[comp.dcom.modems] Trailblazer Performance

rick@pcrat.UUCP (Rick Richardson) (02/24/88)

Did anybody catch the article in Mini-Micro a couple of weeks ago
about modems and data compression?

There was a table of performance for a hayes 1200, the 'Blazer, a hayes
9600, and a null modem (wire).  They tried XMODEM, Kermit, and a protocol
I'd never heard of before (Hyperprotocol?).

As I recall, the 'Blazer beat 'em all on the Hyperprotocol, but the
chart showed abysmal performance for XMODEM and Kermit.  At the time
I thought "they must have set it up wrong".  But since all I've ever
tried is "uucp" on the 'Blazer, I don't know.  Anybody have any
real performance numbers for the 'Blazer.

BTW, the null modem was the fastest, and the cheapest in all the tests :-).

[Aside: some of you people in big companies and Universities are going to
be getting ISDN service this year.  Those that are lucky enough to
get data modules which support DMI "Mode 2" are going to be happy as
pigs in shit to know that Mode 2 is about as close to a "wire" as you're
going to get in performance, 96%-100% of a wire.]
-- 
		Rick Richardson, President, PC Research, Inc.

(201) 542-3734 (voice, nights)   OR     (201) 834-1378 (voice, days)
uunet!pcrat!rick (UUCP)			rick%pcrat.uucp@uunet.uu.net (INTERNET)

wilkes@mips.COM (John Wilkes) (02/28/88)

In article <475@pcrat.UUCP> rick@pcrat.UUCP (Rick Richardson) writes:
>As I recall, the 'Blazer beat 'em all on the Hyperprotocol, but the
>chart showed abysmal performance for XMODEM and Kermit.  At the time
>I thought "they must have set it up wrong".  But since all I've ever
>tried is "uucp" on the 'Blazer, I don't know.  Anybody have any
>real performance numbers for the 'Blazer.

I just did a few large-ish file transfers with kermit between two 'Blazers,
and kermit consistently claims an effective baud rate of about 3100.  This
is with a MIPS M/800 on one end and an AT&T Unix-PC (aka 3b1, 7300) on the
other.  I have S111=10 on my end, but I think the 'Blazer on the other end
is not set up to use this.  When the connection was made, I think it said,
"CONNECT FAST" instead of "CONNECT FAST/KERM/COMP" which might give better
results.  I will post a full report when I've got more data.

John Wilkes
-- 
-- @work:
--		  {decwrl,ames,pyramid,prls}!mips!wilkes
--		       OR, for those of great faith:
--		           wilkes@mips.com