[ih] byte order, was Octal vs Hex, not Re: Dotted decimal notation

[Jack Haverty]

The MIT-DM TTY controller for our KA PDP-10 was a homebrew device, built
by Steve Morton.   Steve really really wanted to design and build a disk
interface.  But we already had a disk interface, and just needed more
TTY lines.  So he was told to make a TTY interface.  He did, but
designed it as if it was to be a disk interface.   I don't recall
exactly how it interfaced to the CPU, but it was fast and efficient.

I added a bit of logic on the RS232 side to boost line speeds up near
100kb/s over RG174 cable spanning between the 2nd (Imlacs) and 9th
(PDP-10) floors of the building.   We ran about 8-10 Imlacs on those TTY
lines, which became a favorite for MazeWars.   It didn't consume very
much of the CPU, but it occupied all of the terminals (Imlacs).

I've wondered if that "TTY Net" was one of the earliest "LAN"
implementation.  We certainly used it like a LAN.  Metcalfe hadn't
gotten around to inventing Ethernet yet, he was still involved with the
IMP interface.

/Jack

On 12/31/20 12:34 PM, the keyboard of geoff goodfellow wrote:
> jack, sure thought so that that so called "legend" is Total Fantasy...!
>
> btw, serial lines connected to the PDP-10's Line Scanner caused an
> Interrupt Per Character... the fact that Mazwar (most especially with
> your "bandwidth enhancement") became consumer of CPU jives with yours
> truly's remembrance of our KA-10 (SRI-AI) when yours truly requested
> our display terminal speeds get upped to 9600 baud (from 2400) and was
> told that wasn't gonna happen cuz 4 9600 baud terminals going flat out
> would consume all the the CPU (and leave nothing for the users
> programs to run)!
>
> we did have ONE "terminal" that went at 9600 baud:
> a https://en.wikipedia.org/wiki/DEC_GT40 vector graphic terminal and
> when we downloaded programs to it (it was located in our machine room
> not to far from the KA-10's console) you could see the light on the
> console corresponding to its Job # be on SOLID -- for a program that
> was literally just spewing/typing out the contents of the executable
> being swallowed by the GT40.
>
> now speaking of something that DARPA DID summarily ban: the NCP port
> 21 "Short Text Message" (dirty) Limerick Server... :D
>
> geoff
>
> On Thu, Dec 31, 2020 at 9:42 AM Jack Haverty via Internet-history
> <internet-history at elists.isoc.org
> <mailto:internet-history at elists.isoc.org>> wrote:
>
>     I just asked this question on a forum of ex-BBN employees, which is
>     populated by many of the people who were involved with building and
>     operating the ARPANET from its beginning and through the 70s and
>     80s.  
>     That elicited answers from the two people who were in charge of the
>     ARPANET project through that time, with ARPA as their client/boss, as
>     well as engineers who worked on building and operating it.
>
>     The consensus -- no such thing as ARPA banning MazeWars over the
>     ARPANET
>     actually happened:
>
>     "I would have heard about it if it were true.  I was deeply connected
>     with ARPA at the time"
>
>     So I'd consider that pretty good evidence that the "legend" is
>     fantasy.
>
>     MazeWars was (unsuccessfully) banned at MIT-DM as it became a prime
>     consumer of CPU and Console time, but that mostly just shifted gaming
>     into the wee hours of the day.   No ARPANET involved.
>
>     /Jack Haverty
>     (MIT-DM 1970-1977; BBN 1977-1990)
>
>     On 12/31/20 4:10 AM, Lars Brinkhoff via Internet-history wrote:
>     > Geoff Goodfellow wrote:
>     >> the MIT PDP-10 reference must be of Al Vezza's MIT-DM host, but
>     yours truly
>     >> is kinda perplexed over the last sentence of:
>     >>
>     >> "Mazewar games between MIT and Stanford were a major data load
>     on the
>     >> early Arpanet."
>     >>
>     >> wondering just what host at Stanford this must have been -- if
>     not SU-AI --
>     >> which yours truly recalls had a couple of Imlac's -- one of
>     which was at
>     >> JMC's (John McCarthy's) house and other at RWW's (Richard
>     Weyhrauch's)
>     >> house -- both of which were connected with 1200 baud leased
>     lines... hardly
>     >> big enough to "contribute" to "a major data load on the early
>     Arpanet." --
>     >> most especially given that JMC &/ RWW didn't seem to be the
>     mazewar playing
>     >> kinda folks...
>     >>
>     >> anyone got more "history" here on this...¿¿¿
>     > I have seen this story many times, but no evidence to back it up.
>     >
>     > It seems DEC WRL's MazeWar for X10/X11/Sunview is one source for the
>     > claim.  The manpage says "MazeWar first appeared at MIT in the early
>     > 1970s, using IMLAC displays and the ArpaNet network.  Legend has it
>     > that, at one point during that period, MazeWar was banned by
>     DARPA from
>     > the ArpaNet because half of all the packets in a given month were
>     > MazeWar packets flying between Stanford and MIT."
>
>     -- 
>     Internet-history mailing list
>     Internet-history at elists.isoc.org
>     <mailto:Internet-history at elists.isoc.org>
>     https://elists.isoc.org/mailman/listinfo/internet-history
>
>
>
> -- 
> Geoff.Goodfellow at iconia.com <mailto:Geoff.Goodfellow at iconia.com>
> living as The Truth is True
>
>
>