http://groups.google.com/d/topic/comp.dcom.modems/gA5a9ugoWcM/discussion
`comp.dcom.modems ›
`Telebit registers
`9 posts by 9 authors
`Ed Wells
`5/21/89
` After calling another Telebit modem in PEP mode, have you ever gone
`back to local mode and checked out the S60's and S70's registers while
`the connection was connected? I did and found what appears to be
`several line/carrier condition tables with the grade of each of the 511
`carriers.
` I tried making a local call several times to the same number and noticed
`that each time I found small differences in these numbers (I assume the
`line sampled slightly differently each time).
` For those hacker types, you may want to try to obtain exactly what type
`of connection you have on your long distance calls. I don't know, but
`Telebit themselves may actually be interested in this information for
`modem evaluation.
` No matter what these numbers come up with, I'm very happy with my
`Telebit modem. This still appears to the the Cadillac of the 19,200
`baud modems.
`--
`=========================================================================
`Edward E. Wells Jr., President Voice: (215)-943-6061
`Wells Computer Systems Corp., Box 343, Levittown, Pa. 19058
`{dsinc,francis,hotps,lgnp1,mdi386,pebco}!wells!edw
`Click here to Reply
`Russell Lawrence
`5/24/89
`In article <49@wells.UUCP>, edw@wells.UUCP (Ed Wells) writes:
`> After calling another Telebit modem in PEP mode, have you ever gone
`> back to local mode and checked out the S60's and S70's registers while
`> the connection was connected? I did and found what appears to be
`> several line/carrier condition tables with the grade of each of the 511
`> carriers. ^^^^
`Could someone please post how-to info on interpreting the table? I've
`tried checking the contents of register 76, but what I get is a listing
`that varies from eight numbers to about a hundred... *not* 511.
`BTW, Ed mentioned in a previous article that he was unable to get his
`TB+ to work faster than 9600 on outgoing calls. My uucico supports
`19.2 kbs calls, but I've never (ever) been able to get a throughput
`of more than 700 bytes per second. One of my net neighbors thinks
`the problem might be line noise, but I'm beginning to suspect that something
`else is wrong. If anyone else has encountered a similar problem, how
`about sharing the solutions?
`--
`Russell Lawrence, WP Group, New Orleans (504) 443-5000
`uunet!wpg!russ
`George Robbins
`5/25/89
`In article <1182@wpg.UUCP> russ@wpg.UUCP (Russell Lawrence) writes:
`> In article <49@wells.UUCP>, edw@wells.UUCP (Ed Wells) writes:
`
` BTW, Ed mentioned in a previous article that he was unable to get his
`> TB+ to work faster than 9600 on outgoing calls. My uucico supports
`> 19.2 kbs calls, but I've never (ever) been able to get a throughput
`> of more than 700 bytes per second. One of my net neighbors thinks
`> the problem might be line noise, but I'm beginning to suspect that something
`> else is wrong. If anyone else has encountered a similar problem, how
`
`>>
`
`Page 1 of 4
`
`

`
`> about sharing the solutions?
`Are you sure your neighbor isn't running with an interface rate set to
`9600 baud, either via "fixing" the baud rate with the S-registers or
`letting the "incoming" baud rate default to the speed at which he last
`talked to the modem?
`This is a "problem" I've seen a couple of times, it's really not obvious
`what the TB's are up to, since they will automatically do baud rate
`adaption. The numbers are a clue though, since 700 char/sec is in the
`ball park thruput you get at a 9600 baud interface rate.
`--
`George Robbins - now working for, uucp: {uunet|pyramid|rutgers}!cbmvax!grr
`but no way officially representing arpa: cbmvax!g...@uunet.uu.net
`Commodore, Engineering Department fone: 215-431-9255 (only by moonlite)
`Henry Spencer
`5/25/89
`In article <1182@wpg.UUCP> russ@wpg.UUCP (Russell Lawrence) writes:
`>Could someone please post how-to info on interpreting the table? I've
`>tried checking the contents of register 76, but what I get is a listing
`>that varies from eight numbers to about a hundred... *not* 511.
`Sure you aren't losing some of them because of lack of flow control on
`the line between modem and host? It's really easy to drop a chunk of
`stuff when you ask a TB for something like a register listing and your
`host can't keep up with full-blast input.
`--
`Van Allen, adj: pertaining to | Henry Spencer at U of Toronto Zoology
`deadly hazards to spaceflight. | uunet!attcan!utzoo!henry he...@zoo.toronto.edu
`Bill Mayhew
`5/26/89
`The line status registers (S70 .. S78) are indeed documented as
`follows in the manual. In case of misplaced manuals, they are:
`S70 Instantaneous Transmit Rate. Outbound raw bit rate, not
` necessarily the actual throughput.
`S71 Transmit Bits Per Channel. Reports number of bits
` currently in use for all 511 carriers. Read only.
`S72 Istantaneous Receiver Rate. Inbound raw bit rate, not
` necessarily the actual throughput.
`S73 Receive Bits Per Channel. Reports number of bits
` currently in use ofr all 511 carriers. Read only.
`S74 Received Packets Retrasmitted. Number of received packets
` requiring retransmission since start of this call.
` Read only.
`S75 Packets Accepted. Number of good packets received since
` current call began. Read only.
`S76 Equivalent Line Noise Profile. CNR to the nearest 1/10th
` dBm at all 511 frequency points. Read only. Now this is
` definitely a cool register!
`S77 Frequncy Offset. Observed frequency offset of the communi-
` cations channel in Hz. to the nearest 1/16 for the current
` connection. Valid for either PEP or regualr connection.
`S78 Slow Mode Line Quality. Merit figure 0 .. 100. 50 or
` or higher menas that the line will support acceptable
` communication. Redialing is recommended if you have a
` merit of less than 30.
`Bill
`wtm@impulse.UUCP
`Jean-Pierre Radley
`6/1/89
`
`Page 2 of 4
`
`

`
`High Speed transfers (was: Telebit registers)
`In article <1182@wpg.UUCP> russ@wpg.UUCP (Russell Lawrence) writes:
`>In article <49@wells.UUCP>, edw@wells.UUCP (Ed Wells) writes:
`>BTW, Ed mentioned in a previous article that he was unable to get his
`>TB+ to work faster than 9600 on outgoing calls. My uucico supports
`>19.2 kbs calls, but I've never (ever) been able to get a throughput
`>of more than 700 bytes per second. One of my net neighbors thinks
`>the problem might be line noise, but I'm beginning to suspect that something
`>else is wrong. If anyone else has encountered a similar problem, how
`>about sharing the solutions?
`The calling computer must be running at 19200, and also the receiving
`computer.
`On the dial-out port, stty should reveal a speed of 19200. On the
`receiving computer, the gettydefs for the receiving port should also be at
`19200.
`CPU <-@19200-> TB <-@9600-> telephone lines <-@9600-> TB <-@19200-> CPU
`Each computer talks with its trailblazer at 19200.
`Each trailblazer talks with the telephone line at 9600.
`The faster throughputs, up to 1700 or 1800 bytes/sec, occur on ASCII files,
`or data files with lots of blocks of nulls or spaces. The trailblazers can
`compress such files while in PEP mode.
`Files which are already compressed, or most binary files, will not transfer
`at such a rate.
`--
`Jean-Pierre Radley CIS: 72160,1341 jpr@jpradley.UUCP
`Chris Lakewood
`6/2/89
`High Speed transfers (was: Telebit registers)
`In article <9858@dasys1.UUCP> jpr@dasys1.UUCP (Jean-Pierre Radley) writes:
`>CPU <-@19200-> TB <-@9600-> telephone lines <-@9600-> TB <-@19200-> CPU
` ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
`>Each computer talks with its trailblazer at 19200.
`>Each trailblazer talks with the telephone line at 9600.
`>The faster throughputs, up to 1700 or 1800 bytes/sec, occur on ASCII files,
`>or data files with lots of blocks of nulls or spaces. The trailblazers can
`>compress such files while in PEP mode.
`>Files which are already compressed, or most binary files, will not transfer
`>at such a rate.
`>--
`>Jean-Pierre Radley CIS: 72160,1341 jpr@jpradley.UUCP
`Wrong. The two modems communicate using PEP at varying speeds up to
`18,000 bps. The content of the data (i.e. binary vs. ASCII) has
`NOTHING to do with the speed at which the modems communicate.
`An extension to PEP called PEP2 supports data compression and can
`achieve speeds of 19,200 bps on compressible data.
`It is not uncommon to see speeds of 1400 characters/sec or more on
`data that is ALREADY compressed. This is where Telebit is way ahead
`of the other "high-speed" modem makers. Their claims of 9600, 19200, or
`38,400 are based on sending data which is highly compressible. If the
`data is already compressed, their thoughput goes WAY down.
`Please check you facts.
`hami...@osiris.cso.uiuc.edu
`6/4/89
`High Speed transfers (was: Tele
`chris@netcom.UUCP says:
`> It is not uncommon to see speeds of 1400 characters/sec or more on
`> data that is ALREADY compressed. This is where Telebit is way ahead
`> of the other "high-speed" modem makers. Their claims of 9600, 19200, or
`
`Page 3 of 4
`
`

`
`> 38,400 are based on sending data which is highly compressible. If the
`> data is already compressed, their thoughput goes WAY down.
` Please check you facts.
`i can't speak for all high-speed modems, but my old HSTs routinely got
`~1100 c/s on already compressed files, and my new HSTs (50% faster) get
`over 1600.
`Daniel A. Graifer
`6/6/89
`High Speed transfers (was: Tele
`I have a pair of Prime EXLs that were running ATT SysV 3.0 hdb uucp over
`TB+ long distance (VA to CA) and routinely achieving 1400cps. Recently,
`the VA machine was upgraded to 3.1, and uucico at 19,200 stopped working. I
`can only get garbage unless I slow down to 9600. This is true when I call
`uunet as well. Prime says they are aware of this, but cannot tell me why.
`Anybody have any ideas what they broke in this release? FYI, 3.1 was the
`release of "Network Support Utilities" and the routing of everything thru
`STREAMS.
`Thanks in advance.
`Dan
`(Note, use address below, "Reply" may not work.)
`------
`Daniel A. Graifer Franklin Capital Investments
`uunet!fciva!dag 7900 Westpark Drive, Suite A130
`(703)821-3244 McLean, VA 22102
`
`>>
`
`Page 4 of 4