jon@gaia.UUCP (Jonathan Corbet) (11/11/86)
I have had a recurring problem trying to communicate between two
machines using Hayes-compatible (actually, one IS a Hayes) 1200 baud
modems. Both modems work great; we both talk to a wide variety of machines
with our respective modems. However, when we try to talk to each other,
there appears to be a real line noise problem. In particular, the junk
characters "~r" and "}i" appear repeatedly.
Does anybody have any idea what's going on here? Why only between
our two modems? What can we do about it? Any help would be much
appreciated.
--
Jonathan Corbet
{hao | nbires}!gaia!jonzhahai@gaia.UUCP (Zhahai Stewart) (11/12/86)
Some more possibly useful information on the ~r and }i modem problem: One modem is a Hayes 1200 which has had no similar problem in connecting to many other modems over the last couple of years. The other is a 2400 bps modem, operating at 1200 of course, which has not had the same problem with anyone else. The garbage chars are inserted every 10-40 seconds more or less, thruout a session. Some connections are great, some are terrible, a few are in between; a given session is pretty much the same from start to finish. Thoughts: it *might* have to do with adaptive equalization on the 2400 bps side, with some phone connections. It might have to do with losing sync in the pseudo random scramblers used with 1200 bps. I had an almost identical problem a couple of years ago with two modems by different manufacturers from either of the ones currently involved. (ie: 4 vendors' products total). This leads me to suspect that it has something to do with a weak area of the Bell 212A protocol, something which is susceptible to marginal conditions of some sort. Any help would be appreciated. Zhahai Stewart HiSystems nbires!gaia!zhahai
steven@mulga.OZ (Steven Bird) (11/13/86)
In article <R-133@gaia.UUCP> zhahai@gaia.UUCP (Zhahai Stewart) writes: > Some more possibly useful information on the ~r and }i modem problem: > [...] Some connections are great, some are terrible, a few are in > between; a given session is pretty much the same from start to finish. I have found that the presence of the garbage characters depends on the direction of communication. ie. any garbage characters generally end up on the terminal screen this end rather than in a file or command on the computer at the other end. They occasionally interfere with other information coming down the line, locking the terminal or clearing the screen... Any help would be appreciated. [UUCP: {seismo,mcvax,ukc,ubc-vision,prlb2,enea,hplabs,tataelxsi}!munnari!steven] [ARPA: steven%munnari.oz@seismo.css.gov CSNET: steven%munnari.oz@australia]
spear@ihlpf.UUCP (Spearman) (11/13/86)
> Some more possibly useful information on the ~r and }i modem problem: > ..... The garbage chars are inserted every 10-40 > seconds more or less, thruout a session. Some connections are great, > some are terrible, a few are in between; a given session is pretty much > the same from start to finish. Sounds kind of like a line balance problem. If your lines between the 2 modems in question are particularly short whereas most of your other calls are farther away, this can come into play. Some modems have a settable dip switch to compensate for this. There also do seem to be 'marginal' conditions in 1200 baud protocol where two modems can be a little off on timing in opposite directions and can talk fine with modems that are more 'dead on', but have trouble with each other. Steve Spearman ihnp4!ihlpf!spear
roy@phri.UUCP (Roy Smith) (11/13/86)
In article <1660@mulga.OZ> steven@mulga.OZ (Steven Bird) writes: > I have found that the presence of the garbage characters depends on the > direction of communication. ie. any garbage characters generally end up > on the terminal screen this end rather than in a file or command on the > computer at the other end. I've noticed this too. I have a theory about why it happens, but I'm not sure. A 212A (and, I think, 103J and V22.bis) full duplex modem has two distinct communication channels frequency division multiplexed over one phone line. The originate->answer side uses half the bandwidth of the phone line and the answer->originate side uses the other half (probably with a guard band in-between). If the spectrum of "typical" phone noise lies mostly in one band, you'll only see noise in one direction. Any modem mavens care to comment on this? -- Roy Smith, {allegra,cmcl2,philabs}!phri!roy System Administrator, Public Health Research Institute 455 First Avenue, New York, NY 10016 "you can't spell deoxyribonucleic without unix!"
Phil_CW_Sih@cupertino.pcc (11/14/86)
Jonathan Corbet
{hao | nbires}!gaia!jon writes:
I have had a recurring problem trying to communicate between two
machines using Hayes-compatible (actually, one IS a Hayes) 1200 baud
modems. Both modems work great; we both talk to a wide variety of machines
with our respective modems. However, when we try to talk to each other,
there appears to be a real line noise problem. In particular, the junk
characters "~r" and "}i" appear repeatedly.
-----
I've noticed this problem too in using my modem to call somewhere between
100 and 200 different systems each month. I am not sure of this answer,
but I think it has to do with some internal adjustments of the modems
in question. If we assume that this adjustment has a somewhat normal
looking distribtion about a calibrated standard, then modems on either
end of the distribution could have this problem.
As an experiment (this is a real BS hack) I opened up one of my units
because I was having this problem a lot. It started to happen after
I had gotten it back from warranty repair. Inside I found one potentiometer
which was still set at its original setting. Since this was the only
adjustment in the whole thing, I decided to go online to some of the
modems I was having noise problems with and play with it. After carefully
marking its original position, I started to diddle it.
What I found, not too surprisingly, was that I could dramatically
influence the rate at which spurious characters came out of the receive
end of the modem. This appeared to be a receiver only adjustment.
I found that the acceptable range for this setting was different
for each of the "problem" modems I called and there was a small (very small)
overlap in the range acceptable to the majority of them.
I also found that which junk characters you got depended on which
side of the adjustment you were on.
I ended up setting this potentiometer in this small range of maximum
acceptance or compatibility and the problem has all but vanished.
Mind you, I don't recommend anyone go and hack their modems.
Phil Sih 408/973-9111 (voice)
The Portal System: "Phil Sih" 408/725-0561 (data)
UUCP: ucbvax!sun!portal!phil_sih%cupertino.pcc
Internet: portal!phil_sih%cupertino.pcc@sun.com
"Coming to you through the Portal-Usenet Gateway"notes@hpscda.HP.COM ( ) (11/14/86)
I've sent Hayes modems exhibiting such behavior back to Hayes for repair.
Sometimes they're heat sensitive, giving garbage after being turned on
for a while. I've forced some failures by heating the modems in my 140
degree (F) oven and then connecting & using them. Good luck.
From the artificial intelligence of {hplabs, hpda, seismo}!hpscda!garyg
Gary Gitzen
Hewlett-Packard
5301 Stevens Creek Blvd.
Santa Clara, CA 95051
408-553-2507 (W)
/* ---------- */vanam@pttesac.UUCP (Marnix van Ammers) (11/14/86)
The "}~r" problems on 212A type modems is due to time division
multiplexing on telephone company trunks. It is not the fault
of the modem. It is the fault of the digital carrier equipment.
--
Marnix A. van\ Ammers Work: (415) 545-8334
Home: (707) 644-9781 CEO: MAVANAMMERS:UNIX
{ihnp4|ptsfa}!pttesac!vanam CIS: 70027,70ciaraldi@rochester.ARPA (Mike Ciaraldi) (11/15/86)
In article <334@pttesac.UUCP> vanam@pttesac.UUCP (Marnix van Ammers) writes: >The "}~r" problems on 212A type modems is due to time division >multiplexing on telephone company trunks. It is not the fault >of the modem. It is the fault of the digital carrier equipment. >-- Great, now what can we do about it? I had/have a similar problem on by home phone. When I call the University of Rochester I get lots of line noise, but when I call other parts of town it is OK. Similarly, other neighborhoods can call the University with no problem. It seems to be something in the connection between these two particular local offices. I called Rochester Telephone, and they admitted that the trunk line between those two offices was one of the oldest still in use (30 years). Also, the problem is much worse in wet weather. How I got around it: I borrowed several different 1200 bps modems and compared them for noise rejection. Of the consumer-oriented ones, Hayes was best, Qubie was worst (kept losing the carrier), and US Robotics was in the middle but still quite bad (I hear they have issued an update which fixes some of this). I wound up buying a Microcom industrial modem, model RX/1200, which is much more solid but also more expensive. As it is, I still get some line noise, but it is not too bad. So, the question is: What can the average person do about this sort of problem? Does complaining to the local telephone company help? In my case I was finally able to track down the right person, but it took over a year (part time). They are going to be looking into it for me. Mike Ciaraldi seismo!rochester!ciaraldi
ken@rochester.ARPA (SKY) (11/15/86)
Short of moving to a different phone area, not much, I suspect. My anecdotal evidence also indicates that noise level goes up in bad weather, leading me to suspect it is in part due to connections. I find it often helps to hang up and dial again. Also, don't get call waiting for a modem line unless you can cancel it with *70 or 1170. It's the kiss of death for a session. Ken
jkg@gitpyr.gatech.EDU (Jim Greenlee) (11/15/86)
In article <22452@rochester.ARPA> ken@rochester.UUCP (SKY) writes: > >Also, don't get call waiting for a modem line unless you can cancel it >with *70 or 1170. It's the kiss of death for a session. > This has probably been posted a million times already, but I just started subscribing to this group, and there is at least one person (me) who doesn't know the answer, so please help me out, here. ISN'T THERE ANY WAY TO KEEP THE MODEM FROM GETTING CLOBBERED BY CALL WAITING!! (short of disabling it, as mentioned above) I can't afford the $100 or so that it would cost me to have a second line installed at my house, but there are times when I don't want to tie up the existing line for hours on end. Am I ill-informed, or is it infeasible for the connection to be maintained while you switch to the other line? Some modems claim to have a "Call Waiting Bridge". What the heck is that? Does it work like the manufacturers claim it ought to, or is it just some marketing guy's pipe dream? If this has been hashed over recently, then somebody please enlighten me via e-mail. My particular set-up is an AT&T 6300 with a CTS 2424ADH (about 90% Hayes-compatible) modem. Any suggestions from anyone would be appreciated. Jim Greenlee -- The Shadow...!{akgua,allegra,amd,hplabs,ihnp4,seismo,ut-ngp}!gatech!gitpyr!jkg Abj lbh'ir tbar naq qbar vg! Whfg unq gb xrrc svqqyvat jvgu vg hagvy lbh oebxr vg, qvqa'g lbh?!
ccrdave@ucdavis.UUCP (Lord Kahless) (11/17/86)
> I've sent Hayes modems exhibiting such behavior back to Hayes for repair. > Sometimes they're heat sensitive, giving garbage after being turned on > for a while. I've forced some failures by heating the modems in my 140 > degree (F) oven and then connecting & using them. Good luck. > > From the artificial intelligence of {hplabs, hpda, seismo}!hpscda!garyg I second the overheat. My Hayes always ran as hot as an oven. Now it's all burned out, after only one year, and I'm back to a five year old Prentice that takes a licking and keeps on ticking. Anyway, as I'm now in the market for a modem, what's the most reliable rated 2400 baud Hayes-type modem? {scdsvax,lll-crg,ucbvax}!ucdavis!vega!ccrdave
dan@prairie.UUCP (Daniel M. Frank) (11/17/86)
In article <18@ucdavis.UUCP> ccrdave@ucdavis.UUCP (Lord Kahless) writes: >Anyway, as I'm now in the market for a modem, what's the most reliable >rated 2400 baud Hayes-type modem? Not to start "modem wars" here, I've had very good luck with my US Robotics 2400. The only problem I ever had was with a bad solder joint on a switch, and the service folks walked me through the schematic to find it when I told them I didn't want to send it back. The USR is on 24 hours a day here at Prairie, and is used for both dial-in and dial-out. The only problems I've seen are a slight delay in locking up with Hayes 1200s which dial in (some, but not all), and perhaps a little less immunity to noise than I'd like. I've heard that the Vadic is a little better quality. I wouldn't be surprised. The USR is quite a bit cheaper, I think. -- Dan Frank uucp: ... uwvax!prairie!dan arpa: dan%caseus@spool.wisc.edu
dave@lsuc.UUCP (11/17/86)
ciaraldi@rochester.UUCP (Mike Ciaraldi) writes: >So, the question is: What can the average person do about this >sort of problem? Does complaining to the local telephone company >help? In my case I was finally able to track down the right person, >but it took over a year (part time). They are going to be looking into >it for me. We had the problem in one of our out-of-town offices, and got it fixed by getting onto a different 3-digit prefix. I don't know about the U.S., but here, there are usually multiple prefixes available within a particular area. What we asked for, in fact, was to be switched to Touch-Tone dialing, which required an upgrade from the old exchange the line was on, to a more modern one which supports tone dialing and, incidentally, has fewer problems with line noise. Dave Sherman The Law Society of Upper Canada Toronto -- { ihnp4!utzoo seismo!mnetor utai hcr decvax!utcsri } !lsuc!dave
jason@grinch.UUCP (jason) (11/18/86)
In article <133@gaia.UUCP> zhahai@gaia.UUCP (Zhahai Stewart) writes: >Some more possibly useful information on the ~r and }i modem problem: >One modem is a Hayes 1200 which has had no similar problem in connecting >to many other modems over the last couple of years. The other is a >2400 bps modem, operating at 1200 of course, which has not had the same >problem with anyone else. [....] > >Thoughts: it *might* have to do with adaptive equalization on the 2400 >bps side, with some phone connections. It might have to do with losing >sync in the pseudo random scramblers used with 1200 bps. > > Zhahai Stewart HiSystems > nbires!gaia!zhahai (WARNING: I am new to the net, and it's been awhile since I worked on this, so, please.. not TOO many flames!!) Aside from that... I might have SOME info for you.. I had been evaluating 2400 baud modems for my old company, and I have found what seems to be an oversight from Rockwell... Seems the chip set marked version 'G' (for the reciever only, the transmitter is fine) does not (at 1200 baud) support the modulation envelope that they use to try and reduce line noise and impedance matching. Rumor has it that they didn't think it was necessary.... If you think that is the problem (i.e, you have the 'G' reciever chip) then what you need to do is get the reciever chip version 'J' from your favorite modem dealer (or, probably just call Rockwell and complain) and that should clear up most of your problems. Hope this helps!! Jason Of course, if you don't have a modem with the Rockwell chips sets.... "Never mind...." ...!sun!altos86!grinch!jason Disclaimer?? Naaaa..... Who would listen to me anyway??
jkg@gitpyr.gatech.EDU (Jim Greenlee) (11/18/86)
In article <18@ucdavis.UUCP> ccrdave@ucdavis.UUCP (Lord Kahless) writes: > >Anyway, as I'm now in the market for a modem, what's the most reliable >rated 2400 baud Hayes-type modem? > > {scdsvax,lll-crg,ucbvax}!ucdavis!vega!ccrdave I tried to e-mail this, but our system has problems going through lll-crg sometimes, so I had to resort to posting. If you're interested, "PC World" magazine reviewed four 2400 baud modems in its May, 1986 issue (Hayes, US Robotics, Novation, Cermetek) and leaned heavily in favor of the Novation modem. "PC" magazine also reviewed several 2400 baud modems at about the same (sorry, can't locate the exact issue). They seemed to like the Hayes, but I have observed the same heat problems with the Hayes 2400s that you men- tioned for the 1200s. I just purchased a CTS 2424ADH (appr. $300, about 90% Hayes compatible). I couldn't afford anything more expensive, and it was favorably reviewed in the above-mentioned "PC" article. It seems to stay relatively cool, but I have only had it for about 2 months, so I can't really comment on its reliability. It comes with a 2 year warranty, though, and has suited my needs just fine. Jim Greenlee -- The Shadow...!{akgua,allegra,amd,hplabs,ihnp4,seismo,ut-ngp}!gatech!gitpyr!jkg Jryy, abj lbh'ir tbar naq qbar vg! Whfg unq gb xrrc svqqyvat jvgu vg hagvy lbh oebxr vg, qvqa'g lbh?!
jason@grinch.UUCP (jason) (11/18/86)
In article <184@grinch.grinch.UUCP> jason@grinch.UUCP writes: > >Aside from that... I might have SOME info for you.. > Sheesshh... where was i that day?? must have been a longer day at work than I thought.. Everywhere it says 'receiver' substitute 'transmitter' and everywhere it said 'transmitter' insert 'receiver'... as for 'modulation'.. well, that one is not even close... But.. the results were still the same, so, 1 out of 3 isn't TOO bad! Jason
berger@clio.UUCP (11/20/86)
Since you specifically asked about a reliable modem, let me recommend the NEC 2420/30. It has a Hayes compatibility mode as well as proprietary NEC menu-driven dialing mode. It accomodates Western Electric 2400 baud standard and V.22 BIS, so is compatible with the majority of modems in the field. It reliably switches between 2400 and 1200 baud as necessary (and accomodates 300 baud too). It meets Western Electric 212 specifications to the letter, but also has a DIP switch that gives you more baud rate error tolerance than the standard allows. A 5 year warranty is standard. The proprietary NEC software lets you autodial and auto logon (with special pause, wait for input, etc. characters) from a battery backed-up list of 12 numbers. Full call status feedback is given (ringing, busy, etc.). As a university, we get a substantial discount on NEC products and this particular modem costs about $ 375. The commercial user price will be higher. Mike Berger Center for Advanced Study University of Illinois ihnp4!uiucuxc!clio!berger
ben@catnip.UUCP (Bennett Broder) (11/20/86)
In article <2660@gitpyr.gatech.EDU> jkg@gitpyr.UUCP (Jim Greenlee) writes: >In article <18@ucdavis.UUCP> ccrdave@ucdavis.UUCP (Lord Kahless) writes: >> >>Anyway, as I'm now in the market for a modem, what's the most reliable >>rated 2400 baud Hayes-type modem? >> >> {scdsvax,lll-crg,ucbvax}!ucdavis!vega!ccrdave > >I just purchased a CTS 2424ADH (appr. $300, about 90% Hayes compatible). >I couldn't afford anything more expensive, and it was favorably reviewed >in the above-mentioned "PC" article. It seems to stay relatively cool, but >I have only had it for about 2 months, so I can't really comment on its >reliability. It comes with a 2 year warranty, though, and has suited my >needs just fine. I have a CTS on loan from a friend, and although it performs flawlessly, I can't believe the lights they chose to include. Although it has 8 leds, it doesn't have RD (receiving data), SD (sending data), or AA (auto answer mode). It does have PW (power) and MR (modem ready) which are redundant. Also, the modem has no speaker. If you can live with this, the modem seems reasonably well built, and at least as resistant to line noise as my USR Courier 1200. -- Ben Broder {ihnp4,decvax} !hjuxa!catnip!ben {houxm,topaz}/
paul@devon.UUCP (Paul Sutcliffe Jr.) (11/24/86)
In article <418@catnip.UUCP>, ben@catnip.UUCP (Bennett Broder) writes: [ refering to the CTS 2424 modem ] > It does have PW (power) and MR (modem ready) which are redundant. Not necessarily. MR might indicate the presence (on) or absence (off) of DTR. Most modems won't dial, answer, or do much of anything for that matter, without DTR being high. It could also indicate that the modems power-up self-test (if it has one) has passed. Just guessing. I don't own (or have I ever seen) a CTS modem. My point was that one should not jump too quickly to conclusions concerning the lights. -Paul -- Paul Sutcliffe, Jr. UUCP: {seismo,ihnp4,allegra,rutgers}!cbmvax!devon!paul Devon Computer Services COMPUSERVE: 76176,502 Allentown, Penna. +1 215 398 3776 "What this country needs is a good 5-cent *nickel*."
paver@milano.UUCP (11/24/86)
Quoting from the CTS2424ADA Instruction Manual: MR Modem Ready This LED follows the true state of the Data Set Ready (DSR) RS232 interface signal (pin6). It is not lighted if pin 6 is forced active with Switch 2. The MR LED indicates the Communications State has been entered and the modem, or data-set as it is sometimes called, is ready to transmit and receive data. It does not indicate that the data link is established. -- ------------------------------------------------------------- Bob Paver ----> paver@mcc.arpa OR ut-sally!im4u!milano!paver Microelectronics and Computer Technology Corp (MCC) 9430 Research Blvd Austin, Texas 78759 (512) 834-3316
hes@ecsvax.UUCP (Henry Schaffer) (11/25/86)
In article <418@catnip.UUCP>, ben@catnip.UUCP (Bennett Broder) writes: ... > >I just purchased a CTS 2424ADH (appr. $300, about 90% Hayes compatible). >... > I have a CTS on loan from a friend, and although it performs flawlessly, > I can't believe the lights they chose to include. Although it has 8 leds, > it doesn't have RD (receiving data), SD (sending data), or AA (auto answer > mode). It does have PW (power) and MR (modem ready) which are redundant. I also miss Receive/Transmit Data ... > Also, the modem has no speaker. If you can live with this, the modem seems > reasonably well built, and at least as resistant to line noise as my USR > Courier 1200. The newer models of this same model number do come with a built-in speaker, which can be disabled. We also find them more resistant to line noise at 1200 than some others we used previously. --henry schaffer n c state univ
dave@dlb.BUCK.COM (Dave Buck) (11/25/86)
In article <117@devon.UUCP> paul@devon.UUCP (Paul Sutcliffe Jr.) writes: >In article <418@catnip.UUCP>, ben@catnip.UUCP (Bennett Broder) writes: >[ refering to the CTS 2424 modem ] >> It does have PW (power) and MR (modem ready) which are redundant. > >Not necessarily. MR might indicate the presence (on) or absence (off) >of DTR. Most modems won't dial, answer, or do much of anything for >that matter, without DTR being high. Just guessing. Aw, guys! How about responding if you have some knowledge. MR is Modem Ready. Same as Data Set Ready, DSR (back in the olden days, modems were called Data Sets). TR is Terminal Ready, same as Data Terminal Ready (DTR). If you read a bit, you'll find that DSR, or MR (the modem manufacturers have difficulty with 3-letter acronyms), means the Modem is ready to communicate. This usually (usually, I said ... depends on manufacturer's alligence to the specs and on jumper/software switch settings) means the phone has been answered or a call-out has been made ... on full-duplex modems, it is usually on when the carrier is up (but not necessarily - there's another signal, Carrier Detect, which really means there's a carrier). Now, some of these new-fangled modems also indicate MR when the modem is ready to accept commands, and you can also strap it on for peculiar computers or terminals. In any case, Paul, presence of power is not the same as modem ready. And Bennett, you were on the right track, wrong led. -- Dave Buck (408)972-2825 dave@dlb.BUCK.COM, {amdahl,sun}!dlb.UUCP!dave D.L.Buck&Assoc.,Inc. 6920 Santa Teresa Blvd. San Jose, Calif.95119
jbn@glacier.ARPA (John B. Nagle) (11/26/86)
Somebody misunderstands the meaning of the RS-232 Data Set Ready
signal. Data Set Ready is supposed to indicate that a modem is attached,
powered up, not in some self-test or loopback state, and otherwise a
functional unit. It does NOT indicate that any kind of communication has
been established with the other end. The Carrier Detect signal indicates
that modem-to-modem communication has been established. Lack of DSR is
properly an indication of strictly local trouble.
I realize that microcomputer vendors tend to be pretty casual about
the RS-232 standard, but those signals have clear meanings in the modem
context.
John Nagleben@catnip.UUCP (11/27/86)
In article <117@devon.UUCP> paul@devon.UUCP (Paul Sutcliffe Jr.) writes: >In article <418@catnip.UUCP>, ben@catnip.UUCP (Bennett Broder) writes: >[ refering to the CTS 2424 modem ] >> It does have PW (power) and MR (modem ready) which are redundant. > >Not necessarily. MR might indicate the presence (on) or absence (off) >of DTR. Most modems won't dial, answer, or do much of anything for >that matter, without DTR being high. No, it has a TR light for that. >It could also indicate that the modems power-up self-test (if it has >one) has passed. Ok, you've got my curiousity aroused. I'll pull out the manual and check... No, actually, the MR light shows the status of the DSR line. The only way to get this light off (without turning off the power) is to put the modem into digital loopback mode. I'll concede that PW and MR are not entirely redundant, but I maintain that this light is not nearly as important as RD, SD, and AA, which I sorely miss. -- Ben Broder {ihnp4,decvax} !hjuxa!catnip!ben {houxm,topaz}/
bsteve@gorgo.UUCP (11/27/86)
The insertion of "}i" into the data stream stems from problems with certain older time division multiplexed T1 telephone trunks. The problem is almost always in a trunk that goes from your prefix to another in your local dialing area. Newer trunks (at least those multiplexed using newer AT&T DS1 muxes) don't exhibit the problem. The fix for the problem is to possibly get another prefix for the troubled line. This will almost always get you a different mux. However if the problem is in a trunk running from a local central office to a switching hub (where the toll switch is located, e.g. Berkeley to SF) then the only fix is to persuade the local phone company to change out the old muxes on each end... or live with the problem. So far, I'm living with it. Steve Blasingame (Oklahoma City) ihnp4!occrsh!gorgo!bsteve bsteve@eris.berkeley.edu