mrm@sceard.UUCP (M.R.Murphy) (11/11/88)
We just got a bright, shiny AT&T 6386WGS with SV3.2 for evaluation.
It has an IPC802 intellegent ports card with 2 parallel ports and 8
RS232 serial ports. The serial ports have DB25s connectors rather
than DB25p connectors. The cable that used to connect my modem
(a genuine DCE) to my other computer (a genuine DTE) doesn't fit into
the connectors on the IPC802 'cause they both have little holes rather
than one of the two of 'em having little pins and the other having
little holes. What ever happened to standards?-) If things like this
keep happening, then nothing will ever connect to anything else without
little short adapters, or a maze of twisty little cables, all the same.
This report is a genuine picking of nits, but one would expect AT&T to
follow standards, wouldn't one?
--
Mike Murphy Sceard Systems, Inc. 544 South Pacific St. San Marcos, CA 92069
UUCP: {nosc,ucsd}!sceard!mrm INTERNET: mrm%sceard.UUCP@ucsd.ucsd.eduwu@spot.Colorado.EDU (WU SHI-KUEI) (11/11/88)
In article <851@sceard.UUCP> mrm@sceard.UUCP (M.R.Murphy) writes: >We just got a bright, shiny AT&T 6386WGS with SV3.2 for evaluation. >It has an IPC802 intellegent ports card with 2 parallel ports and 8 >RS232 serial ports. The serial ports have DB25s connectors rather >than DB25p connectors. stuff deleted >This report is a genuine picking of nits, but one would expect AT&T to >follow standards, wouldn't one? For your information, the RS232 standard says NOTHING about connectors, only the leads are defined. In support of this, consider the fact that DEC VT100 and Heath H19 terminals have DB25p connectors on their back ends rather than the DB25s used by most other terminal manufacturers. In other words, don't knock the developer of the standard for not following what is not in the standard. Just a guest here: In real life Carl Brandauer uunet!nbires!bdaemon!carl
joel@arizona.edu (Joel Snyder) (11/12/88)
To the user who said that RS232 doesn't say anything about connectors: wrong. Since 1985, EIA-232-D (the revision of RS-232-C) has had a section, Interface Mechanical Characteristics, which defines the connector, the sex, the dimensions of the pins and even the drilling of the holes. The only thing it doesn't specify is the finger clearance, which is in a non-integral annex (ie, ignore if you want). RS232 is probably the single most mis-interpreted standard, and the 1985 revision clears up almost all of the myths about it. Any responsible manufacturer should be using this, but the largest probably won't (what is that weird thing on the back of my mac?). Joel Snyder, U of Az MIS Dep't, jms@Mis.arizona.edu (also, DECUS standards representative for Data Comm/Networks)
rich@jolnet.ORPK.IL.US (Rich Andrews) (11/12/88)
In article <851@sceard.UUCP> mrm@sceard.UUCP (M.R.Murphy) writes: >We just got a bright, shiny AT&T 6386WGS with SV3.2 for evaluation. >It has an IPC802 intellegent ports card with 2 parallel ports and 8 >RS232 serial ports. The serial ports have DB25s connectors rather >than DB25p connectors. >This report is a genuine picking of nits, but one would expect AT&T to >follow standards, wouldn't one? >-- >Mike Murphy Sceard Systems, Inc. 544 South Pacific St. San Marcos, CA 92069 Well I would not get too irate as the IPC-802 is not AT&T's design. rich andrews -- Any opinions expressed are my own. Now, for a limited time, they can be yours too, for the incredible price of only $19.95. Simply send $19.95 (in Alterian dollars) to ...killer!jolnet!rich or rich@jolnet.orpk.il.us.
fmcgee@cuuxb.ATT.COM (Netnews Administrator) (11/12/88)
In article <851@sceard.UUCP> mrm@sceard.UUCP (M.R.Murphy) writes: >We just got a bright, shiny AT&T 6386WGS with SV3.2 for evaluation. >It has an IPC802 intellegent ports card with 2 parallel ports and 8 >RS232 serial ports. The serial ports have DB25s connectors rather >than DB25p connectors. The cable that used to connect my modem You can blame none other than IBM for this one. Every serial port on every PC, PC clone, etc. has a female connector. Also, some clones use DB9 connectors instead of DB25's. One of the reasons for choosing the IPC 802 was that it had ports that were physically just like the port on a true IBM PC. Frank McGee Tier 3 Indirect Channel Sales Support attmail!fmcgee -- Frank McGee Tier 3 Indirect Channel Sales Support attmail!fmcgee
bill@ssbn.WLK.COM (Bill Kennedy) (11/13/88)
In article <2190@cuuxb.ATT.COM> fmcgee@cuuxb.UUCP (Frnak W. McGee) writes: >In article <851@sceard.UUCP> mrm@sceard.UUCP (M.R.Murphy) writes: [ explains new 6386 ] >>RS232 serial ports. The serial ports have DB25s connectors rather >>than DB25p connectors. The cable that used to connect my modem > >You can blame none other than IBM for this one. Every serial port on >every PC, PC clone, etc. has a female connector. Also, some >clones use DB9 connectors instead of DB25's. Just backwards Frank. IBM uses DB25p's (male) for DTE ports. There's a defnsible reason for it too. Every modem I've ever seen (DCE) has a DB25s (female) so that lets you plug in a straight through cable, male to the modem, female to the PC and you're on your way. Most terminals I have seen are DTE and have a DB25s (female) on them except the one I'm using (a Z-49, but Z-29's are the same) which has a DB25p (male) on the back for the main port and a DB25s wired as DCE on the auxiliary port. >One of the reasons for choosing the IPC 802 was that it had ports that >were physically just like the port on a true IBM PC. > >Frank McGee >Tier 3 Indirect Channel Sales Support >attmail!fmcgee Now days it's easy enough to visit your favorite Radio Shack and get a hank of ribbon cable and D connectors to taste. A quick session with the vise in the garage and you're on your way, regardless of the sex of the connector. If, on the other hand, you have to hook DCE to DCE, then you can't use ribbon cable and you'd better have a little line monitor gadget to show you what you have crossed... -- Bill Kennedy usenet {killer,att,rutgers,sun!daver,uunet!bigtex}!ssbn!bill internet bill@ssbn.WLK.COM
wtm@neoucom.UUCP (Bill Mayhew) (11/17/88)
You still left us hanging; does the DTE have a DB25S or a DB25P on it? The lastest issue I have is the August 1969 EIA RS-232-C specification, and it only recommends that the connectors have 25 leads; it doesn't specify physical form factor. It seems to me that previously most DTEs had sockets and therefore required an extension cord with plugs on each end and leads 1,2,3, 4,5,6,7,8 and 20 wired straight through. That made VT-100s and IBM XT serial ports the odd men out. Using DB25S connectors on the 6386 WGS is consistent with other DTE devices that AT&T sells. I have an Intelliport 8 port board in my IBM model 80, and that fans out to DB25S connectors too. Unfortunately, I believe we shan't ever see a standard-conforming RS-232 connector. I like the MIDI serial connectors used to wire synthesizers together: reasonable baud rate (31250), only one type of cable, and only one way to hook things up. Totally idiot proof (well, almost!). --Bill
joel@arizona.edu (Joel Snyder) (11/18/88)
Sorry... According to December, 1985, EIA-232-D, section 3.2.1: "Figure 3.1 illustrates the DTE connector which has male (pin) contacts and a female shell (plug connector). Figure 3.2 illustrates the DCE connector which has female (socket) contacts and a male shell (receptacle connector). (text on numbering and dimensions deleted)" Here are a few other myth-busters: Section 1.3: "This standards is applicable for use in data signalling rates in the range from zero to a nominal upper limit of 20,000 bits per second." (ie, there is no explicit limit, but the advice is that 19.2K is normal and you probably won't get away with signalling at 56K) There is no explicit cable length in EIA-232-D. IN PARTICULAR, the real answer is that the designer has the responsibility of building a circuit which is capable of driving all the capacitance in the driver PLUS the capacitance in the cable. It then becomes the responsibility of the designer to specify the cable length. For all intents and purposes, you have 2500 picofarads to play with. If you can build a ten thousand foot cable with less than 2000 picofarads capacitance, you can run 232 that long. jms
mrm@sceard.UUCP (M.R.Murphy) (11/18/88)
In article <1415@neoucom.UUCP> wtm@neoucom.UUCP (Bill Mayhew) writes: > >You still left us hanging; does the DTE have a DB25S or a DB25P on >it? The lastest issue I have is the August 1969 EIA RS-232-C See below. >specification, and it only recommends that the connectors have 25 >leads; it doesn't specify physical form factor. > >It seems to me that previously most DTEs had sockets and therefore >required an extension cord with plugs on each end and leads 1,2,3, >4,5,6,7,8 and 20 wired straight through. That made VT-100s and IBM >XT serial ports the odd men out. Not quite, see below. > >Using DB25S connectors on the 6386 WGS is consistent with other DTE >devices that AT&T sells. I have an Intelliport 8 port board in my >IBM model 80, and that fans out to DB25S connectors too. Not quite, see below. > >Unfortunately, I believe we shan't ever see a standard-conforming >RS-232 connector. I like the MIDI serial connectors used to >wire synthesizers together: reasonable baud rate (31250), only one >type of cable, and only one way to hook things up. Totally idiot >proof (well, almost!). > >--Bill I think that that was me that left us hanging. Sorry. The 6386WGS is a nice box. It is a DTE. With the 802 it has 9 serial ports. One is on the back of the CPU box. The RS-232 serial port on the back of the CPU box has a DB25p. The 8 serial ports on the 802 have DB25s's. The 2 parallel ports on the 802 have DB25s's. This is to make it easy to plug a serial device into a parallel port. Or maybe to make it easy to plug a parallel device into a serial port, who knows? I calculate that the correctness factor for a serial DB25p on a DTE is therefore 0.111... :-). This may be because some folks view a terminal (as in VT100 terminal) as a DTE and think that the computer to which the terminal is attached is a DCE. Not so, they are both DTE's. It's the thing that connects 'em to each other that is a DCE. That may be as simple a device as a null modem, a cable wired as a null modem (horror of horrors), a pair of modems with a pair of wires between 'em, or the entire humongous DDD network (probably a trademark of somebody) between. DTE's outta have DB25p's and DCE's oughtta have DB25s's. If anyone is really interested, the standard is EIA-232-D. (ANSI/EIA-232-D-1986, approved November 12, 1986). This standard is a revision to RS-232-C, which brings it in line with CCITT V.24, V.28 ans ISO IS2110. Paragraph 3.2.1 states Figure 3.1 illustrates the DTE connector which has male (pin) contacts and a female shell (plug connector). Figure 3.2 illustrates the DCE connector which has female (socket) contacts and a male shell (recptacle connector). One might also check out EIA-530-D. Up until this was a standard, this was just the way that folks who knew what they were doing did it. Now it's a standard. Nice if the folks who are now doing it would follow the standard (which can be ordered from ELECTRONIC INDUSTRIES ASSOCIATION Engineering Department 2001 Eye Street, N.W. Washington, D.C. 20006 and which was $20.00 (US) at time of publication. Who knows now?) Thanks are due to A. Philip Arneth, who for over 24 years worked on this standard. I give up. I'll just go to Radio Shack (registered trademark of TANDY:-) and buy a whole bunch of 9-pin, 15-pin, and 25-pin connectors and wire and build adapters as I need them. That's what we've all done in the past, and there certainly is no reason to change now. -- Mike Murphy Sceard Systems, Inc. 544 South Pacific St. San Marcos, CA 92069 mrm@sceard.UUCP {hp-sdd,nosc,ucsd}!sceard!mrm +1 619 471 0655