10 Dec
2008
10 Dec
'08
6:23 p.m.
tiggerlasv at aim.com wrote:
I don't have a source for MMJ cables,
but I do have a US/Domestic source for MMP's
(the offset modular connectors). . .
http://www.gruber.com/GruberInd/pg.asp?FeatureID=38
They are .14 / each for flat cable, and .41 / each for round cable.
Last I checked, they don't ship internationally.
There are two different styles of DecConnect cables;
both inverted, and not inverted.?? The only difference
of course, is that one connector is flipped on one end.
Correct. And excellent info.
When used with genuine DEC adapters, one?style worked
for printers (DCE), and the other style worked with terminals (DTE).
Um. No. :-)
Both printers, terminals and computers are DTE.
DTE means data terminal (or terminating) equipment, while DCE means data
communcation equipment. A typical DCE is a modem.
DEC was one of the few companies that actually got this right, and made
cables and connectors that actually followed the standard (well, until
the switched to MMJ anyway, which was a standard of it's own).
If just every computer manufacturer had read the specs, the world would
have been much nicer, but what can one expect from companies who just
didn't put in the effort because of money and lockins (most anyone), or
just was too big to even bother (IBM).
The other company I know of who was pretty good at actually following
the RS-232 standard is HP.
Back in the early 90's, our office used uVax
2000's,
connected to VT220's. along with various printers,
connected using DEC MMJ <> DB25 adapters.
You couldn't simply swap the DecConnect cables
because of the inverted MMP's on some of the cables.
All of our cabling and adapters were unmodified DEC products.
Well, which adapter worked was obviously based on which cable you had. :-)
You still got four combinations, two of which would make you happy.
Johnny
14 Dec
14 Dec
2:20 p.m.
[...DTE...DCE...]
DEC was one of the few companies that actually got
this right, and
made cables and connectors that actually followed the standard [...]
If just every computer manufacturer had read the
specs, the world
would have been much nicer,
I'm not so sure. It was inevitable that someone would try to connect
two DTEs together directly, as indeed is easily the commonest today.
It would have been much simpler if everything had been specced the same
and all cables had been what today are loosely called null-modem.
The same mistake plagues 10baseT and 100baseTX today. You'd think
they'd've learned from the serial-line mses, but noooo, there are hosts
and hubs and you sometimes want a crossed cable and sometimes don't,
depending on what you're connecting. I really wish they'd specced
everything the same, making all cables the same (what today we call
crossed). At least now with auto-X this mess is easing some extent.
Okay, okay, </rant>....
/~\ The ASCII Mouse
\ / Ribbon Campaign
X Against HTML mouse at rodents-montreal.org
/ \ Email! 7D C8 61 52 5D E7 2D 39 4E F1 31 3E E8 B3 27 4B
15 Dec
15 Dec
2:41 p.m.
On Sunday 14 December 2008, der Mouse wrote:
The same mistake plagues 10baseT and 100baseTX today.
But, this is fixed with 1000Base-T, and works just fine with 10Base-FL,
100Base-FX, 1000Base-SX/LX/ZX, and other fiber ethernet
interconnects. :) Even connecting 10 and 100Mbit (copper) equipment to
a 1Gb switch works whether or not you have a crossover cable.
Pat
--
Purdue University Research Computing --- http://www.rcac.purdue.edu/
The Computer Refuge --- http://computer-refuge.org
10:30 p.m.
[RS232 cables]
I would agree, but then the RS232 signals would need a re-think. They are
not totally symmetircal. For example, DCD (Carrier Detect) and RI (Ring
Indicator) are outptus on the DCE, inputs on the DTE, and there's no
obvious signal that does the same thing going the other way. That's why
many null-modem cables are not simply pin-swaps (there are jumpers
between pins as well, for example crossing STR with DSR-strapped-to-DCD),
and why, as a friend of mine once found out the hard way, joining 2
null-modem cables trgether doesn't get you an exact straight-through cable.
As regards reading the specs, I've spent too many late nights trying to
convince an HP82164 to do what I want. That's the HPIL-RS232 interface,
and it appears the designer read the specs as regards the correct use of
the flow cotnrol lines, and ingorned the fact that everyone else didn't
follow the standard. The result is a device that follwos the spec and
which is a right pain to actually used.
-tony
If just every
computer manufacturer had read the specs, the world
would have been much nicer,
I'm not so sure. It was inevitable that someone would try to connect
two DTEs together directly, as indeed is easily the commonest today.
It would have been much simpler if everything had been specced the same
and all cables had been what today are loosely called null-modem.
5859
days inactive
5864
days old
3 comments
4 participants
participants (4)
-
ard＠p850ug1.demon.co.uk -
bqt＠softjar.se -
mouse＠Rodents-Montreal.ORG -
pat＠computer-refuge.org