17 May
2009
17 May
'09
9:17 a.m.
Chris Elmquist <chrise at pobox.com> wrote:
I may have missed some facts earlier in this thread--
but does the system
obey hardware flow control such as CTS? ie, if CTS is low, will it be
blocked from transmitting? If that is obeyed by all of your OS and
software, then it is not hard to build a microcontroller that would sit
between the system and the terminal, adding the delays you need on <cr>
while buffering for and hardware flow controlling back toward the PDP-8.
In general, when talking about DEC equipment, the answer to that
question is *always* no, for the simple fact that DEC didn't do hardware
flow control. Hardware flow control is actually against the RS-232 spec,
and DEC didn't abuse standards (unlike most other companies).
(And to make it clear: hardware flow control is definitely not possible
towards a PDP8)
Johnny
Chris
On Thursday (05/14/2009 at 03:04AM +0200), Philipp Hachtmann wrote:
--
Johnny Billquist || "I'm on a bus
|| on a psychedelic trip
email: bqt at softjar.se || Reading murder books
pdp is alive! || tryin' to stay hip" - B. Idol
> Hi,
>
> >> I completely mis-understood the source of your troubles.
> Oh.
> The VT05 terminal operates at speeds from 110 baud to 2400 baud,
> selectable with a rotary switch in the back (my unit is currently missing
> a counter ic to generate 110baud, but...).
> So far so good. The problem is, that when you want to operate the VT05 at
> speeds higher than 300 baud, you have to wait a certain amount of time
> after a line feed character or cursor positioning command. This is
> usually done by inserting zero fill characters (4 at 2400 baud) after the
> critical commands. If you don't obey this rule, the VT05 has not
> completed changing the line when you send the next character. This is a
> bit like a too slow cr on a teletype - funny but true.
> So my wish was to modify OS/8 so far that it generates those fill characters.
> The TTY: handler is also capable of correctly doing this for me. But many
> other programs - as the OS/8 keyboard monitor itself - use direct IO
> commands to read from and write to the terminal.
> So one would have to patch many programs. In case of the keyboard
> monitor, I don't know if there is any room left to put such a code in.
> And I don't know how to reassemble and correctly install a new keyboard
> monitor from scratch.
> So all I can do is to run the VT05 at 300 baud. Or get the advances
> serial interface (see other posting).
> Did this illustrate the problem?
>
> Best wishes,
> Philipp :-)
>
>
>
> --
> http://www.hachti.de
-- Chris Elmquist 
18 May
18 May
10:26 a.m.
At 05:17 AM 5/17/2009, Johnny Billquist wrote:
Chris Elmquist <chrise at pobox.com> wrote:
And, of course, there's a pretty good chance that this VT05 is
connected via 20ma, not RS232.
(Assumption based on the fact that there's an easy fix in the KL8-JA;
given that this hasn't been used to fix the problem, it's probably not
an omnibus computer. Pre-8/E serial devices were often current loop.)
-Rick
I may have missed some facts earlier in this
thread-- but does the
system
obey hardware flow control such as CTS? ie, if CTS is low, will it be
blocked from transmitting? If that is obeyed by all of your OS and
software, then it is not hard to build a microcontroller that would sit
between the system and the terminal, adding the delays you need on <cr>
while buffering for and hardware flow controlling back toward the PDP-8.
In general, when talking about DEC equipment, the answer to that
question is *always* no, for the simple fact that DEC didn't do
hardware flow control. Hardware flow control is actually against the
RS-232 spec, and DEC didn't abuse standards (unlike most other companies). 
3:24 p.m.
New subject: 35KSR
Does anybody have info on what the weight of a Teletype 35KSR is ?, I'm
looking into the costs of shipping against going and getting it myself.
Have found that at 33 with stand is 100lb so it's obviousley more than that.
Mike.
19 May
19 May
2:38 a.m.
New subject: 35KSR
I am a very bad judge of weight, but I just went down and tried to lift each
end of mine. I would make a wild guess at maybe 210#?
Jay West
----- Original Message -----
From: "Mike Hatch" <mike at brickfieldspark.org>
To: "General Discussion: On-Topic Posts Only" <cctech at classiccmp.org>
Sent: Monday, May 18, 2009 10:24 AM
Subject: Re: 35KSR
Does anybody have info on what the weight of a
Teletype 35KSR is ?, I'm
looking into the costs of shipping against going and getting it myself.
Have found that at 33 with stand is 100lb so it's obviousley more than
that.
Mike.
2:55 a.m.
New subject: 35KSR
And FYI - did you mean ASR 35? I didn't think there was a KSR version of the
35, but I could be wrong.
Jay West
3:25 a.m.
New subject: 35KSR
And FYI - did you mean ASR 35? I didn't think
there was a KSR version of the
35, but I could be wrong.
Teletype made a huge range of weird variants of the standard products,
and countless upgrades and stunt boxes. For example, I have had light
projection Teletypes with 6 by 8 foot screens, as well as some bizarre
wall mounted Teletypes. I have never seen a KSR35, but I do would not
be surprised if they made them.
I would guess about 180-200 pounds for a 35.
--
Will
18 May
18 May
5:34 p.m.
In general, when talking about DEC equipment, the
answer to that
question is *always* no, for the simple fact that DEC didn't do hardware
flow control. Hardware flow control is actually against the RS-232 spec,
and DEC didn't abuse standards (unlike most other companies).
This reminds me of a comment I made about the HP82164 (HPIL to RS232
interface). That thing follows the RS232 standard to the letter (for
example, it correctly handles the cotnrol lines in half duplex mode).
It's just a pity that nobody else does :-). Getting it to work with some
RS232 devices is 'interesting'..
(And to make it clear: hardware flow control is
definitely not possible
towards a PDP8)
'Nothing is impossible except skiing through a revolving door' :-)
More seriously,. I would have thought you could make hardware
modifications ot the serial interface PCB, possibly to gate ome of the
ready signals with the flow control input. Or (in this case), detect the
CR (or is it LF that's the problem?) and then hold the ready line
deasseted either for enough time for the VT05 to complete its operation
or until it gets some ecternal acknowledge signal.
No, I am not suggesting this is a sensible solution to the problem,but it
is _possible_
-tony
6:46 p.m.
From: "Tony Duell" <ard at p850ug1.demon.co.uk>
Someone suggested a microcontroller to kludge up a flow control signal
(And to make
it clear: hardware flow control is definitely not possible
towards a PDP8)
'Nothing is impossible except skiing through a revolving door' :-)
More seriously,. I would have thought you could make hardware
modifications ot the serial interface PCB, possibly to gate ome of the
ready signals with the flow control input. Or (in this case), detect the
CR (or is it LF that's the problem?) and then hold the ready line
deasseted either for enough time for the VT05 to complete its operation
or until it gets some ecternal acknowledge signal.
No, I am not suggesting this is a sensible solution to the problem,but it
is _possible_ from the VT05 earlier.
Once such a signal is derived (however that is done), the 8/i, 8/L, and
PDP-12 *do* have a TTL signal input that will perform this function.
There isn't a single "serial interface PCB", however. There's one for
input, one for output, another for the baud rate generator, and another
yet for the current-loop interface. They are connected together by the
custom wire-wrap "backplane". So you'd need to convert the flow control
signal to a TTL level, then apply it to the magic wire-wrap post on
the backplane.
The nice thing is that it could be done with minimal, reversable mods to
the vintage gear.
(Somebody else mentioned Reader-Run -- that doesn't come into it, as
that handles flow control in the other direction.)
Vince
8:30 p.m.
More
seriously,. I would have thought you could make hardware
modifications ot the serial interface PCB, possibly to gate ome of the
ready signals with the flow control input. Or (in this case), detect the
CR (or is it LF that's the problem?) and then hold the ready line
deasseted either for enough time for the VT05 to complete its operation
or until it gets some ecternal acknowledge signal.
No, I am not suggesting this is a sensible solution to the problem,but it
is _possible_
Someone suggested a microcontroller to kludge up a flow control signal
from the VT05 earlier.
Once such a signal is derived (however that is done), the 8/i, 8/L, and
PDP-12 *do* have a TTL signal input that will perform this function.
There isn't a single "serial interface PCB", however. There's one for
input, one for output, another for the baud rate
generator, and another
yet for the current-loop interface. They are connected together by the
custom wire-wrap "backplane". So you'd need to convert the flow control
signal to a TTL level, then apply it to the magic wire-wrap post on
the backplane.
Yep....
My comment about 'modifying the serial interface board' could be taken as
correct in that case, though. That 'magic wirewrap pin' presumably
conencts to one on the PCBs in the serial interface circuit. So you
_could_ pick up the signal on that PCB (Personally I wouldn't, on the
grounds that unwrapping a wire from a backplane pin is easier to do than
desoldering a wire from a PCB if you ever wanted to reverse the mod).
The nice thing is that it could be done with minimal, reversable mods to
the vintage gear.
Yes.
Which PDP8 does the OP have? From an earlier comment I gathered it was an
Omnibus machine, but maybe I misread that. In that case I think you do
have to make some modification to the serial interface PCB (and it is
likely to be one PCB on those machines)
-tony
11:29 p.m.
From: "Tony Duell" <ard at p850ug1.demon.co.uk>
Which PDP8 does the OP have? From an earlier comment I
gathered it was an
Omnibus machine, but maybe I misread that. In that case I think you do
have to make some modification to the serial interface PCB (and it is
likely to be one PCB on those machines)
I had thought that Philipp had an older machine, since he didn't seem
interested in looking for a KL8J/M8655 board (which would easily solve
his problem with the VT05 on an Omnibus machine).
However, checking the archives, I see that he has an 8/e with a KL8E
board. So you have remembered correctly.
Vince
5711
days inactive
5713
days old
10 comments
8 participants
participants (8)
-
ard＠p850ug1.demon.co.uk -
bfranchuk＠jetnet.ab.ca -
bqt＠softjar.se -
jwest＠classiccmp.org -
mike＠brickfieldspark.org -
rick＠rickmurphy.net -
vrs＠msn.com -
wdonzelli＠gmail.com