24 Dec
2024
24 Dec
'24
7:36 p.m.
There was a fellow from one of the research buildings who had an xray
crystalography project ongoing, and some of his research is still
textbook for the subject.
Anyway he had been running for years in 72 on, and was generating
massive amount of data. He had two full handtrucks of 2000 card decks
with his program and at a box or two of data in the pile. He did a lot
of matrix processing on the 360/50 (512k of main memory, and 1mb of
LCS), which was in the scheme of things not a lot of computing to do
such with.
He took measurements with sensors at a large number of points on the
back side of an irradiated crystal and entered information about the
position and result.
anyway, the computer center had job cost controls in place, and he had
an account on both the 360, and later 370 TSO with no limits. He shared
his login with me for the system, so I could do jobs any time for our
computer project, which had a minicomputer lab, and half inch tape. Our
use of the mainframe was preparing jobs on our minicomputer, and
creating job tapes and submitting them or data via the 9 track half inch
tape.
Anyway used it thru the mid 80s when the TSO for the university was shut
down and the dialin went away.
The 2540 in this case was hooked to the 360/50 and the job he had was
read into the 360's 2314 subsystem. We ran what I was told was the only
HASP workstation with a remote running on an actual 360. It fed into
HASP, later JES on the 370 145 at another university.
He could submit his job to run either locally on the 360, or transmit it
to the other system to run, and route the results back to our 360/50 for
printing on our local 1403. The one which could fire paper across the
room from earlier in the thread.
I don't recall the reason for the channel 12 being blank and unlimited,
or if it was as simple as my recolletion. I believe there was some
channel control fiddling involved, and I like to play with obscure thing
like that when I discovered documentation about such.
One thing I found and fiddled with for instance was a utility exit in
the tape subsystem library called SPIE. Turned out I could submit a
SPIE handler which could get control when a tape fault occurred in
supervisor mode. Wasn't hard to generate a fault on the tape and get
the exit triggered.
The system programmers added code to block that after I reported it to them.
Thanks
jim
On 12/23/24 19:54, CAREY SCHUG via cctalk wrote:
we had one job that read cards into the middle bin,
then punched totals or something into that bin too.
other times, as a college student and operator, under DOS/360 we seldom used the 2K F2
partition, so I wrote a program so, after I altered storage to ALSO assign the punch to
F2, I would type in my programs and punch them to the middle bin. ran them through the
interpreting card punch later.
<pre>--Carey</pre>
> On 12/23/2024 3:11 AM CST Nico de Jong via cctalk <cctalk(a)classiccmp.org>
wrote:
>
>
>> The carriage control tapes had the sprocket holes dead center which
>> lead to people putting them on backwards by accident and since most
>> people only used a couple channels for any given form, this would lead
>> to paper runaways, as would neglecting to lower the brush block after
>> mounting the carriage control tape.
>>
>> I once saw one of the large system CEs repairing a print train that
>> the customer had neglected to fill the oiler and the train seized.
>> The filling the oiler was the customer's responsibility so the repair
>> of the train was billable. The type slugs I saw where not coupled
>> together and had helical gear teeth on the bottom that coupled with a
>> gear in the train that was driven by the motor. I am not sure what
>> model of 1403 this came from but it was one of the models that had
>> covers that went all the way to the floor. The print trains had a
>> separate machine type and I seem to recall that the customer was
>> obliged to purchase them even if the 1403 was leased. The 3203
>> printer used the same print trains as 1403.
>>
>> Paul.
>>
> When I was an operator, we once had a visit from a CE who had to repair
> the carriage control mechanism. In order to do that, he had to use a big
> screwdriver, and of course he lost it. It hit the 1403 N1's power
> supply, blew all fuses. This was not enough; the screwdrive hit
> obviously the plus and minus pole of the main capacitor (it's about 55
> years ago), so the current was so large that, after the things had
> cooled down, he could lift the capactor out of the printer just by
> lifting the screwdriver
>
> It was by the way the same CE that got his tie wrapped up in the print
> chain....
>
> The same company once had a bunch of visitors who were allowed to visit
> the machine room, which normally was a bit nono. One of the guests took
> his coffeecup with him, put it on top of the 1403, and while things were
> explained to the crowd, the cover lifted and .... well you can guess the
> rest. He was quite pisssed off, but it was his own fault
>
> Another thing I'll never forget, was the 2540. It had 5 bins, and the
> middle one could be used for accepting read cards and punched cards.
> Once an operated started to read cards while cards were being punched,
> and both routines used the middle bin. That is not to be recommended !
>
> /Nico