HP3000.BWR HP-3000 Kermit "Beware" File October 1991 HP-3000 Kermit documentation reflects version 1.0 (circa 1985), but the program has been updated several times since then. See the file HP3000.ANN (the program release announcements) for a list of new features. Anyone who would like to update the documentation, please feel free! The following comments apply to version 1.1 of the program, and may or may not be valid for the current version. ------------------------------ Date: Mon 25 Mar 85 12:08:26-PST From: Ted Shapin Subject: KERMIT parity problems To: info-kermit-request@CU20B.ARPA Postal-address: Beckman Instruments, Inc. Postal-address: 2500 Harbor X-11, Fullerton, CA 92634 Phone: (714)961-3393 Frank, I found the problem. Your suggestion to set KERMIT-20 to log debugging to a file in 8-bit mode was the key. I found out that both CP-6 and HP3000 Kermits were sending even parity so the ^A was sent as 201 octal and not recognized on the PC end. At the end is a note for others on HP3000 Kermit. Honeywell CP-6 is still being modified by its author, Lee Hallin at Los Angeles HW office (altho he is doing it on his own time and I believe will make it available to Columbia when done). Also, it's interesting to note that when I go from the PC THRU KERMIT-20 to the HP3000 or Honeywell, KERMIT-20 strips off the high order bit in terminal emulation mode so everything works. This had me puzzled for a while! Ted. = = = I had trouble communicating with HP3000 Kermit from MSKERMIT until I found the following: Set parity even on the micro. When you first connect to the HP3000, it expects input from a terminal that uses the ACK/ENQ protocol. Log on to the HP system by typing CR and control-F until you see a : which is the HP3000 prompt. Then logon to the HP specifying your terminal type, e.g. HELLO ABC.DEF;TERM=9 control-F. You will then be able to terminate your lines with just the CR. Ted Shapin ------- Reportedly, HP-3000 Kermit does not set its packet number back to zero between sessions. This means that you have to restart the program each time you want to use it. (18 Mar 87) ------------------------------ Date: Wed, 08 Aug 90 12:23:31 EDT From: Jeffrey R Kell Subject: HP-3000 Kermit usage note After noting some particularly sluggish throughput using HP3000 Kermit, particularly on the newer systems running MPE/XL, I began experimenting with various configurations and options. A few results were particularly noteworthy and constitute a "usage note" that should be included somewhere in the package (BWR?) and does not modify any code. HP3000 Kermit runs with HANDSHAKE XON, and generally requires HANDSHAKE XON on the PC-side as well to eliminate overruns during uploads at high speeds. You will not notice this (or need it) at low speeds although there may be a number of retries if the system is not ready for the next block before it is sent. Thus a local HANDSHAKE XON insures integrity, but slows down the transfer rate to a near-ridiculous level. However, as of MPE/XL 1.2, typehead is available (a poor implementation IMHO but that's another issue). Since Kermit turns off echo during transfers, the annoying typeahead echo is also turned off, and thus doesn't adversely affect transfers. Simply turn typeahead on, and no local HANDSHAKE XON is needed! The system buffers any packets sent before it is "ready" for them up to the limit of the typeahead buffer. If this buffer is filled, the system does a flow-control xoff/xon (apparently, at least I was not able to force it to lose data even at 19.2 Kbaud). Throughput increases up to 4x have been observed using typeahead rather than local HANDSHAKE XON. | Jeffrey R Kell, Dir Tech Services | UTC Postmaster/Listserv co-ord. | | Admin Computing, 117 Hunter Hall |Bitnet: JEFF@UTCVM.BITNET | | Univ of Tennessee at Chattanooga |JEFF%UTCVM.BITNET@CUNYVM.CUNY.EDU | | Chattanooga, TN 37403 | Bell: (615)-755-4551 | ------------------------------