What do use a PiPD-11 for...

[Eric Clark]

I'm all about blinky lights and retro emulation. But I'm curious what folks use these for on their home networks. I'm not above getting one for the novelty - but what can I employ this to do on my network? Email server, network logging, smart home integration? 

[Mark Rosenthal]

You could do all of that.

Everyone has their own interests.

I'm using it to:

Explore PDP-11 machine architecture and re-visit what it was like to run a 1980s era multi-tasking, multi-user minicomputer with a distinct 'non-*ix' environment that provides much more 'out-of-the-box' software features than today's Linux and Windows.  If you doubt that, check the available documentation.  If you're willing to spend more on paper and toner than you will for the PiDP-11, you can even print the manuals and fill up a bookshelf.

And the blinky lights.

Cheers!

[timr...@gmail.com]

I have several.  One runs my PiDP-8, and another runs my PiDP-10.  The third I run Pihole as my DNS server and spam blocker.

[Marco]

Mine actually hosts a (very small) website:

https://pdp11.crouze.com/

Only two pages, and authentically (?) slow!

Op vrijdag 9 augustus 2024 om 20:18:10 UTC+2 schreef timr...@gmail.com:

[timr...@gmail.com]

Cool.  I viewed your web site.  What provides the service on the pdp-11 side?  Maybe I missed that?

[Marco]

The PDP-11 is running RSX-11M-PLUS, with Johnny Billquist's TCP/IP implementation which has its own web server. And FTP server too which makes editing and uploading content much easier...

Op dinsdag 13 augustus 2024 om 19:42:19 UTC+2 schreef timr...@gmail.com:

[timr...@gmail.com]

I knew about Johnny's TCP thing.  I had forgotten it had a web server in there too.   I use a similar thing on my old DOS PC to give it networking.

[oscarv]

Hi,

I read your question differently from the other respondents - not 'what you do with/within the PDP-11 simulation' but 'what do you run *concurrently* on the Pi next to the PDP-11 simulation'?

So if that was the intention of the question:

know that certainly on a Pi 3 and up, you've got plenty of CPU cycles left to run anything else you'd want - no need to have a separate Pi for use as NAS, adblocker, mediaserver or whatever.

I think 99.5% of PiDP-11 builders use their machine solely for the PDP-11 simulation. Which always surprised me. Personally, I always considered the PiDP-11 box to be 'dual use'. It was a major reason to go for a Pi Inside rather than something else.

Kind regards,

Oscar.

[terry-...@glaver.org]

I've found the cost of the Pi (3B on my PiDP-11's, 5 on my PiDP-10) to be small enough compared to the cost of the system that I just dedicate the Pi to that use.

On the PiDP-11, the Pi is hidden inside the case with only power, Ethernet, and 4 serial ports (DB-25M, DEC pinout of course, running off 4 CH340 USB/serial adapters in the Pi and through the MAX232's on the PiDP-11) brought out. Completely headless and accessed via VNC.

On the PiDP-10, I do much the same, except there's only 2 USB/serial ports from the Pi brought out to the back panel through the MAX232. Since the Pi does stick out the back, I suppose I could do other things with it. Likewise, headless via VNC.

On all of the PiDP systems, I'm using Samsung 64GB Max Endurance SDXC cards. It is a write-intensive card used for recording video surveillance and similar stuff, to avoid clobbering the Pi's operating system and the emulator with lots of writes.

[Tom Lake]

I have eight user capability on my PiPD-11 running RSTS/E 10.1. I usually have three or four students at a time running BASIC while I teach programming techniques.

[timr...@gmail.com]

Now that is indeed a cool use.  Thank you for teaching these younger folk about programming.

[Johnny Billquist]

Eight??? What has the world come to...? Back in my time, we were usually
around 40 users online and active on RSTS/E at the same time, on an
11/70. When it went up to 60, it was really slow...
(Said with a smile, but the facts are correct...)

Johnny

> --
> You received this message because you are subscribed to the Google
> Groups "[PiDP-11]" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to pidp-11+u...@googlegroups.com
> <mailto:pidp-11+u...@googlegroups.com>.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/pidp-11/fcaea31b-b698-4d52-9c64-f6f220cadc89n%40googlegroups.com <https://groups.google.com/d/msgid/pidp-11/fcaea31b-b698-4d52-9c64-f6f220cadc89n%40googlegroups.com?utm_medium=email&utm_source=footer>.

--
Johnny Billquist || "I'm on a bus
|| on a psychedelic trip
email: b...@softjar.se || Reading murder books
pdp is alive! || tryin' to stay hip" - B. Idol

[terry-...@glaver.org]

8 users on a DZ-11 is horrible. 8 users on a DH-11 will barely be

noticed by the CPU. Of course, the 11/93 has 8 DL-11 ports, but

they're emulated by a microcontroller and have have FIFOs, so

they aren't as unbearable as you might think. On the PiDP-11 it

is all emulation, so the differences between "horrible" and "good"

multiplexors is a lot less obvious. I'm running my 4 physical ser-

ial ports as a DHU11 emulation, and even autobaud works (most

of the time, on a subset of baud rates) which is pretty amazing

given the numbers of things between the terminal and RSTS/E,

compared to an actual 11 - CH340 USB/serial, Raspberry Pi OS,

simh, etc.

At SPC, we moved from Gandalf front-end concentrators to DEC-

server 550s (one full of CXY08 cards and the other full of CXA16

cards), so we didn't have to keep track of physical ports on CPUs

any more.

Mostly it was edit/compile/debug iterations, so it wasn't too bad

even with lots of users. We didn't have many users trying to do

things like running Kermit to transfer files. We were running a

Netware/LAT/DECnet gateway so people could just copy their

files from the floppy in a lab PC to whichever host, so that didn't

involve emulated serial communication. 

I think the record for physical port craziness was on my DG

Eclipse S/200 (which had morphed into the Frankenclipse by

then) where I had 64 multiplexor ports which were often fully

loaded. There's the [infamous] STR I filed with DG saying "DG

XBASIC SYSGEN allows any 32 ports, as long as they're the

first 32". I got that STR back rubber-banded to an XBASIC

source tape with a bunch of other customer's XBASIC STRs

in the same box.

Most of our VAXen only had the console and a single DMF32

(combo 8 async / 1 bsync / 1 LP11) which got used to drive

the attached printer (there were additional shared printers on

DECserver 250s) and the sync port to talk to BITNET via Jnet.

[Johnny Billquist]

On 2024-08-18 08:01, terry-...@glaver.org wrote:
> 8 users on a DZ-11 is horrible. 8 users on a DH-11 will barely be
> noticed by the CPU. Of course, the 11/93 has 8 DL-11 ports, but
> they're emulated by a microcontroller and have have FIFOs, so
> they aren't as unbearable as you might think. On the PiDP-11 it
> is all emulation, so the differences between "horrible" and "good"
> multiplexors is a lot less obvious. I'm running my 4 physical ser-
> ial ports as a DHU11 emulation, and even autobaud works (most
> of the time, on a subset of baud rates) which is pretty amazing
> given the numbers of things between the terminal and RSTS/E,
> compared to an actual 11 - CH340 USB/serial, Raspberry Pi OS,
> simh, etc.

The 8 DL ports on the 11/93 are just as horrible to the CPU as any other
DL11 on any other CPU. The fact that the hardware have buffering don't
change anything from the PDP-11 CPU point of view. It only makes it
better in the sense that the risk of loosing characters are less.

And yes, I agree that DL11s are horrible, as are DZ11. But it's not as
bad as on a VAX, because the PDP-11 is much better at handling
interrupts. Running a DZ11 or two on an -11 is still ok. I would
probably not recommend running 8 of them at the same time.

As for automatically detecting the speed, I don't know how RSTS/E does
it, but on RSX, it actually depends on the specifics of the hardware, so
if it works or not on emulation is very much a case of "luck". RSX
autobaud detection handles DZ different than DH for example.

> At SPC, we moved from Gandalf front-end concentrators to DEC-
> server 550s (one full of CXY08 cards and the other full of CXA16
> cards), so we didn't have to keep track of physical ports on CPUs
> any more.

Yeah. That is rather nice. LAT is a pretty good protocol for keeping
host resource usage down.

> Mostly it was edit/compile/debug iterations, so it wasn't too bad
> even with lots of users. We didn't have many users trying to do
> things like running Kermit to transfer files. We were running a
> Netware/LAT/DECnet gateway so people could just copy their
> files from the floppy in a lab PC to whichever host, so that didn't
> involve emulated serial communication.

In that scenario, DZ shouldn't really be much of an issue at all. File
transfer, or running "graphic" games are when you feel the pain.

> I think the record for physical port craziness was on my DG
> Eclipse S/200 (which had morphed into the Frankenclipse by
> then) where I had 64 multiplexor ports which were often fully
> loaded. There's the [infamous] STR I filed with DG saying "DG
> XBASIC SYSGEN allows any 32 ports, as long as they're the
> first 32". I got that STR back rubber-banded to an XBASIC
> source tape with a bunch of other customer's XBASIC STRs
> in the same box.

Well, like I said - the 11/70 we used when I was at the gymnasium in
Sweden had 64 terminals. And it was fairly common to actually see 63
users logged in. I don't remember if we had DZ11 or DH11 on that
machine, but I think/hope it was DH11. Anyway, most terminals were at
1200 bps, because there were actually 4 schools sharing the machine, and
so we had most users remote. We had fixed, rented lines between the
places, on which we ran one 9600 bps line. That had a MUX on it, so we
had 8 1200 bps lines on each side. So each remote school had 8
terminals, and at the place where the computer was they had all the rest.

> Most of our VAXen only had the console and a single DMF32
> (combo 8 async / 1 bsync / 1 LP11) which got used to drive
> the attached printer (there were additional shared printers on
> DECserver 250s) and the sync port to talk to BITNET via Jnet.

Yeah. That was a popular combo. The DMF32 is actually pretty nice. Good
device for VAXen.

Johnny

> https://groups.google.com/d/msgid/pidp-11/fcaea31b-b698-4d52-9c64-f6f220cadc89n%40googlegroups.com <https://groups.google.com/d/msgid/pidp-11/fcaea31b-b698-4d52-9c64-f6f220cadc89n%40googlegroups.com> <https://groups.google.com/d/msgid/pidp-11/fcaea31b-b698-4d52-9c64-f6f220cadc89n%40googlegroups.com?utm_medium=email&utm_source=footer <https://groups.google.com/d/msgid/pidp-11/fcaea31b-b698-4d52-9c64-f6f220cadc89n%40googlegroups.com?utm_medium=email&utm_source=footer>>.

>
> --
> Johnny Billquist || "I'm on a bus
> || on a psychedelic trip
> email: b...@softjar.se || Reading murder books
> pdp is alive! || tryin' to stay hip" - B. Idol
>

> --
> You received this message because you are subscribed to the Google
> Groups "[PiDP-11]" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to pidp-11+u...@googlegroups.com
> <mailto:pidp-11+u...@googlegroups.com>.
> To view this discussion on the web visit

> https://groups.google.com/d/msgid/pidp-11/323679ea-77eb-4e18-8816-4b237288cd84n%40googlegroups.com <https://groups.google.com/d/msgid/pidp-11/323679ea-77eb-4e18-8816-4b237288cd84n%40googlegroups.com?utm_medium=email&utm_source=footer>.

[Johnny Billquist]

You actually have them on TTYs? Yikes.

The serial port bit should be easy, I think, but if you aim for TTYs,
then yeah, that will be expensive.

And the upper limit on users on RSTS/E is 63. Just so you know. :-)

Johnny

On 2024-08-18 10:13, Thomas Lake wrote:
> Heh, heh. If you're willing to pay for them, I'd be willing to add the
> necessary USB to RS-232 converters and terminals to add hundreds of users!
>
> Right now, however, eight is enough (not the TV show!) although I am
> getting a lot more applications these days. I just might expand when the
> number of applicants
> allows me to buy the hardware. Have you seen the price of working
> Teletypes these days? It's a good thing I know a guy in southern Vermont
> who services them, although
> he may not for much longer.
> ------------------------------------------------------------------------
> *From:* pid...@googlegroups.com <pid...@googlegroups.com> on behalf of
> Johnny Billquist <b...@softjar.se>
> *Sent:* Saturday, August 17, 2024 9:00 PM
> *To:* pid...@googlegroups.com <pid...@googlegroups.com>
> *Subject:* Re: [PiDP-11] Re: What do use a PiPD-11 for...

> Eight??? What has the world come to...? Back in my time, we were usually
> around 40 users online and active on RSTS/E at the same time, on an
> 11/70. When it went up to 60, it was really slow...
> (Said with a smile, but the facts are correct...)
>
>    Johnny
>
> On 2024-08-17 11:36, Tom Lake wrote:
>> I have eight user capability on my PiPD-11 running RSTS/E 10.1. I
>> usually have three or four students at a time running BASIC while I
>> teach programming techniques.
>>
>> On Friday, August 9, 2024 at 7:16:25 AM UTC-4 guit...@gmail.comwrote:
>>
>>     I'm all about blinky lights and retro emulation. But I'm curious
>>     what folks use these for on their home networks. I'm not above
>>     getting one for the novelty - but what can I employ this to do on my
>>     network? Email server, network logging, smart home integration?
>>
>> --
>> You received this message because you are subscribed to the Google
>> Groups "[PiDP-11]" group.
>> To unsubscribe from this group and stop receiving emails from it, send
>> an email to pidp-11+u...@googlegroups.com
>> <mailto:pidp-11+u...@googlegroups.com
> <mailto:pidp-11+u...@googlegroups.com>>.
>> To view this discussion on the web visit

>> https://groups.google.com/d/msgid/pidp-11/fcaea31b-b698-4d52-9c64-f6f220cadc89n%40googlegroups.com <https://groups.google.com/d/msgid/pidp-11/fcaea31b-b698-4d52-9c64-f6f220cadc89n%40googlegroups.com> <https://groups.google.com/d/msgid/pidp-11/fcaea31b-b698-4d52-9c64-f6f220cadc89n%40googlegroups.com?utm_medium=email&utm_source=footer <https://groups.google.com/d/msgid/pidp-11/fcaea31b-b698-4d52-9c64-f6f220cadc89n%40googlegroups.com?utm_medium=email&utm_source=footer>>.

>
> --
> Johnny Billquist                  || "I'm on a bus
>                                    ||  on a psychedelic trip
> email: b...@softjar.se             ||  Reading murder books
> pdp is alive!                     ||  tryin' to stay hip" - B. Idol
>

> --
> You received this message because you are subscribed to a topic in the
> Google Groups "[PiDP-11]" group.
> To unsubscribe from this topic, visit
> https://groups.google.com/d/topic/pidp-11/DikJzgYwVSM/unsubscribe
> <https://groups.google.com/d/topic/pidp-11/DikJzgYwVSM/unsubscribe>.
> To unsubscribe from this group and all its topics, send an email to
> pidp-11+u...@googlegroups.com.

> To view this discussion on the web visit

> https://groups.google.com/d/msgid/pidp-11/dc5d8bc3-d855-4f5f-8642-30a8855484e3%40softjar.se <https://groups.google.com/d/msgid/pidp-11/dc5d8bc3-d855-4f5f-8642-30a8855484e3%40softjar.se>.

[Johnny Billquist]

I realize I maybe should answer this one as well...

As for what specifically my PiDP-11 is used for, I use it as a testing
machine. It's setup to run RSX with just TCP/IP and no DECnet, and
talking over WiFi (it's complicated). When I do new builds of TCP/IP,
this is one of the machines I am testing on, as it has a little odd
setup, and I want to make sure things work there too.

But I have several machine, including real ones (an 11/93 sitting right
next to me among others). I do development of networking protocols,
daemons and clients, I also develop and help others with issues around
PDP-11, occasionally even professionally.

The perhaps most interesting one for most people are not a PiDP-11, but
still, an emulated PDP-11/74 which is available online, and which have
my RPM server on it, runs an ftp server, the web server with all RSX
manuals (and some related stuff) available, can be logged into by guest
users, and is also my main development platform for any RSX work I do.
It also acts as a relay server for my SMTP traffic when ISPs are being
stupid, and it is a cookie server, as well as a HECnet-internet mail
relay server, the DECnet nodename central database server (also with a
web interface), and a connection point that I can use to connect new
users to HECnet, if needed.

It's probably used for a few more things as well, which I probably
forgot right now. The address is mim.stupi.net, by the way. Telnet, ftp,
http, mail, cookies (telnet to port 8445 and give an empty line).

Johnny

> --
> You received this message because you are subscribed to the Google
> Groups "[PiDP-11]" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to pidp-11+u...@googlegroups.com
> <mailto:pidp-11+u...@googlegroups.com>.
> To view this discussion on the web visit

> https://groups.google.com/d/msgid/pidp-11/7c121075-ba5a-48fd-8c14-49a7f6f1ba11n%40googlegroups.com <https://groups.google.com/d/msgid/pidp-11/7c121075-ba5a-48fd-8c14-49a7f6f1ba11n%40googlegroups.com?utm_medium=email&utm_source=footer>.

[terry-...@glaver.org]

On Sunday, August 18, 2024 at 5:03:57 AM UTC-4 b...@softjar.se wrote:

On 2024-08-18 08:01, terry-...@glaver.org wrote:
> 8 users on a DZ-11 is horrible. 8 users on a DH-11 will barely be
> noticed by the CPU. Of course, the 11/93 has 8 DL-11 ports, but
> they're emulated by a microcontroller and have have FIFOs, so
> they aren't as unbearable as you might think. On the PiDP-11 it
> is all emulation, so the differences between "horrible" and "good"
> multiplexors is a lot less obvious. I'm running my 4 physical ser-
> ial ports as a DHU11 emulation, and even autobaud works (most
> of the time, on a subset of baud rates) which is pretty amazing
> given the numbers of things between the terminal and RSTS/E,
> compared to an actual 11 - CH340 USB/serial, Raspberry Pi OS,
> simh, etc.

The 8 DL ports on the 11/93 are just as horrible to the CPU as any other
DL11 on any other CPU. The fact that the hardware have buffering don't
change anything from the PDP-11 CPU point of view. It only makes it
better in the sense that the risk of loosing characters are less.

  DEC (or at least ROI) documented changing the buffer size, implement-

ing local echo, various delimiters, etc. All with Big Warnings that any of

this was Completely Unsupported.

  I could see some large customer doing that on their own proprietary

operating system/application.

  By the time the 11/93 came out, at least the RSTS Engineering group

(I can't say for sure about the other PDP-11 OS groups) was being

driven, to a large extent, by "large customer X wants feature Y and is

willing to pay $Z for it". As an example, RSTS/E got support added to

send a configurable answerback string (up to 15 characters, IIRC)

when receiving an ASCII WRU. This happened while many top-of-

wishlist items from symposia weren't getting addressed, I asked them

about it and they admitted some big customer paid a lot of money

for the feature

For the 11/93 CPUs that went in as upgrades to 11/83 or /84 systems,

people probably put tape over those ports with "DO NOT USE!" on the

tape in big letters.

And yes, I agree that DL11s are horrible, as are DZ11. But it's not as
bad as on a VAX, because the PDP-11 is much better at handling
interrupts. Running a DZ11 or two on an -11 is still ok. I would
probably not recommend running 8 of them at the same time.

  It didn't help that DEC sold the DZ11 as the "standard" multiplexor for

systems at least through the 11/44 era. That's probably because the

DH11 of that era was a whole 9-slot backplane full of cards. And on

an 11/70, that usually meant another BA11 to hold it.

 

  Emulex came out with a DH-emulating gizmo that was one slot

(IIRC) and used powered distribution panels to implement as many

ports as the customer wanted.

  DEC had 2 responses to that - the DHU11 which got you 16 ports

on a single slot, and an oddball gizmo that I don't remember the name

of. The oddball was a hex Unibus board that spoke DH on the CPU side

and almost-but-not-quite T1 on the other side. It was connected to a

powered distribution unit that could be located thousands of feet away

from the CPU (presumbaly in industrial environments). I don't recall

what it was called (either model or M-number). Can someone jog my

memory?

As for automatically detecting the speed, I don't know how RSTS/E does
it, but on RSX, it actually depends on the specifics of the hardware, so
if it works or not on emulation is very much a case of "luck". RSX
autobaud detection handles DZ different than DH for example.

  I'd have to look and see what it does, specifically. I know it looks at the

overrun bit and probably some other error bits to decide the next speed

to try. IIRC, it took 3 CR's maximum to get the speed right on real hard-

ware.

  The Gandalf concentrators used a Break signal to get the attention of

the fabric controller. It might take it a few seconds to get around to it,

so we put up signs that said "Break, 1-potato, 2-potato, 3-potato, Return"

to give people a cadence that would work - if nothing happened after

around 7 seconds, the fabric controller went back to doing other things

(after all, it was supervising 512 terminal ports and 128 host ports).

Once you had its attention, the Return keypress bitstream was looked

at to see how long a character frame was, and it always got the baud

rate correct. The whole fabric was just bit-shifting.

> At SPC, we moved from Gandalf front-end concentrators to DEC-
> server 550s (one full of CXY08 cards and the other full of CXA16
> cards), so we didn't have to keep track of physical ports on CPUs
> any more.

Yeah. That is rather nice. LAT is a pretty good protocol for keeping
host resource usage down.

Yup - by the time the seriously good serial controllers came out, many

sites were already running (or considering) Ethernet, even if it was just

a bunch of systems plugged into a DELNI at first. So, many sites used

termninal servers to connect to the host over Ethernet. Not that either

the original DEUNA or DEQNA were much to brag about, but at 10Mbit

speeds they got the job mostly done.

  We used Micom condentrators (16 serial ports onto a 4800 baud sync

leased line). One day, response times suddenly jumped (on a hard-copy

terminal, if you pressed a key and didn't have the character echo and print

almost immediately, it becomes rather unnerving). A call to the phone 

company got a response of "we moved that to satellite". For a circuit

that was 11 miles long. I asked "What ground station is the uplink?" and

they said "ABC123" or some other CLLI code. I said "And the downlink?"

and they answered "ABC123". I said "Can you see the problem here?"

22,500 miles up and another 22,500 back down, just to land in the same

facility that it went up from. I have NO idea what they were thinking.

> Most of our VAXen only had the console and a single DMF32
> (combo 8 async / 1 bsync / 1 LP11) which got used to drive
> the attached printer (there were additional shared printers on
> DECserver 250s) and the sync port to talk to BITNET via Jnet.

Yeah. That was a popular combo. The DMF32 is actually pretty nice. Good
device for VAXen.

  Yup. Definitely beat a combination of LP11 and some synchronous

interface. Of course, once the first Cisco router arrived on campus, the

sync ports pretty much instantly became useless (you could run BITNET

through the Cisco's leased line). Likewise for the async ports and their

Telebit Trailblazer modems (for UUCP).

[Johnny Billquist]

Just a few comments on some things that might be interesting.

On 2024-08-18 23:04, terry-...@glaver.org wrote:
>
>
> On Sunday, August 18, 2024 at 5:03:57 AM UTC-4 b...@softjar.se wrote:
>
> On 2024-08-18 08:01, terry-...@glaver.org wrote:
> > 8 users on a DZ-11 is horrible. 8 users on a DH-11 will barely be
> > noticed by the CPU. Of course, the 11/93 has 8 DL-11 ports, but
> > they're emulated by a microcontroller and have have FIFOs, so
> > they aren't as unbearable as you might think. On the PiDP-11 it
> > is all emulation, so the differences between "horrible" and "good"
> > multiplexors is a lot less obvious. I'm running my 4 physical ser-
> > ial ports as a DHU11 emulation, and even autobaud works (most
> > of the time, on a subset of baud rates) which is pretty amazing
> > given the numbers of things between the terminal and RSTS/E,
> > compared to an actual 11 - CH340 USB/serial, Raspberry Pi OS,
> > simh, etc.
>
> The 8 DL ports on the 11/93 are just as horrible to the CPU as any
> other
> DL11 on any other CPU. The fact that the hardware have buffering don't
> change anything from the PDP-11 CPU point of view. It only makes it
> better in the sense that the risk of loosing characters are less.
>
>
>   DEC (or at least ROI) documented changing the buffer size, implement-
> ing local echo, various delimiters, etc. All with Big Warnings that anyof
> this was Completely Unsupported.

I don't even remember those added things, but anyway, unless you wrote
something very specific for the 11/93, it was just like any other DL11,
just with a less risk of dropping characters.

> And yes, I agree that DL11s are horrible, as are DZ11. But it's notas
> bad as on a VAX, because the PDP-11 is much better at handling
> interrupts. Running a DZ11 or two on an -11 is still ok. I would
> probably not recommend running 8 of them at the same time.
>
>
>   It didn't help that DEC sold the DZ11 as the "standard" multiplexor for
> systems at least through the 11/44 era. That's probably because the
> DH11 of that era was a whole 9-slot backplane full of cards. And on
> an 11/70, that usually meant another BA11 to hold it.

Well. It was also what you almost always saw in VAX-11/750... I was at
DEC around 1984 or so (not working there at that time), and the training
systems were running 11/750s with DZ11. Me and some friends were allowed
to use those systems at night, and we played games, and 3-4 users on one
11/750 and the system was almost dead.

>   Emulex came out with a DH-emulating gizmo that was one slot
> (IIRC) and used powered distribution panels to implement as many
> ports as the customer wanted.

The distribution panel is not powered, but otherwise yeah. It's a very
nice alternative to the DH11. Only partial modem support, but then
again, I don't know many who ever cared about full modem support. It was
yet another few cards in the DH11 to get that, which I've never actually
seen.

> As for automatically detecting the speed, I don't know how RSTS/E does
> it, but on RSX, it actually depends on the specifics of the
> hardware, so
> if it works or not on emulation is very much a case of "luck". RSX
> autobaud detection handles DZ different than DH for example.
>
>
>   I'd have to look and see what it does, specifically. I know it looks
> at the
> overrun bit and probably some other error bits to decide the next speed
> to try. IIRC, it took 3 CR's maximum to get the speed right on real hard-
> ware.

In RSX, the port is set at 4800 bps, and then when people hit enter,
you'll get "something". And then there are lookup tables what a CR looks
like for different transmission speeds, while the PDP side is running at
4800. And the value differs between controllers. Which is why it might
be very tricky to get it working right in an emulator with some other
serial port hardware.

>   The Gandalf concentrators used a Break signal to get the attention of
> the fabric controller. It might take it a few seconds to get around to it,
> so we put up signs that said "Break, 1-potato, 2-potato, 3-potato, Return"
> to give people a cadence that would work - if nothing happened after
> around 7 seconds, the fabric controller went back to doing other things
> (after all, it was supervising 512 terminal ports and 128 host ports).
> Once you had its attention, the Return keypress bitstream was looked
> at to see how long a character frame was, and it always got the baud
> rate correct. The whole fabric was just bit-shifting.

Yikes. That's a bit complicated.

>   We used Micom condentrators (16 serial ports onto a 4800 baud sync
> leased line). One day, response times suddenly jumped (on a hard-copy
> terminal, if you pressed a key and didn't have the character echo and print
> almost immediately, it becomes rather unnerving). A call to the phone
> company got a response of "we moved that to satellite". For a circuit
> that was 11 miles long. I asked "What ground station is the uplink?" and
> they said "ABC123" or some other CLLI code. I said "And the downlink?"
> and they answered "ABC123". I said "Can you see the problem here?"
> 22,500 miles up and another 22,500 back down, just to land in the same
> facility that it went up from. I have NO idea what they were thinking.

:-)
I have no idea what concentrators we were using, but it would be
interesting to find out.

Johnny

[Brian Denley]

In 1980, we won a production contract for some military hardware (infrared detectors). We had to set up a production line for volume assembly and test (hundreds per month). We chose PDP 11/03 systems: 5 of them, with RL02 drives and VT102 terminals.  These and all our data terminals connected to an 11/70 system for data storage.  The chosen software was FORTRAN 77 which I hated.  I was a Pascal fan at the time.

We ran this line for over 20 years for the US Army.

Brian

Brian Denley

KB1VBF

Sent from my iPad

On Aug 9, 2024, at 7:16 AM, Eric Clark <guit...@gmail.com> wrote:

I'm all about blinky lights and retro emulation. But I'm curious what folks use these for on their home networks. I'm not above getting one for the novelty - but what can I employ this to do on my network? Email server, network logging, smart home integration? 

--
You received this message because you are subscribed to the Google Groups "[PiDP-11]" group.

To unsubscribe from this group and stop receiving emails from it, send an email to pidp-11+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/pidp-11/7c121075-ba5a-48fd-8c14-49a7f6f1ba11n%40googlegroups.com.

[Heinz-Bernd Eggenstein]

Among other things, I use the PiDP-11 and the Pi hugging its panel in 

combination to alert me when the next star in our galaxy collapses to a 

black hole or a neutron star. 

I think I've described this before but since the question of "what to do 

with this" came up again:

The emulated PDP-11 (under 2.11BSD) will execute a just slightly 

edited TCP/IP client software 

https://gcn.gsfc.nasa.gov/gcn/voevent_client_demo.c

continuously. This is legacy code to use a legacy protocol to access the 

GCN network of NASA that shares alerts for certain astronomical events

in real-time. One of the messages carried by GCN will inform us about 

certain types of supernovae in the Milky Way (triggered by neutrino detectors). 

This is expected to happen once ever ca 30..50 years only, so you don't want to 

miss it.

I wanted to make an alarm sound when this happens, but could not find a direct 

way to let the PDP-11 do that, so instead the PDP-11 will write a file locally that 

can be queried by an HTTP server running on the PDP-11.

The Raspi host will periodically lookup that file via HTTP-get, and if it finds a 

new one, it sounds an alarm.

 https://youtu.be/Ta87XFJVwf0

There are test alerts issued periodically, so I can actually enjoy that sound once a week.

 

 Other than that, I use it during the day as GUI-detox therapy (maintaining 

TODO lists with JOVE and tracking my 2 cats weight with sc, etc).

 

During early morning, the PDP-11 will also download via FTP some time series data 

from three amateur seismic detectors (Raspberry Shakes) nearby (one in my 

own basement) that a script on a Raspberry Pi had earlier downloaded from an 

Internet service and put on an FTP server in the LAN. The PDP-11 will then run a 

Butterworth low-pass filter in C (FORTRAN would be cooler) over that 3 x 24 hours data, 

will then generate earth quake candidate triggers based on a very simple 

heuristic comparing the  ratio of long-term to short term average power of the filtered 

signal, and will finally look for coincidences among the  detectors to distinguish 

true earthquakes from things like road traffic or other human made seismic 

signals (e.g. washing machines).

 
It will then generate retro-style reports like this one using nroff
 
 http://bikeman.selfhost.eu/rshake/rep-coinc-2024-08-18.pdf
 

To generate a nicer looking report, it sends the report data via uucp to a SIMH 

instance running a VAX-11/780 under 4.3BSD where a better version of troff 

generates a Postscript report, that is converted to PDF via a CUPS print server 

on a Raspberry Pi. 

   
http://bikeman.selfhost.eu/rshake/rep-coinc-2024-08-18-pretty.pdf
 

The PDP-11 is also running a script that picks a daily "Motto of the day", appends it to 

the remote uptime  (ruptime) of 2.11BSSD and sends it over telnet to a PiDP-8 where 

it is displayed as a ticker-tape-message on an LED-matrix-display  because you cannot 

have enough blinking lights .

 
 http://bikeman.selfhost.eu/astro/VID_20220517_231138.mp4
 

The web server on the PDP-11 is also running a little CGI script: you can enter celestial 

coordinates into a web form and the script will then compute when in the next 24 hours 

that position would be observable at my own site and at any of the iTelescope.net 

remote telescope sites, so you can plan observations and reserve telescope time in advance.

 
  
HB

[Thomas Pfarr]

I worked at NASA Goddard Space Flight Center supporting scientific spacecrarft in the 1970's.  Most of my programming work was done initially on IBM 360 MVT systems in Fortran and Assembler supporting custom code that commanded the spacecraft to perform the science mission.  In the late 70's NASA wanted to retire the IBM 360/65 we were using and wanted the next mssion to be supported by PDP 11/70 systems running rsx-11m version 3.x at the time as I remember it.  The type of programs we developed required dynamic file creation and reading which was problematic on the IBM (simulated with partitioned data set i/o packages).  Going to rsx-11 was like  dying and going to heaven with files and especially interactive terminals for development.  A long story short we developed a new system in 1/3 the time on the PDP (6 months on PDP versus 18 months typical on IBM).  The only joke on the PDP is whenever we had a delivery of a new release the system pack would crash the day before the delivery.  Of course we backed up everything the day before that allways.  Love those disk crashes.  From there I worked on the Hubble Space Telescope on VAX 11/780., 785's and an 8650 finally before launch.  We started without Ethernet and Clusters.  I built the first cluster at GSFC with 3 11/780's must have been 1983 or so.  We coded in Fortran using a 16port statistical multiplexer for terminals and printers.  connected via LAT Terminal servers.

Those systems were replaced by Unix/WIndows in 1995 to 2000.  We were the first JAVA conference presenters at the first JAVA world running our demo user interface on windows under the table because the SUN JAVA runtime would not work.

After 2000 I worked on JWST for 22 years!  All done on windows and all my code in TCL.  That code still runs today after 20 some years to support the mission.

I recently found a set of old FORTRAN listings of code I wrote in 1977 to support the High Energy Astronomical Observatory (HEAO/Einstein) and have slowly scanned it all,  typed most in manually and got some to run again on the IBM MVS simulation (TK5) and have gotten most to compile at this point on the pidp11 running rsx11m-plus with F77.  I'm rewriting utility ibm assembler routines on the PDP in various languages for fun.  The Fortran ports to the PDP from IBM with practically no changes.

Lots of fun.

Tom Pfarr

On Friday, August 9, 2024 at 7:16:25 AM UTC-4 guit...@gmail.com wrote:

[Alan B]

When I was at UC Berkeley we had a fairly well equipped 11/70 in Cory Hall for the computer science department that regularly ran about 50 interactive terminal users on early Unix, most doing their CS classwork. The grad students were doing projects on Unix itself improving the performance for this workload and it was much more capable than the computer center machine running the stock Unix version. Ken Thompson was there, though at the time I didn't know who he was. Bill Joy was there. Good old days. :)

To view this discussion on the web visit https://groups.google.com/d/msgid/pidp-11/1a30f91d-f197-4429-9c79-eeb3b465fb16n%40googlegroups.com.

[DR]

Wow, sounds like fun times.  Sometimes we don't always realize how
privileged we were to work on projects which were so important or only a
few people in the world got to do.


I am no IBM expert, but a long shot in my understanding them.

I was under the impression that the 360 series was more business
oriented (COBOL and all that) more than scientific, where mainframes
like the what if I recall correctly, NASA used a 7090 or such.

But apparently the boxes could do whatever they were loaded to do.  Our
university's 360s were all over in the business school and social
science buildings.


I know our Burroughs B5500 had almost zero science usage, more business
so the architecture of a machine can  apparently be tweaked to support
various orientations.


Thanks for your personal notes on your experiences.  We had a guy, now
deceased, in our astronomy club that did the heavy duty programming to
import the GUIDE 8 predictions for observing and then make it track for
long exposures.  He was a previous employee of maybe Lockheed? and wrote
assembler level stuff for satellites that he insisted he just couldn't
talk about or was a way to avoid small talk or chatter.


One of my friends used to help design memory for FabriTek and then went
on to be at Cray Research, knowing just about everything about
everything while there.


And those machines DID a lot of scientific type of computing for
weather, car manufacturers (which surprised the heck out of me), and of
course the things which went on at Los Alamos  and Lawrence Livermore. 
I assume they must have a lot of computing needs but don't hear of big
contracts for those machines much any more.


And I felt awfully blessed to have access to the PDP-8 and spent too
many hours with it (especially after bootstrapping it to run and then
not wanting to turn it off.

[trp...@gmail.com]

Goddard had 5 IBM large 360 systems in the 70's. A 95 (very fast but not reliable) two 75 workhorses and a 65 that was the most reliable and used for day to day satellite operations. Goddard pioneered orbit analysys, spacecraft attitude control and sensor calibrations, science data processing, calibration of instrument data and analysis of science products. Most of this mathematic processing was done on the IBM's. Real time control of spacecraft were all done with early mini-computers (SDS 930, Sigma 5's, followed by PDP11s and Vax). Most current systems are a combination of Unix servers and Windows.

My earliest mentor when I went to work was Sarah Shefield (a British American). She was the lead designer of the earliest systems at GSFC in the late 60's and programmed in IBM assembler. Her code was the most brilliant code I've ever encountered. Simple, elegant, purposeful. A true giant that I was humbled to stand on her shoulders.

--
You received this message because you are subscribed to the Google Groups "[PiDP-11]" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pidp-11+u...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/pidp-11/ad45748f-842b-4e0e-ab66-7aba11a3ae73%40gmail.com.

[Adam Thornton]

NASA had at least one 360/195 at one point.  One of the truly great front panels, even if it was largely useless to the customer (it was mostly tracking the state of the internal microcode, if I remember correctly).

[DR]

that panel looks a lot like a Univac 1108, but twice as wide.

In addition to looks, we had sounds. The thing to do when a solo
operator (night shift usually) was to carry a transistor radio around
where we were and if tuned to the right frequency, it would begin making
certain noises when it went idle.

We didn't know when a job would terminate, short or long, and the time
was so sought after and expensive, they wanted us to keep the jobs
running as much as possible.  If you released more than 5 or 6, it
bogged down with the OS it was running that you actually weren't being
efficient. I started 8 at one time and learned my lesson.

[Richard McDonald]

When I got my BSEE (1972-78) at UMR (now Missouri S&T University) the ground floor of the Computer Science building in Rolla, MO housed an IBM System 360 with a boatload of high speed (256K words?) and low speed (1M words?) core memory and (if memory serves me right) 21 hard disk packs.  Of course it also had the requisite high-speed line printers and card reader/sorter/punches, and 9-track tape drives.  And for the introductory programming courses: a put-the-cards-in-yourself card reader just outside the machine room window where you could load your deck and watch it print your program out (seemingly) immediately.  Magic to a freshman.  

In 1973 the University completed a dedicated (90+ mile) microwave link to the UMC CompSci center in Columbia, MO where they had a System 370.  At that time the University decided in their infinite wisdom that all computing would be done on the 370 and the 360 was converted into essentially a batch processing que for sending jobs to the 370 and printing out the results for distribution to local students.  Excess capability of the 360 was dedicated to long-running grad student programs.  Rolla used 100% of the 360 and something like 85% of the 370.

I remember that the 360 printed out a header page indicating the the system used "HASP-II".  I believe that acronym stood for Houston Automated System Program.  Does anybody know if that was developed for NASA's use at the Manned Spacecraft Center in Houston?

Rich

[Curtis Dean Smith (史國興)]

This all sounds super fun.  Now, if I could only figure out how to set up wireless networking between the RPi and 2.11 BSD!

[terry-...@glaver.org]

The 195 was entirely discrete logic. Not a bit of microcode anywhere in the main CPU. That's one of the things that allowed it to be so fast. But it also meant that CE's (the IBM repair techs) didn't have the ability to load special microdiagnostics to locate the source of a problem. The immense front panel allowed them to "see" into the logic with much more detail than on other machines, because it was necessary for diagnosis.

I also suspect that's why the 370/195 never got virtual memory, although it did pick up the rest of the 370 additions.

To answer an earlier question about scientific vs. commercial machines, the IBM 360 (as in 360 degrees) was designed to cover both markets, while earlier IBM CPUs were targeted either to the commercial or scientific markets.

[Derek]

It's really cool to see all the things people have been using theirs for.

Mine is being used for quite a plethora of things. Firstly, I always wanted to use a mainframe computer since I was younger. I missed it entirely having grown up in the 90s where they were getting very scare to come by. So after I found out about the amazing PiDP-11 and SIMH combo I immediately had to get one.

Since then it's been messing with switches for programming directly into memory (I made a pomodoro timer at the most practical of these experiments), exploring the different Unix systems (always wanted to use true Unix after being a big Linux and Mac/Darwin guy), and as of lately RSX has become a huge playground of exploration and wonder. My PiPD-11 has been booted into it almost exclusively since I dipped my toes in. I've done a variety of things in RSX from BASIC coding, to messing with FTP, and also sending HTTP requests through the TC device thanks to Johnny Billquist's TCP/IP setup. In fact, I didn't even really know how exactly HTTP requests worked until I tried sending one from RSX, so I have it to thank for the understanding the process now. Kinda took away some of the magic though haha. Eventually I'll make a program read the switches on the panel to turn my Philips Hue lights on and off. Been doing some web development and hosting not it too, and just getting accustomed to the OSs specifics like the UFD concepts and OS level file revision history. Really cool stuff. I also downloaded a treasure trove of PDP-11 manuals for the various OSs and hardware from https://bitsavers.computerhistory.org/pdf/dec/pdp11/, and I have been periodically viewing the PDP-11 training course videos here (https://www.youtube.com/playlist?list=PLWgt3n8uQhm35OH13EuIbDO5LxGHZfVZB), so I have 8+ gigs of playing around to go through still.

Honestly it's been a little too much fun and distracting haha.

[Johnny Billquist]

Fun stuff, indeed. If you are running RSX a lot, there are a couple of
things I can add.

First of all, you might find some interesting manuals at
http://mim.stupi.net/manuals.htm, not all of them might even be on
bitsavers (and yes, this is served from an RSX machine).

If you want to read the front panel switches in RSX, there is a system
call do do that:
GSSW$

also available from high level languages, using the FORTRAN calling API.
See HELP EXEC GSSW, or read the manual.

So you did some kind of http client? That's fun. Have you done anything
more "interesting" with around it?

UFDs are just the RSX term for directories. Unfortunately, you can't
have multiple levels of subdirectories. I have patches to allow for
that, but it's a bit of an effort to enable it, and I haven't finished
some tools needed to support it fully in a couple of special applications.

But yes, it is fun (and distracting), but you do learn a lot of stuff as
well.

I would recommend checking in around mim.stupi.net in general, as there
might be more/other things of use there as well...

Johnny

> --
> You received this message because you are subscribed to the Google
> Groups "[PiDP-11]" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to pidp-11+u...@googlegroups.com

> <mailto:pidp-11+u...@googlegroups.com>.

> To view this discussion on the web visit

> https://groups.google.com/d/msgid/pidp-11/ca2325f1-1ce1-494c-ad32-d74a1ece5e15n%40googlegroups.com <https://groups.google.com/d/msgid/pidp-11/ca2325f1-1ce1-494c-ad32-d74a1ece5e15n%40googlegroups.com?utm_medium=email&utm_source=footer>.

[Johnny Billquist]

Argh. Sorry about the address for the manuals. It should be
http://mim.stupi.net/manuals/

>>     logic with much more detail than on other machines,because it was

>>     necessary for diagnosis.
>>
>>     I also suspect that's why the 370/195 never got virtual memory,
>>     although it did pick up the rest of the 370 additions.
>>
>>     To answer an earlier question about scientific vs. commercial
>>     machines, the IBM 360 (as in 360 degrees) was designed to cover both
>>     markets, while earlier IBM CPUs were targeted either to the
>>     commercial or scientific markets.
>>
>>     On Monday, August 19, 2024 at 5:01:30 PM UTC-4 atho...@gmail.com
>> wrote:
>>
>>         NASA had at least one 360/195 at one point.  One of the truly

>>         great front panels, even ifit was largely useless to the

[Bradford Miller]

The 11 was a mini, not a mainframe. If you want a mainframe equivalent, get a PiDP-10!

[Mark Rosenthal]

Another mini operating system without subdirectories is Hp3000 MPE.  Maybe someday Oscar will do a Hp3000 Series III front panel.  :-)

To unsubscribe from this group and stop receiving emails from it, send an email to pidp-11+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/pidp-11/f6bec4f6-448f-4503-a3c6-663190a3b2d0%40softjar.se.

[Derek]

Oh whoops, I had totally skipped over the part of originally going for the PiDP-10. I wanted both it and the PiPD-11 equally, but figured the more mainframe experience would be the most interesting to start with since I didn't get a chance to use a real one. However the PiPD-10 was sold out at the time, so I went with the PiPD-11 as the first of the two. But loving the PDP-11 now and glad I by chance snagged it first (still getting a PiPD-10 at some point too though).

Phew, there we go! Don't mind me doing a little mix up there haha.

[Johnny Billquist]

Well, RSX isn't entirely without subdirectories, but unfortunately, it
only allows a single level of subdirectories under the MFD. It's not a
limitation of the file system as such, but in the parsing and processing
of directories in libraries.

It wasn't too hard to add more levels of subdirectories in RSX in
general. It's just that directory wildcarding in PIP is a somewhat
special case that I haven't fixed yet.

Johnny

On 2024-08-21 00:16, Mark Rosenthal wrote:
> Another mini operating system without subdirectories is Hp3000 MPE.
> Maybe someday Oscar will do a Hp3000 Series III front panel.  :-)
>
> On Tue, Aug 20, 2024, 1:54 PM Johnny Billquist <b...@softjar.se
> <mailto:b...@softjar.se>> wrote:
>
> Fun stuff, indeed. If you are running RSX a lot, there are a coupleof
> things I can add.
>
> First of all, you might find some interesting manuals at

> http://mim.stupi.net/manuals.htm <http://mim.stupi.net/manuals.htm>,

> not all of them might even be on
> bitsavers (and yes, this is served from an RSX machine).
>
> If you want to read the front panel switches in RSX, there is a system
> call do do that:
> GSSW$
>
> also available from high level languages, using the FORTRAN calling
> API.
> See HELP EXEC GSSW, or read the manual.
>
> So you did some kind of http client? That's fun. Have you done anything
> more "interesting" with around it?
>
> UFDs are just the RSX term for directories. Unfortunately, you can't
> have multiple levels of subdirectories. I have patches to allow for
> that, but it's a bit of an effort to enable it, and I haven't finished
> some tools needed to support it fully in a couple of special
> applications.
>
> But yes, it is fun (and distracting), but you do learn a lot of
> stuff as
> well.
>
> I would recommend checking in around mim.stupi.net

> <http://mim.stupi.net> in general, as there

> <https://bitsavers.computerhistory.org/pdf/dec/pdp11/>, and I have

> > been periodically viewing the PDP-11 training course videos here
> >

> (https://www.youtube.com/playlist?list=PLWgt3n8uQhm35OH13EuIbDO5LxGHZfVZB <https://www.youtube.com/playlist?list=PLWgt3n8uQhm35OH13EuIbDO5LxGHZfVZB>), so I have 8+ gigs of playing around to go through still.

> >
> > Honestly it's been a little too much fun and distracting haha.
> > On Tuesday, August 20, 2024 at 10:08:13 AM UTC-5

> terry-...@glaver.org <mailto:terry-...@glaver.org> wrote:
> >
> >     The 195 was entirely discrete logic. Not a bit of microcode
> anywhere
> >     in the main CPU. That's one of the things that allowed it to
> be so
> >     fast. But it also meant that CE's (the IBM repair techs)
> didn't have
> >     the ability to load special microdiagnostics to locate the
> source of
> >     a problem. The immense front panel allowed them to "see" into the
> >     logic with much more detail than on other machines, because
> it was
> >     necessary for diagnosis.
> >
> >     I also suspect that's why the 370/195 never got virtual memory,
> >     although it did pick up the rest of the 370 additions.
> >
> >     To answer an earlier question about scientific vs. commercial

> >     machines, the IBM 360 (as in 360 degrees) wasdesigned to
> cover both
> >     markets, while earlier IBM CPUs were targetedeither to the

> >     commercial or scientific markets.
> >
> >     On Monday, August 19, 2024 at 5:01:30 PM UTC-4

> atho...@gmail.com <mailto:atho...@gmail.com> wrote:
> >
> >         NASA had at least one 360/195 at one point.  One of the truly
> >         great front panels, even if it was largely useless to the
> >         customer (it was mostly tracking the state of the internal
> >         microcode, if I remember correctly).
> >
> > --
> > You received this message because you are subscribed to the Google
> > Groups "[PiDP-11]" group.
> > To unsubscribe from this group and stop receiving emails from it,
> send
> > an email to pidp-11+u...@googlegroups.com

> <mailto:pidp-11%2Bunsu...@googlegroups.com>
> > <mailto:pidp-11+u...@googlegroups.com
> <mailto:pidp-11%2Bunsu...@googlegroups.com>>.

> > To view this discussion on the web visit
> >

> https://groups.google.com/d/msgid/pidp-11/ca2325f1-1ce1-494c-ad32-d74a1ece5e15n%40googlegroups.com <https://groups.google.com/d/msgid/pidp-11/ca2325f1-1ce1-494c-ad32-d74a1ece5e15n%40googlegroups.com> <https://groups.google.com/d/msgid/pidp-11/ca2325f1-1ce1-494c-ad32-d74a1ece5e15n%40googlegroups.com?utm_medium=email&utm_source=footer <https://groups.google.com/d/msgid/pidp-11/ca2325f1-1ce1-494c-ad32-d74a1ece5e15n%40googlegroups.com?utm_medium=email&utm_source=footer>>.

>
> --
> Johnny Billquist                  || "I'm on a bus
>                                    ||  on a psychedelic trip

> email: b...@softjar.se <mailto:b...@softjar.se>             ||

> Reading murder books
> pdp is alive!                    ||  tryin' to stay hip" - B. Idol
>
> --
> You received this message because you are subscribed to the Google
> Groups "[PiDP-11]" group.
> To unsubscribe from this group and stop receiving emails from it,
> send an email to pidp-11+u...@googlegroups.com

> <mailto:pidp-11%2Bunsu...@googlegroups.com>.

> To view this discussion on the web visit

> https://groups.google.com/d/msgid/pidp-11/f6bec4f6-448f-4503-a3c6-663190a3b2d0%40softjar.se <https://groups.google.com/d/msgid/pidp-11/f6bec4f6-448f-4503-a3c6-663190a3b2d0%40softjar.se>.

[Derek]

Thanks for the info Johnny! The manuals and reading the front panel switches will definitely be helpful. I love C so it'd be my go to, but I'd also enjoy exploring others like FORTRAN, Pascal, and more assembly as well (even besides boot.ini or switch panel programs).

I also have yet to get much into the peripherals too and very interested. I barely setup extra disks and a tape reader/puncher once (to punch a 'hello world' BASIC program to paper). There's still lots of display types to play with, and more exploration with the paper tapes and cassette tapes too. Even as a kid in the 90s when I saw references to cassette tapes somehow being used to store data instead of just music, I had always wanted to play around with it and use it. And naturally I'll have to explore Lunar Lander and the PDP-11 version of Spacewar. And I still need to get Adventure working since using SAVE causes a crash on each of the OSs I've tried on the PiPD-11  - perhaps it's expecting a floppy disk to save to?

Oh I didn't have anything quite that fancy as a HTTP client haha. I was just using Telnet from RSX to confirm and send commands... then I tried to adapt the network example from the PiPD-11 manual where a text file was sent to a printer's 9100 port using PIP TC. I was able to successfully PIP TC the HTTP request text file to my Mac (which was waiting to hear from the RSX via netcat). Though so far I haven't gotten my Philips Hue lights to notice RSX sending PIP TC commands, though RSX's Telnet will correctly turn lights on and off as expected. At least thanks to your tip, now I know how I can read the switch states for that eventual program. ;)

And even so, I still need to deep dive further into the older Unix versions, perhaps even bring over the old Doom Linux source and try to get it up on one of them. And I'd love to install Ultrix and RSX-11M+ on my systems fresh. The default RSX-11M+ drive with the PiPD-11system images seems to come with a LOT of directories, even an empty one named CHRIS. So I'd like to setup a system fresh to see which of these directories come with the start. So far in the RSX-11M+ quick reference manual had a list of important systems UFDs which has been really handy to know.

[Johnny Billquist]

Just a quickie... Check http://mim.stupi.net/pidp.htm, and use that
image. Install adventure using RPM, and it should work just fine. You
can also install DUNGEON (ZORK) that way, as well as ZEMU, which will
give you the Infocom ZORK releases.

That disk image is pretty much a clean RSX-11M+ with lots of patches and
fixes applied, as well as already having DECnet, TCP/IP and RPM
installed. But nothing else. So it's a very clean system to start from
scratch with, without having to go through a bunch of hoops. And also,
most of the patches and fixes I've done are not available in any other way.

Johnny

> http://mim.stupi.net/manuals.htm <http://mim.stupi.net/manuals.htm>,

> not all of them might even be on
> bitsavers (and yes, this is served from an RSX machine).
>
> If you want to read the front panel switches in RSX, there is a system
> call do do that:
> GSSW$
>
> also available from high level languages, using the FORTRAN calling
> API.
> See HELP EXEC GSSW, or read the manual.
>
> So you did some kind of http client? That's fun. Have you done anything
> more "interesting" with around it?
>
> UFDs are just the RSX term for directories. Unfortunately, you can't
> have multiple levels of subdirectories. I have patches to allow for
> that, but it's a bit of an effort to enable it, and I haven't finished
> some tools needed to support it fully in a couple of special
> applications.
>
> But yes, it is fun (and distracting), but you do learn a lot of
> stuff as
> well.
>
> I would recommend checking in around mim.stupi.net

> <http://mim.stupi.net> in general, as there

> <https://bitsavers.computerhistory.org/pdf/dec/pdp11/>, and I have

> > been periodically viewing the PDP-11 training course videos here
> >

> (https://www.youtube.com/playlist?list=PLWgt3n8uQhm35OH13EuIbDO5LxGHZfVZB <https://www.youtube.com/playlist?list=PLWgt3n8uQhm35OH13EuIbDO5LxGHZfVZB>), so I have 8+ gigs of playing around to go through still.

> https://groups.google.com/d/msgid/pidp-11/ca2325f1-1ce1-494c-ad32-d74a1ece5e15n%40googlegroups.com <https://groups.google.com/d/msgid/pidp-11/ca2325f1-1ce1-494c-ad32-d74a1ece5e15n%40googlegroups.com> <https://groups.google.com/d/msgid/pidp-11/ca2325f1-1ce1-494c-ad32-d74a1ece5e15n%40googlegroups.com?utm_medium=email&utm_source=footer <https://groups.google.com/d/msgid/pidp-11/ca2325f1-1ce1-494c-ad32-d74a1ece5e15n%40googlegroups.com?utm_medium=email&utm_source=footer>>.

>
> --
> Johnny Billquist || "I'm on a bus
> || on a psychedelic trip
> email: b...@softjar.se || Reading murder books
> pdp is alive! || tryin' to stay hip" - B. Idol
>

> --
> You received this message because you are subscribed to the Google
> Groups "[PiDP-11]" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to pidp-11+u...@googlegroups.com
> <mailto:pidp-11+u...@googlegroups.com>.
> To view this discussion on the web visit

> https://groups.google.com/d/msgid/pidp-11/c74a324f-09cf-4418-ada0-0f505c4f8e59n%40googlegroups.com <https://groups.google.com/d/msgid/pidp-11/c74a324f-09cf-4418-ada0-0f505c4f8e59n%40googlegroups.com?utm_medium=email&utm_source=footer>.

[terry-...@glaver.org]

If you're going to do a HP 3000 front panel, it has to be the Classic: https://www.hewlettpackardhistory.com/item/a-legend-is-born/

I actually saw one of those running (as well as it could, anyway) in the early 1970's. That HP link definitely looks at the Classic through rose-tinted glasses. For a more realistic view, I suggest: http://www.robelle.com/library/smugbook/classic.html

[Ed Tottenham]

Hi Johnny,

I'm trying to work out if the implication of that is that you have the source code to PIP and other RSX utilities

Ed

Get BlueMail for Android

[Johnny Billquist]

Yes I do, and no, I can't share. I actually have a contract with XX2247
to have access to sources and do stuff.

Johnny

On 2024-08-21 13:49, Ed Tottenham wrote:
> Hi Johnny,
>
> I'm trying to work out if the implication of that is that you have the
> source code to PIP and other RSX utilities
>
> Ed
>

> Get BlueMail for Android <https://bluemail.me>

> On 20 Aug 2024, at 23:28, Johnny Billquist <b...@softjar.se

> <http://mim.stupi.net <http://mim.stupi.net>> in general, as there

> <https://bitsavers.computerhistory.org/pdf/dec/pdp11>/
>
> <https://bitsavers.computerhistory.org/pdf/dec/pdp11
> <https://bitsavers.computerhistory.org/pdf/dec/pdp11>/>, and I have

>
> been periodically viewing the PDP-11 training course videoshere
>
>

> (https://www.youtube.com/playlist?list=PLWgt3n8uQhm35OH13EuIbDO5LxGHZfVZB <https://www.youtube.com/playlist?list=PLWgt3n8uQhm35OH13EuIbDO5LxGHZfVZB> <https://www.youtube.com/playlist?list=PLWgt3n8uQhm35OH13EuIbDO5LxGHZfVZB <https://www.youtube.com/playlist?list=PLWgt3n8uQhm35OH13EuIbDO5LxGHZfVZB>>), so I have 8+ gigs of playing around to go through still.

>
>
> Honestly it's been a little too much fun and distracting haha.
> On Tuesday, August 20, 2024 at 10:08:13 AM UTC-5
>
> terry-...@glaver.org <mailto:terry-...@glaver.org> wrote:
>
>
>      The 195 was entirely discrete logic. Not a bit of
> microcode
>
> anywhere
>
>      in the main CPU. That's one of the things that allowed
> it to
>
> be so
>

>      fast. But it also meant that CE's (theIBM repair techs)

>
> didn't have
>
>      the ability to load special microdiagnostics to locate the
>
> source of
>
>      a problem. The immense front panel allowed them to
> "see" into the
>      logic with much more detail than on other machines,
> because
>
> it was
>
>      necessary for diagnosis.
>
>      I also suspect that's why the 370/195 never got
> virtual memory,
>      although it did pick up the rest of the 370 additions.
>
>      To answer an earlier question about scientific vs.
> commercial
>      machines, the IBM 360 (as in 360 degrees) wasdesigned to
>
> cover both
>
>      markets, while earlier IBM CPUs were targetedeither to the
>      commercial or scientific markets.
>
>      On Monday, August 19, 2024 at 5:01:30 PM UTC-4
>
> atho...@gmail.com <mailto:atho...@gmail.com> wrote:
>
>
>          NASA had at least one 360/195 at one point.  One
> of the truly

>          great front panels, evenif it was largely useless

> to the
>          customer (it was mostly tracking the state of the
> internal
>          microcode, if I remembercorrectly).
>
> --
> You received this message because you are subscribed to the
> Google
> Groups "[PiDP-11]" group.
> To unsubscribe from this group and stop receiving emails
> from it,
>
> send
>
> an email to pidp-11+u...@googlegroups.com
>
> <mailto:pidp-11%2Bunsu...@googlegroups.com>
>
> <mailto:pidp-11+u...@googlegroups.com
>
> <mailto:pidp-11%2Bunsu...@googlegroups.com>>.
>
> To view this discussion on the web visit
>
>

> https://groups.google.com/d/msgid/pidp-11/ca2325f1-1ce1-494c-ad32-d74a1ece5e15n%40googlegroups.com <https://groups.google.com/d/msgid/pidp-11/ca2325f1-1ce1-494c-ad32-d74a1ece5e15n%40googlegroups.com> <https://groups.google.com/d/msgid/pidp-11/ca2325f1-1ce1-494c-ad32-d74a1ece5e15n%40googlegroups.com <https://groups.google.com/d/msgid/pidp-11/ca2325f1-1ce1-494c-ad32-d74a1ece5e15n%40googlegroups.com>> <https://groups.google.com/d/msgid/pidp-11/ca2325f1-1ce1-494c-ad32-d74a1ece5e15n%40googlegroups.com?utm_medium=email&utm_source=footer <https://groups.google.com/d/msgid/pidp-11/ca2325f1-1ce1-494c-ad32-d74a1ece5e15n%40googlegroups.com?utm_medium=email&utm_source=footer> <https://groups.google.com/d/msgid/pidp-11/ca2325f1-1ce1-494c-ad32-d74a1ece5e15n%40googlegroups.com?utm_medium=email&utm_source=footer <https://groups.google.com/d/msgid/pidp-11/ca2325f1-1ce1-494c-ad32-d74a1ece5e15n%40googlegroups.com?utm_medium=email&utm_source=footer>>>.

>
> --
> Johnny Billquist                 || "I'm on a bus
>                                   ||  on a psychedelic trip
> email: b...@softjar.se <mailto:b...@softjar.se>            ||
> Reading murder books
> pdp is alive!                    ||  tryin' to stay hip" - B. Idol
>
> --
> You received this message because you are subscribed to the Google
> Groups "[PiDP-11]" group.
> To unsubscribe from this group and stop receiving emails from it,
> send an email to pidp-11+u...@googlegroups.com
> <mailto:pidp-11%2Bunsu...@googlegroups.com>.
> To view this discussion on the web visit

> https://groups.google.com/d/msgid/pidp-11/f6bec4f6-448f-4503-a3c6-663190a3b2d0%40softjar.se <https://groups.google.com/d/msgid/pidp-11/f6bec4f6-448f-4503-a3c6-663190a3b2d0%40softjar.se> <https://groups.google.com/d/msgid/pidp-11/f6bec4f6-448f-4503-a3c6-663190a3b2d0%40softjar.se <https://groups.google.com/d/msgid/pidp-11/f6bec4f6-448f-4503-a3c6-663190a3b2d0%40softjar.se>>.

>

--
Johnny Billquist || "I'm on a bus
|| on a psychedelic trip

email: b...@softjar.se || Reading murder books

[Tim Boan]

Hi,

HASP - Houston Automatic SPOOLING PRIORITY, developed by IBM field office in Houston,

probably for NASA, but code was owned by IBM. For too long ago for my memory.

Tim Boan

Virus-free.www.avast.com

--

You received this message because you are subscribed to the Google Groups "[PiDP-11]" group.

To unsubscribe from this group and stop receiving emails from it, send an email to pidp-11+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/pidp-11/a3d4614c-d90b-489e-bc21-14341ca5c2ccn%40googlegroups.com.

[isy...@gmail.com]

Hi All,

 HASP. Did this become part of MTS for the 360/70/90?????

as in MTS *HSP.....

response <MTS HASP START....>

BW, Ian.