produced, I'm kind of amazed at the incredibly small
number of 360's that apparently exist.
Yes, I know, any would-be collector could drag home a PDP-8
Sometimes by simply physically dragging it out of dumpster!
(And then not having a place for it at home, and leaving it out in the
back yard to rot naturally)
Were 360s owned by the customer, or LEASED?
If leased, then they didn't go into "dumpsters" when
"decommissioned".
Wow! I'd like to see a dumpster that size.
6:57 p.m.
On 05/07/2015 01:02 PM, Fred Cisin wrote:
On Thu, 7 May 2015, Jon Elson wrote:
Lots of 360's were owned in the later days. All the ones I
knew of, with the possible exception of Washington
Universities' first 360/50 were owned (and bought used).
So, they went on lease, then somebody upgraded, and we
bought them used.
There was a HUGE market in leased and off-lease 360 and 370
machines in the 70's and 80's, some brokers got insanely
rich on that trade. Some of those lost it all when the
market collapsed. The smart ones knew to not keep too many
machines in inventory.
Jon
Just a curious thing I've noticed. Given the
historic
significance, and the large number of machines produced,
I'm kind of amazed at the incredibly small number of
360's that apparently exist.
Yes, I know, any would-be collector could drag home a PDP-8
Sometimes by simply physically dragging it out of dumpster!
(And then not having a place for it at home, and leaving
it out in the back yard to rot naturally)
Were 360s owned by the customer, or LEASED?
If leased, then they didn't go into "dumpsters" when
"decommissioned".
Wow! I'd like to see a dumpster that size.
8:20 p.m.
On 5/7/2015 6:57 PM, Jon Elson wrote:
Lots of 360's were owned in the later days. All
the ones I knew of,
with the possible exception of Washington Universities' first 360/50
The UMR
360/50 was one of the first owned 360s. They were given a
360/40 because IBM could not deliver the 50 on schedule. I don't know
how that was handled.
The 50 at UMR was also odd because it was a 512K main memory machine,
and 1mb of LCS, giving a non power of 2 memory size. Supposedly lead to
some bug fixes.
Ran HASP and MVT 21, as well as CPS
thanks
JIm
8 May
8 May
12:03 a.m.
On Thu, May 7, 2015 at 9:20 PM, jwsmobile <jws at jwsss.com> wrote:
The 50 at UMR was also odd because it was a 512K main
memory machine, and
1mb of LCS, giving a non power of 2 memory size. Supposedly lead to some bug
fixes.
I don't doubt that, but AFAIK almost all machines with LCS had a
non-power-of-two total memory size.
8:23 a.m.
On 05/07/2015 10:20 PM, jwsmobile wrote:
On 5/7/2015 6:57 PM, Jon Elson wrote:
Shouldn't have needed much software tinkering, as a number
of other models could be set up with non power of 2 memory
size. Maybe a hardware quirk was that it needed to make a
program exception if you addressed a 1.7 MB location, and
instead it would get a machine check with a main store
parity error. That would take maybe just one added gate to
trigger the program exception on the right combination of
address bits.
Lots of 360's were owned in the later days.
All the ones
I knew of, with the possible exception of Washington
Universities' first 360/50
The UMR 360/50 was one of the first owned 360s.
They were
given a 360/40 because IBM could not deliver the 50 on
schedule. I don't know how that was handled.
The 50 at UMR was also odd because it was a 512K main
memory machine, and 1mb of LCS, giving a non power of 2
memory size. Supposedly lead to some bug fixes.
Ran HASP and MVT 21, as well as CPS
Well, I gotta say, our 360/65 sure beat the pants off the
360/50 that it replaced. In fact, the /65 beat the pants
off TWO 370/145's they had. They used the /145's for
"timesharing" and got FOUR users per machine! Wow, that was
a riot.
Ahh, the "old days".
Jon
7 May
7 May
11:07 a.m.
-----Original Message-----
From: cctalk [mailto:cctalk-bounces at classiccmp.org] On Behalf Of Jon Elson
Sent: 07 May 2015 18:17
To: General at classiccmp.org; Discussion at classiccmp.org:On-Topic and Off-
Topic Posts
Subject: Re: 360/50 microcode listing
On 03/13/2015 01:32 PM, Al Kossow wrote:
I think that you are not comparing like with like. How many KA10's are there left as
that?s the DEC System/10 that?s comparable with the S/360. If you want to compare IBM
Mainframes with DEC System/10 then you need to include S/360, S/370 and 43xx boxes. Now
there may be few S/360's but there are 370's and I am sure there are many 43xx
boxes.
However I think that the quantity of preserved mainframes is dire when compared with other
items of "technology" history. There are hundreds of preserved static steam
engines, steam locomotives, vintage , veteran and classic cars or automobiles (not that
vintage and veteran have specific definitions in relation to cars/automobiles), radios and
TVs etc. I am sure there are more concorde aeroplanes and possibly even space shuttles
than there are S/360's...
Dave
G4UGM
CHM was able to obtain volumes 18-20 of the IBM
2050 drawings, which
are the microcode charts and ROS dump. I got them scanned and uploaded
yesterday to http://bitsavers.org/pdf/ibm/360/fe/2050
This was one of the things that I had been trying to locate for a
while now.
Just a curious thing I've noticed. Given the historic significance, and the
large
number of machines produced, I'm kind of amazed at the incredibly small
number of 360's that apparently exist.
Yes, I know, any would-be collector could drag home a PDP-8 and put it in his
garage, even a whole rack mount system with an RK02 (or 3), dectape and
paper tape reader, and still get his car in the garage. And, the system could
be run off normal mains power.
You can't do that with a real 360 (some 360/20's were pretty small), even a
360/30 was a pretty big box. And, you can't run a 360 off normal residential
power, either. Many of the peripherals used 3-phase motors, and hacking
the converter/inverter to run off single phase would not be a task for any but
the most experienced EE.
But, it sure is a shame that there appear to be a tiny number of machines in
existence. One list shows 15 or 16 machines, excluding the model 20.
Probably there are a couple more hidden somewhere, like the B1900 that
came to light so recently.
As fas as I can tell, NONE of these systems is complete enough to ever run,
with the possible exception of the
360/30 at the CHM, which does seem to have a complement of peripherals,
and maybe control units, too.
Given the number of DEC 10's that are actually up and running, this seems a
bit of a surprise. there might be some emotional attachments that are
behind this disparity.
Anybody have some comments?
Jon
11:56 a.m.
Is there a reliable estimate of the number of extant relatively complete and/or actually
working System/360 machines? A registry, perhaps?
I would find that to be a very interesting data point.
- Rob
12:15 p.m.
On 05/07/2015 11:56 AM, Robert Ferguson wrote:
Is there a reliable estimate of the number of extant
relatively complete and/or actually working System/360 machines? A registry, perhaps?
I would find that to be a very interesting data point.
I'd be interested also, in that a simple 360 CPU is essentially useless.
You need at least a minimum collection of working peripherals to be
able to do anything at all.
--Chuck
2:29 p.m.
+1
I got a 360/20 "running" at LCM, but we didn't have the 2560 MFCM
reader/punch for which it was wired. Unlike other 360s, the /20 didn't use
the channel architecture - all peripheral interfaces were built into the
2020 box.
And the machine won't even process instructions without peripherals being
sensed. The printer and reader/punch are purely electromechanical, and all
electronics is in the interface, including such niceties as cover
interlocks and card-bin status. Every macro-instruction cycle, the
microcode checks for those things and would stall if they were not in
order.
I always wanted to emulate the MFCM, but there were always other
projects... and even with the ALDs, it was a sizeable challenge.
On Thu, May 7, 2015 at 12:15 PM, Chuck Guzis <cclist at sydex.com> wrote:
On 05/07/2015 11:56 AM, Robert Ferguson wrote:
--
Ian S. King, MSIS, MSCS, Ph.D. Candidate
The Information School <http://ischool.uw.edu>
Archivist, Voices From the Rwanda Tribunal <http://tribunalvoices.org>
Value Sensitive Design Research Lab <http://vsdesign.org>
University of Washington
There is an old Vulcan saying: "Only Nixon could go to China."
Is there a reliable estimate of the number of
extant relatively complete
and/or actually working System/360 machines? A registry, perhaps?
I would find that to be a very interesting data point.
I'd be interested also, in that a simple 360 CPU is essentially useless.
You need at least a minimum collection of working peripherals to be able to
do anything at all.
--Chuck
2:53 p.m.
On 05/07/2015 02:29 PM, Ian S. King wrote:
+1
I got a 360/20 "running" at LCM, but we didn't have the 2560 MFCM
reader/punch for which it was wired. Unlike other 360s, the /20 didn't use
the channel architecture - all peripheral interfaces were built into the
2020 box.
Ah, but what would a Model 20 be without Mother Fletcher's Card Mulcher?
(cleaned up for "G" rating).
--Chuck
7:11 p.m.
On 05/07/2015 02:15 PM, Chuck Guzis wrote:
On 05/07/2015 11:56 AM, Robert Ferguson wrote:
Small 360's at least had integrated channels, but you still
needed control units AS WELL as the peripherals.
It ends up being a fair number of boxes to just get even a
360/30 to do any reasonable bit of work. Some of the unit
record control units could live in shared boxes. But, a
reasonable system would need a unit record controller, a
tape controller and a disk controller, as well as the card
machine, tape drives and disk drives.
The 360/65 had no integrated channels, IIRC, so you needed
to have at least a couple channel boxes in addition to the
CPU and control units. Ended up being a forest of blue
boxes. And, the wiring under the floor was totally
unbelievable!
Jon
Is there a reliable estimate of the number of
extant
relatively complete and/or actually working System/360
machines? A registry, perhaps?
I would find that to be a very interesting data point.
I'd be interested also, in that a simple 360 CPU is
essentially useless. You need at least a minimum
collection of working peripherals to be able to do
anything at all.
--Chuck
12:34 p.m.
Henk Stegeman has done this: http://www.ibmsystem3.nl/System360/index.html
On 07/05/15 20:56, Robert Ferguson wrote:
Is there a reliable estimate of the number of extant
relatively complete and/or actually working System/360 machines? A registry, perhaps?
I would find that to be a very interesting data point.
- Rob
--
Lawrence Wilkinson lawrence at ljw.me.uk
Ph +41(0)79 926 1036 http://www.ljw.me.uk
12:40 p.m.
I?m pretty sure I own the console nameplate from the unknown Dallas, Texas machine, but
don?t know the status of the rest of it. I?m betting it?s all been scrapped,
unfortunately.
--
Ben Sinclair
ben at bensinclair.com
On May 7, 2015, at 2:34 PM, Lawrence Wilkinson
<ljw-cctech at ljw.me.uk> wrote:
Henk Stegeman has done this: http://www.ibmsystem3.nl/System360/index.html
On 07/05/15 20:56, Robert Ferguson wrote:
Is there a reliable estimate of the number of
extant relatively complete and/or actually working System/360 machines? A registry,
perhaps?
I would find that to be a very interesting data point.
- Rob
--
Lawrence Wilkinson lawrence at ljw.me.uk
Ph +41(0)79 926 1036 http://www.ljw.me.uk
12:53 p.m.
Thanks, that's quite interesting.
Is there some consensus about how complete that list is? Does anyone have a feeling if it
includes most of the systems in private hands, for example?
- Rob
On May 7, 2015, at 12:34 PM, Lawrence Wilkinson
<ljw-cctech at ljw.me.uk> wrote:
Henk Stegeman has done this: http://www.ibmsystem3.nl/System360/index.html
On 07/05/15 20:56, Robert Ferguson wrote:
Is there a reliable estimate of the number of extant relatively complete and/or actually
working System/360 machines? A registry, perhaps?
I would find that to be a very interesting data point.
- Rob
--
Lawrence Wilkinson lawrence at ljw.me.uk
Ph +41(0)79 926 1036 http://www.ljw.me.uk
12:57 p.m.
On 05/07/2015 12:34 PM, Lawrence Wilkinson wrote:
Henk Stegeman has done this:
http://www.ibmsystem3.nl/System360/index.html
So, from this list, only one working--and it's a 4331, not a S/360 box.
That doesn't surprise me at all.
--Chuck
1:14 p.m.
On Thu, May 7, 2015 at 3:57 PM, Chuck Guzis <cclist at sydex.com> wrote:
On 05/07/2015 12:34 PM, Lawrence Wilkinson wrote:
We had a 4331 at Software Results in the 1980s, sharing the room with
a couple of VAX-11/750s and some PDP-11s, all speaking HASP, 3780, and
some SNA (through COMBOARDs on the DEC side). ISTR it wasn't too hard
of a machine to keep going, and not really more of a footprint than an
11/780.
-ethan
Henk Stegeman has done this: http://www.ibmsystem3.nl/System360/index.html
So, from this list, only one working--and it's a 4331, not a S/360 box.
That doesn't surprise me at all.
7:07 p.m.
On 05/07/2015 01:56 PM, Robert Ferguson wrote:
Is there a reliable estimate of the number of extant
relatively complete and/or actually working System/360 machines? A registry, perhaps?
I would find that to be a very interesting data point.
- Rob
This is certainly not authoritative, but the best I've found
so far :
http://www.ibmsystem3.nl/System360/index.html
Jon
9:19 p.m.
There are more S/360s out there than the list suggests.
We, the big iron collectors, do not share this information on any sort of
registry. We are a secretive bunch.
At least one S/360 is in running order.
--
Will
On May 7, 2015 10:07 PM, "Jon Elson" <elson at pico-systems.com> wrote:
On 05/07/2015 01:56 PM, Robert Ferguson wrote:
Is there a reliable estimate of the number of
extant relatively complete
and/or actually working System/360 machines? A registry, perhaps?
I would find that to be a very interesting data point.
- Rob
This is certainly not authoritative, but the best I've found so far :
http://www.ibmsystem3.nl/System360/index.html
Jon
8 May
8 May
8:24 a.m.
On 05/07/2015 11:19 PM, William Donzelli wrote:
There are more S/360s out there than the list
suggests.
We, the big iron collectors, do not share this information on any sort of
registry. We are a secretive bunch.
At least one S/360 is in running order.
WOW, can you give more info??!!?? I'd sure like to hear
more about that.
Jon
8:34 a.m.
No, I can not. Not without getting permission from the owner.
The secret nature of the big iron collectors probably has to do with money
involved, as well as the extremely conservative nature of mainframing. It
has taken me years to penetrate the veil, and I am still mostly an outsider.
Far more S/360s were saved than S/370s, which is a shame, as S/370s were in
a way more important.
--
Will
On May 8, 2015 11:25 AM, "Jon Elson" <elson at pico-systems.com> wrote:
On 05/07/2015 11:19 PM, William Donzelli wrote:
There are more S/360s out there than the list
suggests.
We, the big iron collectors, do not share this information on any sort of
registry. We are a secretive bunch.
At least one S/360 is in running order.
WOW, can you give more info??!!?? I'd sure like to hear more about that.
Jon
1:32 p.m.
Hey Will, are you still worried that IBM will come asking for 'their'
machine back? :-)
On Thu, May 7, 2015 at 9:19 PM, William Donzelli <wdonzelli at gmail.com>
wrote:
There are more S/360s out there than the list
suggests.
We, the big iron collectors, do not share this information on any sort of
registry. We are a secretive bunch.
At least one S/360 is in running order.
--
Will
On May 7, 2015 10:07 PM, "Jon Elson" <elson at pico-systems.com> wrote:
--
Ian S. King, MSIS, MSCS, Ph.D. Candidate
The Information School <http://ischool.uw.edu>
Archivist, Voices From the Rwanda Tribunal <http://tribunalvoices.org>
Value Sensitive Design Research Lab <http://vsdesign.org>
University of Washington
There is an old Vulcan saying: "Only Nixon could go to China."
On 05/07/2015 01:56 PM, Robert Ferguson wrote:
Is there a reliable estimate of the number of
extant relatively complete
and/or actually working System/360 machines? A registry, perhaps?
I would find that to be a very interesting data point.
- Rob
This is certainly not authoritative, but the best I've found so far :
http://www.ibmsystem3.nl/System360/index.html
Jon
7 May
7 May
7:05 p.m.
On 05/07/2015 01:07 PM, Dave G4UGM wrote:
Well, I'm pretty sure that the Concordes have been 100%
preserved, other than the one that crashed. True also for
the shuttles.
But, no, there is a list of known/suspected/rumored 360's,
and it lists 15 or 16 "real" 360 processors (I'm excluding
the 360/20, of which they list 5 units.)
There are several KA10's RUNNING! Compared to how many of
those were built, there seem to be a LOT higher percentage
of those still extant than of the 360's. (Maybe a bunch are
still to be revealed, I suppose.) And, I'm pretty sure
there are NO 360's running. The LCM is trying to get a
360/20 running. Maybe they are going to start with a small
project before attempting something harder.
Jon
-----Original Message-----
From: cctalk [mailto:cctalk-bounces at classiccmp.org] On Behalf Of Jon Elson
Sent: 07 May 2015 18:17
To: General at classiccmp.org; Discussion at classiccmp.org:On-Topic and Off-
Topic Posts
Subject: Re: 360/50 microcode listing
On 03/13/2015 01:32 PM, Al Kossow wrote:
I think that you are not comparing like with like. How
many KA10's are there left as that?s the DEC System/10 that?s comparable with the
S/360. If you want to compare IBM Mainframes with DEC System/10 then you need to include
S/360, S/370 and 43xx boxes. Now there may be few S/360's but there are 370's and
I am sure there are many 43xx boxes.
However I think that the quantity of preserved mainframes is dire when compared with
other items of "technology" history. There are hundreds of preserved static
steam engines, steam locomotives, vintage , veteran and classic cars or automobiles (not
that vintage and veteran have specific definitions in relation to cars/automobiles),
radios and TVs etc. I am sure there are more concorde aeroplanes and possibly even space
shuttles than there are S/360's...
CHM was able to obtain volumes 18-20 of the IBM
2050 drawings, which
are the microcode charts and ROS dump. I got them scanned and uploaded
yesterday to http://bitsavers.org/pdf/ibm/360/fe/2050
This was one of the things that I had been trying to locate for a
while now.
Just a curious thing I've noticed. Given the historic significance, and the
large
number of machines produced, I'm kind of amazed at the incredibly small
number of 360's that apparently exist.
Yes, I know, any would-be collector could drag home a PDP-8 and put it in his
garage, even a whole rack mount system with an RK02 (or 3), dectape and
paper tape reader, and still get his car in the garage. And, the system could
be run off normal mains power.
You can't do that with a real 360 (some 360/20's were pretty small), even a
360/30 was a pretty big box. And, you can't run a 360 off normal residential
power, either. Many of the peripherals used 3-phase motors, and hacking
the converter/inverter to run off single phase would not be a task for any but
the most experienced EE.
But, it sure is a shame that there appear to be a tiny number of machines in
existence. One list shows 15 or 16 machines, excluding the model 20.
Probably there are a couple more hidden somewhere, like the B1900 that
came to light so recently.
As fas as I can tell, NONE of these systems is complete enough to ever run,
with the possible exception of the
360/30 at the CHM, which does seem to have a complement of peripherals,
and maybe control units, too.
Given the number of DEC 10's that are actually up and running, this seems a
bit of a surprise. there might be some emotional attachments that are
behind this disparity.
Anybody have some comments? 
8 May
8 May
1:50 p.m.
From: Jon Elson
Sent: Thursday, May 07, 2015 7:05 PM
But, no, there is a list of known/suspected/rumored
360's, and it
lists 15 or 16 "real" 360 processors (I'm excluding the 360/20, of
which they list 5 units.)
There are several KA10's RUNNING! Compared to how
many of those were
built, there seem to be a LOT higher percentage of those still extant
than of the 360's. (Maybe a bunch are still to be revealed, I
suppose.) And, I'm pretty sure there are NO 360's running. The LCM
is trying to get a 360/20 running. Maybe they are going to start with
a small project before attempting something harder.
Where are there *any* KA10s running???? That's a very bold claim. We
only know of 3 complete KA10s in existence, and none of them is running.
We have been searching for a KA10 for the museum since before it *was* a
museum.
Rich
Rich Alderson
Vintage Computing Sr. Systems Engineer
Living Computer Museum
2245 1st Avenue S
Seattle, WA 98134
mailto:RichA at LivingComputerMuseum.org
http://www.LivingComputerMuseum.org/
8:06 p.m.
On 05/08/2015 03:50 PM, Rich Alderson wrote:
From: Jon Elson
Sent: Thursday, May 07, 2015 7:05 PM
OH, sorry, I got them mixed up, I thought YOU had one
running there! Geez, I wouldn't think it would be all that
hard to get a KA10 running, just a LOT of discrete
transistors. A couple guys at Wash U got a Linc running a
few years ago. Mostly all they had to do was put new
capacitors in the power supply. Of course, I know there's a
LOT more logic in a KA10.
I'm pretty sure I have heard of some KA10's running,
somewhere, in the last decade, but that could be apocryphal.
Jon
But, no, there is a list of
known/suspected/rumored 360's, and it
lists 15 or 16 "real" 360 processors (I'm excluding the 360/20, of
which they list 5 units.)
There are several KA10's RUNNING! Compared to how many of those were
built, there seem to be a LOT higher percentage of those still extant
than of the 360's. (Maybe a bunch are still to be revealed, I
suppose.) And, I'm pretty sure there are NO 360's running. The LCM
is trying to get a 360/20 running. Maybe they are going to start with
a small project before attempting something harder.
Where are there *any* KA10s
running???? That's a very bold claim. We
only know of 3 complete KA10s in existence, and none of them is running.
We have been searching for a KA10 for the museum since before it *was* a
museum.
8:05 a.m.
On 7 May 2015 at 19:16, Jon Elson <elson at pico-systems.com> wrote:
You can't do that with a real 360 (some
360/20's were pretty small), even a
360/30 was a pretty big box.
Have you read Lawrence Wilkison's account of rescuing an S360/20 in NZ
when he was a student? It was a truly epic undertaking.
It's here, among other places:
http://forums.whirlpool.net.au/archive/2245918
He is now implementing one on an FPGA -- with blinkenlights output on
the VGA port. I think one small part of the reasoning is that it's a
lot lighter to lift...
--
Liam Proven ? Profile: http://lproven.livejournal.com/profile
Email: lproven at cix.co.uk ? GMail/G+/Twitter/Flickr/Facebook: lproven
MSN: lproven at hotmail.com ? Skype/AIM/Yahoo/LinkedIn: liamproven
Cell/Mobiles: +44 7939-087884 (UK) ? +420 702 829 053 (?R)
8:34 a.m.
On 8 May 2015 at 17:05, Liam Proven <lproven at gmail.com> wrote:
Have you read Lawrence Wilkison's account of
rescuing an S360/20 in NZ
when he was a student? It was a truly epic undertaking.
It's here, among other places:
http://forums.whirlpool.net.au/archive/2245918
He is now implementing one on an FPGA -- with blinkenlights output on
the VGA port. I think one small part of the reasoning is that it's a
lot lighter to lift...
That was meant to go to Sun-Rescue... Sorry!
Since I know he is on this list -- it's how we met -- I'd not have had
the temerity to post it here...
--
Liam Proven ? Profile: http://lproven.livejournal.com/profile
Email: lproven at cix.co.uk ? GMail/G+/Twitter/Flickr/Facebook: lproven
MSN: lproven at hotmail.com ? Skype/AIM/Yahoo/LinkedIn: liamproven
Cell/Mobiles: +44 7939-087884 (UK) ? +420 702 829 053 (?R)
7:52 p.m.
On 05/08/2015 10:05 AM, Liam Proven wrote:
On 7 May 2015 at 19:16, Jon Elson <elson at
pico-systems.com> wrote:
YES, and I have saved it on my machine! A GREAT tale. But,
that machine probably no longer exists.
It was a 360/30 (NOT /20) by the way. A 360/20 is a 16-bit
mini, and not instruction set compatible with the 360 series.
Jon
You can't do that with a real 360 (some
360/20's were pretty small), even a
360/30 was a pretty big box.
Have you read Lawrence Wilkison's account of rescuing an S360/20 in NZ
when he was a student? It was a truly epic undertaking.
3514
days inactive
3571
days old
34 comments
18 participants
participants (18)
-
aek＠bitsavers.org -
ben＠bensinclair.com -
cclist＠sydex.com -
cisin＠xenosoft.com -
dave.g4ugm＠gmail.com -
elson＠pico-systems.com -
ethan.dicks＠gmail.com -
isking＠uw.edu -
jws＠jwsss.com -
ljw-cctech＠ljw.me.uk -
lproven＠gmail.com -
paulkoning＠comcast.net -
pontus＠update.uu.se -
RichA＠LivingComputerMuseum.org -
rob＠bitscience.ca -
spacewar＠gmail.com -
tmfdmike＠gmail.com -
wdonzelli＠gmail.com