On 2015-02-13 15:36, Paul_Koning at Dell.com wrote:
On Feb 13, 2015, at 9:06 AM, Clem Cole <clemc at ccc.com
<mailto:clemc at ccc.com>> wrote:
I agree, I never saw _/correctly/_ configured systems with anything
less than 8 bits. 7 with parity (8 total) as Paul said,
was pretty standard. particularly on modems. Early UNIX tty drivers
used to assume it and pretty much always AND 0x7F with the input
character.
Not only that, but some (BSD 2.9?) put out text with 0x80 ORed into the
output stream, which messes up modern terminal emulators.
It's still there in 2.11BSD, until the last (unofficial) patch set I sent out a few years ago which removed it.
Johnny
--
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
On 2015-02-13 14:44, Paul_Koning at Dell.com wrote:
Yes, but that s 8 bits total. I was referring to 7 bits, no parity, no fixed extra bit like a mark bit.
Right. And as to the question about no devices actually ever using 7 bits, it is still a good thing to test, when you are trying to find patterns. But I don't actually know of any device that actually have been 7 bits pure. But it is one way of testing for possible strange parity issues.
We can go on discussing possible ways of troubleshooting serial ports for quite a while. And I'm happy to do so, if we really want to.
But in the current case, the problem seems to indeed have been grounding problems. So from that point of view, this isn't a problem any more.
Johnny
paul
On Feb 12, 2015, at 9:45 PM, pechter at gmail.com wrote:
I saw 7 bit mark and 7 even back in my old Field Service days in the 80s...
Bill
-----Original Message-----
From: Paul_Koning at Dell.com
To: hecnet at Update.UU.SE
Sent: Thu, 12 Feb 2015 20:59
Subject: Re: [HECnet] SC-40
On Feb 12, 2015, at 12:41 PM, Johnny Billquist <bqt at softjar.se> wrote:
...
Start with just 7 bits no parity. Try both one and two stop bits. Then move on from there.
7 bits no parity is something I ve never seen. I would start with 8 bits no parity, one stop bit unless it s 110 bps. 7 bits with parity, that does come up on a few occasions. But while I ve seen 8 and 5 and even 6 bit serial comms used, I can t think of any 7 bits in the wild.
paul
--
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
On Feb 13, 2015, at 9:06 AM, Clem Cole <clemc at ccc.com> wrote:
I agree, I never saw correctly configured systems with anything less than 8 bits. 7 with parity (8 total) as Paul said, was pretty standard. particularly on modems. Early UNIX tty drivers used to assume it and pretty much always AND 0x7F with the input character.
Not only that, but some (BSD 2.9?) put out text with 0x80 ORed into the output stream, which messes up modern terminal emulators.
Less than 8 does appear in some places. There s the old 5 bit code (Murray and friends, referred to as Baudot or RTTY by hams). There are a pile of 6 bit codes that were used by newspaper wire service data sources, carrying a variety of typesetting codes specific to the data carried. For example, stock data had lots of special codes for all the fractions then used for prices. I did some work on drivers for that, DL11s with an unusual collection of mode jumpers most likely.
paul
On Fri, Feb 13, 2015 at 8:44 AM, <Paul_Koning at dell.com> wrote:
Yes, but that s 8 bits total. I was referring to 7 bits, no parity, no fixed extra bit like a mark bit.
paul
I agree, I never saw correctly configured systems with anything less than 8 bits. 7 with parity (8 total) as Paul said, was pretty standard. particularly on modems. Early UNIX tty drivers used to assume it and pretty much always AND 0x7F with the input character.
Clem
Yes, but that s 8 bits total. I was referring to 7 bits, no parity, no fixed extra bit like a mark bit.
paul
On Feb 12, 2015, at 9:45 PM, pechter at gmail.com wrote:
I saw 7 bit mark and 7 even back in my old Field Service days in the 80s...
Bill
-----Original Message-----
From: Paul_Koning at Dell.com
To: hecnet at Update.UU.SE
Sent: Thu, 12 Feb 2015 20:59
Subject: Re: [HECnet] SC-40
On Feb 12, 2015, at 12:41 PM, Johnny Billquist <bqt at softjar.se> wrote:
...
Start with just 7 bits no parity. Try both one and two stop bits. Then move on from there.
7 bits no parity is something I ve never seen. I would start with 8 bits no parity, one stop bit unless it s 110 bps. 7 bits with parity, that does come up on a few occasions. But while I ve seen 8 and 5 and even 6 bit serial comms used, I can t think of any 7 bits in the wild.
paul
On Fri, 13 Feb 2015, Paul_Koning at Dell.com wrote:
On Feb 12, 2015, at 12:41 PM, Johnny Billquist <bqt at softjar.se> wrote:
...
Start with just 7 bits no parity. Try both one and two stop bits. Then move on from there.
7 bits no parity is something I ve never seen. I would start with 8 bits no parity, one stop bit unless it s 110 bps. 7 bits with parity, that does come up on a few occasions. But while I ve seen 8 and 5 and even 6 bit serial comms used, I can t think of any 7 bits in the wild.
Looking like it's some grounding issue somewhere. Waiting until someone with better eyes had physical access and can trace it out...due to no schematics unless Peter has some.
paul
--
Cory Smelosky
http://gewt.net Personal stuff
http://gimme-sympathy.org Projects
I saw 7 bit mark and 7 even back in my old Field Service days in the 80s...
Bill
-----Original Message-----
From: Paul_Koning at Dell.com
To: hecnet at Update.UU.SE
Sent: Thu, 12 Feb 2015 20:59
Subject: Re: [HECnet] SC-40
On Feb 12, 2015, at 12:41 PM, Johnny Billquist <bqt at softjar.se> wrote:
=20
...
Start with just 7 bits no parity. Try both one and two stop bits. Then mo=
ve on from there.
7 bits no parity is something I=E2=80=99ve never seen. I would start with =
8 bits no parity, one stop bit unless it=E2=80=99s 110 bps. 7 bits with pa=
rity, that does come up on a few occasions. But while I=E2=80=99ve seen 8 =
and 5 and even 6 bit serial comms used, I can=E2=80=99t think of any 7 bits=
in the wild.
The SC40 expects a serial byte with values 0..127 (8 bits, MSB always
=0) no parity one stop bit.
--P
I saw 7 bit mark and 7 even back in my old Field Service days in the 80s...
Bill
-----Original Message-----
From: Paul_Koning at Dell.com
To: hecnet at Update.UU.SE
Sent: Thu, 12 Feb 2015 20:59
Subject: Re: [HECnet] SC-40
On Feb 12, 2015, at 12:41 PM, Johnny Billquist <bqt at softjar.se> wrote:
...
Start with just 7 bits no parity. Try both one and two stop bits. Then move on from there.
7 bits no parity is something I ve never seen. I would start with 8 bits no parity, one stop bit unless it s 110 bps. 7 bits with parity, that does come up on a few occasions. But while I ve seen 8 and 5 and even 6 bit serial comms used, I can t think of any 7 bits in the wild.
paul
On Feb 12, 2015, at 12:41 PM, Johnny Billquist <bqt at softjar.se> wrote:
...
Start with just 7 bits no parity. Try both one and two stop bits. Then move on from there.
7 bits no parity is something I ve never seen. I would start with 8 bits no parity, one stop bit unless it s 110 bps. 7 bits with parity, that does come up on a few occasions. But while I ve seen 8 and 5 and even 6 bit serial comms used, I can t think of any 7 bits in the wild.
paul
Hello!
I remember watching a VAX system at work, suffice to say the operator
promptly did something along the lines of what you did Robert, and the
machine complained about not having any bananas (Along the lines of
that song.)
-----
Gregg C Levine gregg.drwho8 at gmail.com
"This signature fought the Time Wars, time and again."
On Thu, Feb 12, 2015 at 2:56 PM, Erik Olofsen <e.olofsen at xs4all.nl> wrote:
Eggs may be scrambled as well!
$ EXIT %x3f30b70
%KATIE-W-FISH, my hovercraft is full of eels
On Thu, Feb 12, 2015 at 11:11:44AM -0800, Robert Armstrong wrote:
$ show sys
OpenVMS V8.3 on node CODA 12-FEB-2015 11:08:30.67 Uptime 0 00:11:00
....
$ WRITE SYS$OUTPUT F$MESSAGE (2928)
%SYSTEM-W-FISH, my hovercraft is full of eels
$
Looks like Fred is right!
Bob
