Path: utzoo!attcan!utgpu!jarvis.csri.toronto.edu!mailrus!ukma!gatech!gtss!chas
From: chas@gtss.gatech.edu (Charles Cleveland)
Newsgroups: comp.sys.amiga
Subject: Re: VLT tektronix emulation
Summary: The author replies
Message-ID: <407@gtss.gatech.edu>
Date: 27 Apr 89 13:02:36 GMT
References: <sYJ9Ley00hkeQMgWtg@cs.cmu.edu>
Reply-To: chas@gtss.UUCP (Charles Cleveland)
Organization: Georgia Tech School of Physics
Lines: 68

In article <sYJ9Ley00hkeQMgWtg@cs.cmu.edu> mjw@cs.cmu.edu (Michael Witbrock) writes:
)I run VLT from home to log in to campus. It's really good; it doesn't reboot my
)machine after kermit like vt100 used to
)BUT
)I have a 19 200 baud mcn connection from home, and when I use the tektronix
)emulation at this speed, spurious lines go everywhere.
)I wonder whether just a little more speed can be got out of it (or it could send
)xon xoffs or something - it seems not to at the moment).
)
)It's a really nice tek emulation, and the stuff it does draw correctly happens
)faster than xterm running on my workstation.

I forwarded this message to Willy Langeveld (WGLP09@SLACVM.BITNET), whose
response is to Michael is below.  I am taking the liberty of posting it here,
instead of mailing it to Michael, because I thought it might be of wider
interest.

Resent-From: "Charles L Cleveland" <PH274CC@GITNVE2.GATECH.EDU>
Resent-To:
Return-Path: <WGLP09@SLACVM.BITNET>
Received:    From VM1.GATECH.EDU By GITNVE2.GATECH.EDU; Wed, 26 Apr 89 16:58:03 EDT
Received:    from SLACVM.BITNET (WGLP09) by VM1.GATECH.EDU (Mailer X1.24) with BSMTP id 9825; Wed, 26 Apr 89 16:53:57 EDT
From:        WGLP09@SLACVM.BITNET
To:          PH274CC@GITNVE2.GATECH.EDU
Subject:     Re: VLT (forwarded message)
Date:        26 Apr 89   13:52 PST


   ----- Begin forwarded message -----
Date: 26 April 1989, 13:39:39 PST
From: WGLP09   at SLACVM
To:   PH274CC at GITNVE2.GATECH.EDU
Subject: Re: VLT (forwarded message)

In-Reply-To: PH274CC AT GITNVE2.GATECH.EDU -- 04/26/89 06:07

Charles -

          If you could forward this to Michael, I would appreciate it.
          The problems at 19.2 kbaud are, as far as I can tell, not due to
any problem within VLT, but a general problem with the Amiga's serial device.
It latches only one character in the hardware and relies on the interrupt
that that generates to be serviced in time. If it does not get serviced in
time, the next character will just overwrite the latch. Interrupts are in
Chip (=slow) memory, and when there's a lot of bus contention, due for example
to display DMA, it happens not unfrequently that the service is too late. You
especially see this at the higher baud rates, of course.
           You should also see this with regular text, but with graphics it
is much easier to spot, is all... There's no known workaround, with the
exception of the obvious ones: don't run anything in the background, don't
use overscan, use less bitplanes, etc. Me, I'm happy to run at 9600.
           By the way, the raw speed of the Tektronix (tm) emulation is only
slightly faster than 9600 baud, so it's not obvious you gain much running
at 19.2k. The redraws are faster because I store the vectors in a different
binary format.
           If you have any problem with the Tek emulation at lower baud rates,
I'd appreciate it if you made a capture file (using Capture Session from the
VLT menu) of the misbehaving graphics, and send it to me in uuencoded form.
           Thanks,

                  Willy.
   -----  End  forwarded message -----

-- 
"Our vision is to speed up time, eventually eliminating it." -- Alex Schure

Charles Cleveland   Georgia Tech School of Physics   Atlanta, GA 30332-0430
UUCP: ...!gatech!gtss!chas                  INTERNET:  chas@gtss.gatech.edu