Re: System 3 Code

Are you sure of that?  I thought those were 250k platters, not 2,500k. The 
Model 6 didn't have a screen or terminal.  It had a typewriter in the 
desktop and the printer housing behind it.  The operator typed commands to 
the printer.  A programmer quickly learned that all print programs started 
with a skip before and at last record time you did a skip after, otherwise 
the operator would skin you alive for making her life miserable.    The 
smallest memory was around 5k, of which 3k was reserved for the operating 
system.  The H-spec position that talks about using 1/4k increments was 
important, as were overlays.  Overlays allowed larger programs but 
performance was horrible. 


_______________________
Booth Martin
Booth@MartinVT.com
http://www.MartinVT.com
_______________________




Mel Rothman <melrothman@home.com>
Sent by: owner-rpg400-l@midrange.com
09/28/2000 03:53 PM
Please respond to RPG400-L

 
        To:     RPG400-L@midrange.com
        cc: 
        Subject:        Re: System 3 Code

Richard, here are a few more things to add to your list of "what happens 
when
people have 'lots of experience.'"

Although you didn't say that RPG one never ran on System/3 (System/3 was
announced with and faded out with RPG II), I wanted to make sure that no 
one
thinks your were implying so.

I never had any experience with the console problems you mentioned.  Lucky 
me!

The "pizza platter's" capacity (5440 cartridge) was 2.45 MB, not 10 MB. If 
one
had two full drawers of disk, F1, R1, F2, and R2, one had 9.8 MB of online
capacity. 

The early System/3s disks were installed in one or two drawers, drawer 1 
and
drawer 2.  Drawer 1 was required.  Drawer two was optional.  A drawer's 
maximum
capacity was evenly split between a fixed (non removable) platter and a
removable platter mounted in a 5440 cartridge.  I believe the fixed 
platter in
drawer 2 was optional.  The fixed platters were called F1 and F2 (fixed 
platter
in drawer 1 and drawer 2, respectively); the removable platters were 
called R1
and R2 (removable platter in drawer 1 and drawer 2, respectively).  Drawer 
1
always had an F1 and a R1.  The possible configurations were 2.45 MB (half
capacity F1 and R1), 4.9 MB (full capacity F1 and R2), 7.35 MB (full 
capacity
F1, R1, R2), and 9.8 MB (full capacity F1, R1, F2, R2).

The fixed platters were backed up by copying (DISKCOPY utility) them on to
removable cartridges.  An inexperienced operator at one of my (not to be 
named)
customers, backed up the system one day by copying F1 to EVERY removable
cartridge.  It took them months to reconstruct their data from printed 
reports. 
What a nightmare!
 
Before 3340s were available on S/3, some models supported 5445s, which 
used 2314
disk packs with approximately 20 MB capacity each.  I believe the maximum 
number
of 5445s was either 2 or 4.

The smallest System/3 memory size I saw was 8 KB on the card Model 10 and 
the
Model 6

I don't remember the "black crinkle paint."  I remember gray paint and 
panels in
red, yellow, blue, or white.

Models 10 and 12 were available with the dual programming feature (DPF), 
which
wasn't as sophisticated as the Model 15's partitioning.  I believe the 
maximum
total memory size for DPF systems was the same as DPF systems, 64 KB.

I remember MRTs (multiple requester program) well.  I avoided them 
whenever
possible.  I preferred SRTs (single requester program), which wrote to the
display and ended.  Code to start the next program using program request 
under
format (PRUF) was left on the display.  When the user pressed ENTER, the 
next
program would load and process the display's input.  The AS/400 KEEP and 
ASSUME
DDS keywords are PRUF's legacy.

I really liked 96 column cards.  They fit completely in my shirt pocket, 
leaving
my IBM white shirt still looking "elegant."  80 column cards were more 
than
twice the height of the pocket and hung out sloppily.  Since I worked in
Phoenix, Arizona, where it gets very hot, I didn't always have a suit 
jacket on
to cover the unsightly 80 column card.  So, I used 96 column cards 
exclusively!

I thought DITTO was a utility, not a programming language.  I thought RPG 
one
was modeled after unit record (wired boards) equipment.

The hot 1403 printer was the Model N1 (affectionately called Nankeen).  It 
was
an engineering marvel that printed 1100 lines a minute (still pretty fast, 
I'd
say).  Like any high performance machine, you had to check the oil or risk
having it burn out!  The 1403 printer family was used on System/360 and 
earlier
IBM systems such as 1401 before System/3 came along.

There is much about System/3 that I remember but have not mentioned. There 
is
even more that I have forgotten!


Mel Rothman



Richard Jackson wrote:
> 
> That was a challenge.  Here is my reply - originally written to Lisa 
Abney.
> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> After about 13 replies to your note below, it should be clear to you 
that
> there are some awfully forgetful people on this mailing list.
> 
> They don't remember that RPG I was created to replace DITTO.  That RPG 
one
> was implemented on 360 and 1130 before System/3.  They didn't point out 
that
> RPG I didn't permit subroutines.  I'll bet that some of them will tell 
you
> that that they don't know what DITTO is.  Incredible.  See what the 
passage
> of time does to memory?
> 
> Somebody said that CICS ran on a 15D ... NEVER!  (read this next bit 
very
> fast, in a quiet voice ... Of course, CCP was created by tearing big 
chunks
> out of the CICS source code and porting it to an 8-bit machine and had 
to be
> gen'd just like CICS where every file and program that ran under CCP was
> specified and the memory region and buffers were described and ... well,
> never mind.)
> 
> They forgot to mention the legend that the smallest originally-announced
> System 3 model never shipped because the RPG err, umm, "compiler" 
couldn't
> be shoehorned into the max memory on the machine.  (Where is Hans Koert 
when
> you need him ... :)  Or that they were done up in black crinkle paint 
and
> looked like and antique coffee grinder.
> 
> Mel mentioned the console on a 15D ... :)  He didn't mention what 
happened
> if the bolt holding the 5-volt cable to the top of the main gate became
> loose or if the grounding was correct.  Mel mentioned partitions - BG, 
F2
> and F2.  Those the only places where stuff could run on a 15D.  Period. 
On
> models 8, 10, 12 and 15, there was just one thread of execution.  See 
how
> the memory goes?
> 
> Nobody mentioned that the 96-column cards were approximately square and 
this
> made it hard to properly orient them after you dropped a tray - the 
corner
> cut was important.  Or that the holes were round instead of square which
> made them easier on the eyes if you threw a handful of scurf at someone. 
 Or
> that they made crummy book marks and note papers - unlike 80-column 
cards.
> Nobody mentioned the 10 meg pizza-platter disks.  Nobody mentioned the 
awful
> tape drives with the slide-up plastic doors - horrible things.  Nobody
> mentioned that 15Ds used 1403 printers or told you what happened if the 
oil
> ran low in the printer or talked about programming carriage tapes. 
Nobody
> mentioned the fact that terminals were VERY heavy and VERY dumb and had 
to
> be connected the machine using special coax cable with funny BNC 
connectors.
> Nobody mentioned debugging a CCP gen - not funny given the absence of a
> debugger and the available documentation.  Only Mel mentioned program
> overlays - sort of important with that 32K max program size.  Nobody
> mentioned the almost-undocumented $-sign macro calls used to read and 
write
> to the displays.  Nobody mentioned MRTs (Is that wailing or moaning I
> hear?).  Nobody mentioned that the System 3 didn't have a debugger.  Or 
that
> the 3310/3340 were the original Winchester disk.  Nobody mentioned the
> rewards of making SORT do everything or the mysteries of making it work
> right.  Or the days where people didn't trust their disk drives because 
they
> crashed all the time - remember 2311 and 2314?  Is that crying I hear?
> 
> See what happens when people have "a lot of experience"?  They forget 
stuff.
> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> 
> I'm going to my room now - to look up more terrifying stuff!
> 
> Richard Jackson
> mailto:richardjackson@richardjackson.net
> http://www.richardjacksonltd.com
> Voice: 1 (303) 808-8058
> Fax:   1 (303) 663-4325
> 
> > |-----Original Message-----
> > |From: owner-rpg400-l@midrange.com 
[mailto:owner-rpg400-l@midrange.com]On
> > |Behalf Of Norm Dennis
> > |Sent: Wednesday, September 27, 2000 7:40 PM
> > |To: RPG400-L@midrange.com
> > |Subject: Re: System 3 Code
> > |
> > |
> > |> I have a long satirical reply to this about how "very
> > |experienced" people
> > |> are forgetful - but it doesn't matter.
> > |
> > |Richard,
> > |go to your room until you learn to respect your elders. <VBG>
> > |
> > |
> > |Norm
> > |
> > |__________________________________________________
> > |
> > |Western Midrange Software Services Pty Ltd
> > |Office Telephone: +61 8 9312 1995
> > |Mobile Telephone: 0417 659 914
> > |Web Site: http://www.iinet.net.au/~wmss
> > |__________________________________________________
> > |
> > |
+---
| This is the RPG/400 Mailing List!
| To submit a new message, send your mail to RPG400-L@midrange.com.
| To subscribe to this list send email to RPG400-L-SUB@midrange.com.
| To unsubscribe from this list send email to RPG400-L-UNSUB@midrange.com.
| Questions should be directed to the list owner/operator: 
david@midrange.com
+---




+---
| This is the RPG/400 Mailing List!
| To submit a new message, send your mail to RPG400-L@midrange.com.
| To subscribe to this list send email to RPG400-L-SUB@midrange.com.
| To unsubscribe from this list send email to RPG400-L-UNSUB@midrange.com.
| Questions should be directed to the list owner/operator: david@midrange.com
+---