NetNews Usenet Archive 1992 #16/NN_1992_16.iso/spool/comp/dcom/modems/11232

Internet Message Format | 1992-07-26 | 1.6 KB

Xref: sparky comp.dcom.modems:11232 can.uucp:177
Newsgroups: comp.dcom.modems,can.uucp
Path: sparky!uunet!zaphod.mps.ohio-state.edu!n8emr!colnet!res
From: [e-mail] (Rob Stampfli)
Subject: Re: UUCP 'g' vs. MNP & V.42
Message-ID: <[e-mail]>
Organization: Little to None
References: <[e-mail]> <[e-mail]> <[e-mail]>
Date: Sun, 26 Jul 1992 16:18:54 GMT
Lines: 21

>Anybody running MNP at any level with UUCP 'g' should have their modem
>and UUCP configurations examined!  :-)

Perhaps, but it can be made to work fine.  I originally experienced
problems with throughput on uucp-g with MNP enabled.  I saw 500-600 cps
throughput as opposed to 700 cps on a straight connection.  Experimenting
revealed that the bottleneck was the uucp window size of 3.  Increasing
this to 7 resulted in a throughput increase to ~850 cps, independent of
whether a straight, MNP or v.42 connection was made.

I now turn off reliable mode when originating a uucp connect, but allow
whatever the other end wants to establish when accepting one, and it
doesn't seem to make any difference.  Note that the numbers mentioned
above are for v.32 (not bis), and for precompressed news batches.  If
I were sending a significant amount of compressible data, I would expect
MNP5 or v.42bis to be a win, even though the uucp-g ACKS and the reliable
mode packetizing might interact with each other to increase overall delays
to some extent.
--
Rob Stampfli  [e-mail]      The neat thing about standards:
614-864-9377  HAM RADIO: [e-mail]  There are so many to choose from.