terminal multiplexer i/o devices, simh pdp10/kl10

[Noah Smith]

hi everybody! my simh ini currently has the following

at tty -U 2030

and it works pretty good for telnetting in via 2030, but am suspecting some tty i/o issues. from some things have seen, feel like maybe should be using

set dz 8b
at -am dz 2030

but simh says 'no device dz' on startup...

question - would 'dz' be 'better'? and if so, any experience getting that going?

or maybe should configure 'tty' better?

cheers,

noah

[Noah Smith]

ahha - actually - one set of notes that definitely seem trust-worthy have the following - much more sophisticated!

set dz 8b
set dz lines=32
at -am dz 2030 speed=*32,buffer=1024k

would like to try those three, if could get 'dz' going:)

[Noah Smith]

quick update - will love to see any discussion around this topic, total newb here!
but meanwhile seem to have miraculously fixed it! - just needed to add the first line below, weren't enough 'tty-lines'!:0

set tty lines=16
at tty -U 2030

[Lars Brinkhoff]

DZ means DZ11 which is a Unibus device, hence it will only work on a KS10 machine.

The KL10 does not have any Unibus, except on the front ends.  The TTY device has support for the front end protocol.  You can set it to 8-bits and configure the number of lines:

https://github.com/rcornwell/sims/blob/master/PDP10/kl10_fe.c#L528-L548

[Noah Smith]

thanks lars! yep quick followup - realized that the simh .ini using the dz was for some other / older simh pdp10 exe - it would make sense that the exe was for a ks! meanwhile, am basically using rich's 'latest greatest' kl code and exe - it has the tty device! and with the number of lines specified to 16, everything is fine! this is for decwar, and three players were no problem, but all kinds of havoc broke out with more than three players. not enough tty lines, it was causing chaos in the decwar i/o tty handling:( lesson learned - have enough tty lines! am running 'standard stress tests' with seven and eight players now. maybe 'default' should be 16 lines? this was a painful gotcha, hahaha

cheers,

noah

--
You received this message because you are subscribed to a topic in the Google Groups "PiDP-10" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/pidp-10/k_2wizDtcdE/unsubscribe.
To unsubscribe from this group and all its topics, send an email to pidp-10+u...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/pidp-10/1c45cc40-3957-4757-a1d4-5d183169a725n%40googlegroups.com.

--

noah smith - state space dev

[Ted Wojcik]

Lars,

Not exactly.  TOPS20 supported serial lines on the front end 11 and could have multiple Unibusses for things like Bisynch.  There was space for up to 4 DTE20s.  The hardware was there.  I don’t think TOPS10 ever supported this and besides it already supported the DL10 which could support 4 Unibusses but TOP20 did.

Wow!  This was a memory flogger!  I had to go to Bitsavers to research this one.  I never supported any -20s.  My Lars field assignment was UNH and we supported both synch and asynchronous through a DL10.

/Ted

[John Francini]

Hi!

The KL-10 (serial #1288) at Brandeis University (where I worked in 1982-83 before going to DEC) had one DH-11 (16 lines) on the console front end. Those of us on the staff were really cooking with VT-100 terminals running at 9600 baud, while everyone else had to live with terminal lines off the DC-10 terminal line concentrator, which interupted the -10 on every. single. character. Hence the baud rates had to be kept to 1200 or less to prevent the interrupt overhead from swiping too much of the CPU.

This worked because DECsystem-1091 models (in the shorter DECSYSTEM-20 cabinets with blue tops) could have terminal lines on DTE0 (console front end) just like a -20, as well as on DN87S or DN20 FEs on the other DTEs. So if you installed a DH-11 in your 1090's CFE and built the Monitor accordingly, you'd have terminal lines on a 1090.

Aside: The internal MOS memory option on the 1091  (MF-20) required an additional daemon to run: TGHA (The Great Heuristic Algorithm) - which handled (perhaps badly) remapping single-bit errors in the memory. *

I don't think there was a way to have both internal and external memory on a 1091, but I could be wrong. No external memory means no SMP.

* [see https://www.computerhistory.org/collections/catalog/600000648/record/102737573/ for a collection of SPRs (remember them?) saved by Tim Litt on programs beginning with "T". ]

Thanks and regards,

John

[10,422] back in the day

--

John Francini <fran...@mac.com>

"It is a capital mistake to theorize before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts." -- Sherlock Holmes, A Scandal in Bohemia

-- 
You received this message because you are subscribed to the Google Groups "PiDP-10" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pidp-10+u...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/pidp-10/b5934726-1791-41bd-9e32-1d78dd9a5b54n%40googlegroups.com.