Path: utzoo!mnetor!uunet!husc6!rutgers!ucla-cs!zen!ucbvax!cbosgd!mandrill!neoucom!wtm
From: wtm@neoucom.UUCP (Bill Mayhew)
Newsgroups: comp.sys.att
Subject: Pathetic 3b1 terminal emulation speed
Message-ID: <890@neoucom.UUCP>
Date: 28 Dec 87 14:44:19 GMT
Organization: Northeastern Ohio Universities College of Medicine
Lines: 36
Keywords: vi ^L and go get a cup of coffee while u wait

Hi,

I'd like to get my HP2645 off the desk and have room for my 3b1 and
couple of manuals and some room to work.  I won't get rid of my
2645 until I can get the 3b1 to act as a reasonable dial-up
terminal.  I have to work at my account on a vax quite a bit, and
I'd just like to log in instead of uucp'ing stuff, editing, and
then uucp'ing back to the vax.

Just by looking, it looks like the effective scroll rate of the 3b1
is about 4800 baud when working on something directly on the 3b1.
Like native vi, for instance.  That's OK by me, since my brain is
slower than 4800 baud anyway.

Now the gripe:  the ATE software is incredibly slow.  It can only
keep up with 300 baud!!  I asked not to log my session to a file,
but every character received still goes into a file in 
/u/name/Filecabinet/somelogfilename.  I've read the dang ATE manual
a couple of times and I can't see anything I'm doing worng.  Is
this a known problem, or am I being a dummy as usual?

I gave up on the ATE and decided to use cu instead.  cu still only
manages to make about 1000 baud throughput.  I know cu does do some
internal buffering and checking for tilde sequences, but that
doesn't explain the lethargy.

I know quasi-reasonable serial throughts are possible, as my uucp
calls to the vax average around 550 characters/sec via a 9600 baud
Trialblazer modem.  While this is less than I'd hope for, I think
the load on the vax is bogging down the uucp xfer.  I think the 3b1
is keeping up on its end.

Thanks for letting me bend your ear.

Happy holidays,
Bill