29 Jan
2014
29 Jan
'14
9:47 p.m.
On 2014-01-29 14:13, Kyle Owen<kylevowen at gmail.com> wrote:
On Wed, Jan 29, 2014 at 1:51 PM, Johnny Billquist<bqt at update.uu.se> wrote:
Indeed. So you have the parity option installed in your ASR33. I punched
a bit of paper tape on an ASR33 only two weeks ago, which did not.
Unfortunately I didn't keep the paper, but that one certainly did not
have a parity option installed, so it was pretty obvious what the
"default" for parity is on an ASR33...
I'm tempted to decode it, but don't have the time right now...
>
>Eh? What? There was an option for 8-bit clean communication in the ASR33,
>as well as other parity choices. But by default an ASR33 have*MARK* parity.
>And all older PDP-8 software that I've ever seen also assumes MARK parity.
>If you try running older PDP-8 software with your terminal set to even
>parity, it will not work.
>
>If you are ever near an ASR33 with a paper punch, you can easily verify
>that it is doing mark parity by just enabling the punch, and check what
>codes you get there as you type, with the ASR33 in local mode.
>
> Johnny
>
I punched this tape on a Model 33 ASR in local mode and it seems to have
even parity. Bonus points for deciphering it!:)
http://i.imgur.com/R3GPWIS.jpg 
I'll try some more software with mark parity, but
I believe CHECKMO II (the
chess game) really wants even parity.
Please do. I would be surprised. But who knows. But I know OS/8 inside
and out (pretty much), and believe me. Older software really do use MARK
parity in general.
(And, like I said, I verified the ASR33 only two weeks ago. Besides, the
ASR33 manuals are online, and you can check it there too. I've read them
in the past, and they also say that it is mark parity.)
Johnny
29 Jan
29 Jan
10:17 p.m.
On Wed, Jan 29, 2014 at 3:47 PM, Johnny Billquist <bqt at update.uu.se> wrote:
Indeed. So you have the parity option installed in your ASR33. I punched a
bit of paper tape on an ASR33 only two weeks ago, which did not.
Unfortunately I didn't keep the paper, but that one certainly did not have
a parity option installed, so it was pretty obvious what the "default" for
parity is on an ASR33...
I'm tempted to decode it, but don't have the time right now...
I certainly don't doubt you now that the default is mark parity. My friend
owns the Teletype I printed that tape on (I'll get my own one of these
days), but it was *not* used with PDP-8s back in the day, furthering your
point. I think it was hooked up to an IMSAI 8080.
When I talk to my PDPs, I tend to use minicom at 9600 7E2 or 8N2 depending
on if I'm sending ASCII or binary data. 7E2 works fine with every program
I've tried. My notes say that trying 8N2 with CHECKMO II resulted in
garbled characters on the receiving end. Maybe the chess program was
managing to parse the incoming characters alright, but it certainly wasn't
being received by minicom without switching to 7E2.
Perhaps my issues have not been the parity but rather the data
length...more to investigate!
Kyle
10:30 p.m.
On Jan 29, 2014, at 5:17 PM, Kyle Owen <kylevowen at gmail.com> wrote:
...
When I talk to my PDPs, I tend to use minicom at 9600 7E2 or 8N2 depending
on if I'm sending ASCII or binary data. 7E2 works fine with every program
I've tried. My notes say that trying 8N2 with CHECKMO II resulted in
garbled characters on the receiving end. Maybe the chess program was
managing to parse the incoming characters alright, but it certainly wasn't
being received by minicom without switching to 7E2.
Perhaps my issues have not been the parity but rather the data
length...more to investigate!
Most likely, yes. The correct setting would be 8N1 or 7E1 for all speeds above 110. For
110, you?d want 2. If you set it for 2 stop bits, it will transmit 10% slower which is
fine, but it may also expect 2 stop bits on receive and get overrun by the incoming data
stream because it?s coming with one stop bit.
The ?1.5 stop bits? notion applies to 5 bit codes for some speeds. In fact, the actual
stop length depended on speed and machine, and might in fact be 1 bit time for some. Then
to confuse things even more, the Selectric when connected with a UART, at 134.5 bps,
required some odd stop length I think, but I don?t remember the details. (Those machine
also use any number of code tables, none of which are ASCII. RSTS supported them for a
while, V5 to V8 or so.)
paul
3986
days inactive
3986
days old
2 comments
3 participants
participants (3)
-
bqt＠update.uu.se -
kylevowen＠gmail.com -
paulkoning＠comcast.net