9 Nov
2022
9 Nov
'22
8:17 a.m.
Is anyone familiar with PC-51 and/or CoreNET?
These are IBM 5110/5120 related tools developed by an individual in the
early 1980s.
My understanding is PC-51 was an emulator that ran BASIC programs from the
IBM 5110. One keyword new in the IBM 5110 was the "FORMS" keyboard, and
you could define input fields (including type-formatting constraints, like
sequence of letters and numbers) -- and once defined, you could
relatively-easily store all the contents of the fields to a file (on tape
or disk). I'm not entirely sure what format PC-51 supported (e.g. could
read in ASCII text files containing the BASIC programs?). But I always
imagined those customer data entry forms in old Radio Shack or Sears stores
-- large department stores -- being developed in something like this.
And CoreNET, I think was some kind of "null modem cable" that let the IBM
5110 communicate with an IBM PC 5150. The IBM 5110 has 3x DB25 connectors
at the backside (and 1x DB15 cable like what became the "standard" joystick
port on some systems in mid-late 1980s). The external tape and disk system
would use these connectors -- with software driven from the ROS. I've
always imagined it would be possible to "bit bang" across these external IO
pins with some PALM-assembly -- the machine should be fast enough to encode
7-bit ASCII at 300 baud across those pins, maybe 1200. I'm not sure if
CoreNET used or required any async card or the parallel communication card
(that did IEEE-488), i.e. not sure if it was more than just a cable.
But what's more interesting - apparently Sony is now the owner of both
these assets, PC51 and CoreNET. Maybe Hal Prewitt sold it to them? Why
would Sony be interested in it? Anyone happen to know anyone who works at
Sony, or ideas on where to start on even "asking them" about it? Might be
a lost cause these days.
Anyone happen to have a copy of the old manuals of either of these?
-Steve
9 Nov
9 Nov
10:21 a.m.
On Wed, 9 Nov 2022, Steve Lewis wrote:
always imagined it would be possible to "bit
bang" across these external IO
pins with some PALM-assembly -- the machine should be fast enough to encode
7-bit ASCII at 300 baud across those pins, maybe 1200. I'm not sure if
The PALM and thus the 51[012]0 is much much faster than that (I guess
about 0.7 MIPS).
I do bit-banging on the Asynchronous Serial I/O card with twice the baud
rate (to sample the center of a bit) and am able to do *full-duplex*
communications *with* terminal emulation *and* character set conversion at
4800 baud.
Using the byte mode of the I/O adapter I can go up to 38400 baud (used for
file transfers).
Remember: this card is absolutely dumb, it essentially only has shift
registers and a clock generator. The benefits come from the zero-overhead
interrupt context switches and the great number of fast processor
registers.
But back to your questions: I personally have never heard of these
products. I only know that there was a third-party hard disk option.
Christian
2:32 p.m.
Remember: this card is absolutely dumb, it essentially
only has shift
registers and a clock generator
Ah, the clock might be important. I was thinking of a terminal that could
be written without the async card, and using maybe 3 pins on those set of
external DB25 connectors. This is possible using the 4-pin connector
"serial IO" connector on the 1980 Color Computer (but it struggles to do
1200 baud). And yes part of the work would be receiving ASCII and
translating those results to "display character codes." I don't think the
5110 has a clock on its own, so you'd have to carefully time things? (like
doing audio on an original Apple 2 with no RTC module).
But back to your questions: I personally have never
heard of these
products. I only know that there was a third-party hard disk option.
I got a response from Hal Prewitt a couple months ago (apparently the
creator of these products), who stated he did still have a copy of PC-51,
but said "none of the hardware" was still available for CoreNET. I asked
if PC-51 could be made available basically into the public domain, but
hadn't heard a response (then later I learned of the Sony connection, which
maybe is the reason).
-Steve
On Wed, Nov 9, 2022 at 4:22 AM Christian Corti via cctalk <
cctalk(a)classiccmp.org> wrote:
> On Wed, 9 Nov 2022, Steve Lewis wrote:
> > always imagined it would be possible to "bit bang" across these
external
> IO
> > pins with some PALM-assembly -- the machine should be fast enough to
> encode
> > 7-bit ASCII at 300 baud across those pins, maybe 1200. I'm not sure if
>
> The PALM and thus the 51[012]0 is much much faster than that (I guess
> about 0.7 MIPS).
> I do bit-banging on the Asynchronous Serial I/O card with twice the baud
> rate (to sample the center of a bit) and am able to do *full-duplex*
> communications *with* terminal emulation *and* character set conversion at
> 4800 baud.
> Using the byte mode of the I/O adapter I can go up to 38400 baud (used for
> file transfers).
> Remember: this card is absolutely dumb, it essentially only has shift
> registers and a clock generator. The benefits come from the zero-overhead
> interrupt context switches and the great number of fast processor
> registers.
>
>
But back to your questions: I personally have never
heard of these
products. I only know that there was a third-party hard disk option.
>
> Christian
>
10 Nov
10 Nov
12:08 p.m.
On Wed, 9 Nov 2022, Steve Lewis wrote:
translating those results to "display character
codes." I don't think the
5110 has a clock on its own, so you'd have to carefully time things? (like
doing audio on an original Apple 2 with no RTC module).
Sure it has. The ASync card does not have an oscillator. It divides the
system clock that is brought out on one of the I/O bus pins (I think
+OSC/4 on A3 pin 2). Anyways, it uses the 3.01824 MHz clock from this bus.
Christian
7:24 a.m.
Just yesterday, I received the following notes from Hal Prewitt of CORE
(see below).
Some highlights:
- confirmation that use of async, comm, parallel card was not necessary
- from the PC51 manual, their hard drives were accessed as "device 08" (D08
instead of D80 for the IBM disk drive)
- gives me hope that doing a similar thing with a modern SD card might be
possible !
- Hal's notes (not shown below) indicate that "5,000 to 30,000 users of
5110s and 5120s,"
- does anyone have insight into NASA's usage of PALM in the 1960s?
Notes from Hal Prewitt (HP):
"CoreNET was our storage & network for the 5110/5120 systems. We wrote
PALM assembler, linker and IO drivers. We engineered an interface
card that connected to the 5110/20 cabling design. Build cables and
assorted storage systems. Did not use Async, Comm or parallel card
features. The HDD interface inside our drive boxes used SASI
(predecessor to SCSI). On the 5110/20 we loaded a driver from the
floppy which provided access to the CoreNet and our HD drives. The
5110/20 could directly access its storage devices (tape & floppies)
and our HD systems just by using new references for our storage
devices.
We also had a card for the PC bus and box that connected to the
5110/20 cables which provided the PC with access to the CoreNet and
IBM machines. Had a box with 8" flopies for the PC and when running
PC-51 turned the PC into a 5110/20 replacement running Basic.
The term "PALM" refers to the processor. I was not involved but think
it was created for NASA for use in the 1960-1970s Apollo missions. IBM
was the computer equipment contractor. And yes, the ROS emulated
S/360 code."
On Wed, Nov 9, 2022 at 2:17 AM Steve Lewis <lewissa78(a)gmail.com> wrote:
Is anyone familiar with PC-51 and/or CoreNET?
These are IBM 5110/5120 related tools developed by an individual in the
early 1980s.
My understanding is PC-51 was an emulator that ran BASIC programs from the
IBM 5110. One keyword new in the IBM 5110 was the "FORMS" keyboard, and
you could define input fields (including type-formatting constraints, like
sequence of letters and numbers) -- and once defined, you could
relatively-easily store all the contents of the fields to a file (on tape
or disk). I'm not entirely sure what format PC-51 supported (e.g. could
read in ASCII text files containing the BASIC programs?). But I always
imagined those customer data entry forms in old Radio Shack or Sears stores
-- large department stores -- being developed in something like this.
And CoreNET, I think was some kind of "null modem cable" that let the IBM
5110 communicate with an IBM PC 5150. The IBM 5110 has 3x DB25 connectors
at the backside (and 1x DB15 cable like what became the "standard" joystick
port on some systems in mid-late 1980s). The external tape and disk system
would use these connectors -- with software driven from the ROS. I've
always imagined it would be possible to "bit bang" across these external IO
pins with some PALM-assembly -- the machine should be fast enough to encode
7-bit ASCII at 300 baud across those pins, maybe 1200. I'm not sure if
CoreNET used or required any async card or the parallel communication card
(that did IEEE-488), i.e. not sure if it was more than just a cable.
But what's more interesting - apparently Sony is now the owner of both
these assets, PC51 and CoreNET. Maybe Hal Prewitt sold it to them? Why
would Sony be interested in it? Anyone happen to know anyone who works at
Sony, or ideas on where to start on even "asking them" about it? Might be
a lost cause these days.
Anyone happen to have a copy of the old manuals of either of these?
-Steve
742
days inactive
743
days old
4 comments
2 participants
participants (2)
-
Christian Corti -
Steve Lewis