[Simh] PDP for beginners

Sat Feb 3 17:01:51 EST 2018

On Sat, Feb 3, 2018 at 8:12 AM, Pär Moberg <ghostdewolf at gmail.com> wrote:

> On 03/02/2018, Bob Eager <rde at tavi.co.uk> wrote:
> > On Sat, 3 Feb 2018 11:41:38 -0000
> > "Dave Wade" <dave.g4ugm at gmail.com> wrote:
> >
> >> > -----Original Message-----
> >> > From: Simh [mailto:simh-bounces at trailing-edge.com] On Behalf Of Pär
> >> > Moberg
> >> > Sent: 03 February 2018 02:26
> >> > To: simh at trailing-edge.com
> >> > Subject: [Simh] PDP for beginners
> >> >
> >> > I am looking to put up a PDP-# machine to play with. What machine
> >> > and OS should I use. I am looking for simple/beginner system, or as
> >> > beginner it got.
> >>
> >> I think I would start with a PDP-8. NO telnet terminal. Simple OS,
> >> lots of software and support. If you want blinken lights try one of
> >> these:-
> >>
> >> http://obsolescence.wixsite.com/obsolescence/pidp-8-get-one
> >>
> >> its SIMH with a nice panel
> >
> > I was going to suggest the same thing. Be warned that there's quite a
> > lot of soldering in a PiDP-8 - although not as much as an SBC-6120! (I
> > have both)
>
> Why I want terminal support is so I can use some of my old clunkers as
> terminals and access the server from different places in the house.

Like others the PDP-8 is definitely were I would start because there is so
much around for it, from single user, batch and timesharing systems.
 Google is you friend of
course
, a couple of interesting URLs are:
Bitsavers's PDP-8 Download
<http://bitsavers.trailing-edge.com/pdf/dec/pdp8/>,

PDP-8 Handbooks <http://bitsavers.trailing-edge.com/pdf/dec/pdp8/handbooks/>
,
 Running_TSS_8_on_the_DEC_PiDP-8_i_and_SIMH.html
<https://raymii.org/s/articles/Running_TSS_8_on_the_DEC_PiDP-8_i_and_SIMH.html>,
and
PDP-8 Net <http://www.pdp8.net/index.shtml,>.
[BTW:  do not pass go ....   down load the PDP-8 users handbook to start
BTW].

There are also a number of different simulators to work with.  But there
are some caveats that have been discussed (more in the minute). I too have
a PiDP-8 and I have it connected to serial ports on the RPi to drive
physical connections, but you can use the telnet stuff from simh described
elsewhere.

That said -- as much as I love Bob Supnik and Mark Pizzolato's
wonderful simulation system (particularly if you want to talk to 'real
hardware'), if you want to to learn a but more about the PDP-8 itself, I
might recommend you also consider Berhbard Baehr's amazing PDP-8/e
simulator for the Mac:  A PDP-8/E Simulator for the Apple Macintosh
<http://www.bernhard-baehr.de/pdp8e/pdp8e.html>. A know a of a couple of
excellent CS Departments that use this emulation to teach their students in
their intro to architecture class because it gives also exposes some more
modern debugging schemes and thus is easier to get a handle on what is
going on 'inside.'  From the web page:

The simulated machine is a PDP-8/E with 4K words of memory and optionally a
KM8-E Memory Extension (with up to 32K words of memory) and a KE8-E
Extended Arithmetic Element. I/O devices are implemented as plugin modules,
and plugins for an ASR 33 Console Teletype, an ASR 33 Auxiliary Teletype, a
PC8-E High Speed Paper Tape Reader and Punch, and a RK8-E Disk Cartridge
System are available.

There is also a KC8-EA Programmer's Console plugin enabling the user to
operate the simulator (nearly) like a hardware PDP-8/E, including single
step execution on memory cycle level.

As other have pointed out OS/8 is out there which like the later DOS/11
and RT-11 which we modeled on it, is basically single user.   But there is
also TSS/8, ETOS and I think a few other
multi-
user systems for the 8.   But I would recommend starting with DOS/8.    I
used's OS/8 and TSS/8 'back-in-the-day.'

That said, if you do want to mess with these with real hardware such as
PiDP-8, there is an interesting feature/bug that is annoying when working
with the DOS/8 (or TSS/8) under emulation on a real-terminal (on an XTERM
or the like this is less of an issue).  It's how the hardware works and how
^S/^Q flow control is implemented in the old systems.    The problem is
that the way the OS works for terminal output is that spins on a read bit
from the UART before it writes the next character.  This is done at the
lowest level of the OS.   In the old days, even at the 'slow' 10cps of an
ASR-33, the truth is the simulated hardware will push the next chars out
faster than you the human can respond.   The ^S you will send back to the
emulated system will be seen much, much later that it would have been in
the real HW.  [The solution is we need to change simh to not set the ready
bit until a real time amount of 10 cps for 110 baud or 960 cps for 9600
baud - but this is very hard to do and in practice, not needed because we
mostly use simulated TTY today].  i.e.
If you are one a simulated TTY such as xterm, there is enough buffering in
the xterm
 software emulation to not 'lose' character and then it adds
the ability to 'scroll back
.
'
  B
ut on a real display of the day, you might out of luck
 (i.e. when I connect even as modern a terminal as an Heathkit H19 or
a Wyse-60)
.

Best of luck,
Clem

ᐧ
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.trailing-edge.com/pipermail/simh/attachments/20180203/34899ab0/attachment.html>