NetNews Usenet Archive 1992 #31/NN_1992_31.iso/spool/comp/dcom/modems/18726

Internet Message Format | 1992-12-31 | 3.4 KB

Xref: sparky comp.dcom.modems:18726 comp.sys.sun.hardware:6576 comp.sys.sun.admin:9842
Newsgroups: comp.dcom.modems,comp.sys.sun.hardware,comp.sys.sun.admin,cu.sun.managers
Path: sparky!uunet!zaphod.mps.ohio-state.edu!mstar!mstar!bob
From: [e-mail] (Bob Sutterfield)
Subject: Re: Anyone out there successfully get a 14.4K modem running on a Sun?
In-Reply-To: [e-mail]'s message of Thu, 31 Dec 1992 05: 45:46 GMT
Message-ID: <[e-mail]>
Sender: [e-mail]
Nntp-Posting-Host: volitans.morningstar.com
Organization: Morning Star Technologies
References: <[e-mail]> <[e-mail]>
    <[e-mail]>
Date: Thu, 31 Dec 1992 14:44:26 GMT
Lines: 49

In article <[e-mail]> [e-mail] (Squish) writes:
   I have some Telebit WorldBlazers connected to some Sun with no
   special hardware (standard tty) ...  Soon I'll be seeing kind of
   performance I get via WorldBlazer to WorldBlazer using slip,
   c-slip, and ppp (people have been telling me c-slip should win).

Using the WorldBlazers' V.32bis carrier, you'll see FTP throughput
around 2.4 Kbytes/sec for /vmunix, or 2.8 Kbytes/sec for text files (I
use 5 concatenated copies of /usr/dict/words).  Text files go faster
because they're more susceptible to in-modem V.42bis data compression.
An 8-byte ICMP ping packet shows a minimum round-trip time around
240ms, and a single-character telnet packet shows a minimum round-trip
time around 150ms.

There's no performance difference (either throughput or latency)
between SLIP and PPP, but there is a big interactive latency increase
if you use SLIP instead of CSLIP, or if your PPP doesn't do "VJ" TCP
header compression.

Using TurboPEP, you'll see FTP throughput drop some, and interactive
latency will increase markedly.

   So, if you are doing dialup, snag the latest and great Kermit and
   blaze at 38400.

The Sun's native serial ports are limited to only 38400 because that's
the highest value in <sys/ttydev.h>.  They are somewhat crippled
because they can't flow-control the modem to throttle the flow of
incoming data, but this turns out not to be much of a problem when
you're running SLIP or PPP.  If the zs silo overflows (which happens
occasionally), TCP will retransmit the packet anyway, and you won't
notice much of a performance hit.

Some add-on async serial interfaces can run at higher speeds, partly
because they use more modern IO hardware from e.g. Cirrus Logic, and
partly because they have their own device drivers that can handle
different speed settings than those specified in <sys/ttydev.h>.  For
example, I'm about to start working with our PPP/SLIP product over
some Sbus cards from Mesa Ridge Technologies ([e-mail]) that
claim to run up to 115200 with bidirectional RTS/CTS flow control and
modem signal handling too.  I'll be testing these boards with some
WorldBlazers, and with some Codex 3260FAST modems.  The 3260FAST
modems on the native serial ports gave me FTP throughput numbers of
3.7 Kbytes/sec, regardless of the data being transferred, showing that
a mere 38400 DTE is clearly a bottleneck with the new breed of modems.
--
Bob Sutterfield, Morning Star Technologies            [phone]
1760 Zollinger Rd, Columbus Ohio USA, 43221           [phone]
[e-mail]                                   [phone] (FAX)