LCDproc development and user support list

Text archives Help


[Lcdproc] Strangeness with 1.5.2 and Matrix Orbital VFD


Chronological Thread 
  • From: ethan.dicks AT gmail.com (Ethan Dicks)
  • Subject: [Lcdproc] Strangeness with 1.5.2 and Matrix Orbital VFD
  • Date: Thu May 3 02:28:01 2007

Is anyone else here using a Matrix Orbital serial-interfaced LCD or
VFD? I've seen a couple of strange things, one I fixed, the other I'm
still trying to figure out.

The one I fixed is the sense of DEFAULT_AUTOSCROLL. It's 1 in
MtxOrb.h and needs to be 0. If it's 1, the display scrolls everytime
position (19,3) is written to.

The other is less obvious. I'm seeing occasional garbage characters,
and I think I have a theory, but not the solution, yet. At the ends
of some blocks of text, a single garbage char appears where one would
routinely (but not exclusively, I think) expect a blank. If I fire up
LCDd on a freshly-powered screen, the garbage char is the fill char
(for a recently-used display, it's either the heartbeat icon or one of
the big num elements). It seems to me that the garbage char might
just be whatever 0x00 renders to.

Positionally, when I start up LCDd, I see the garbage char at (0,0),
and right after the text on the third line. Starting clients may
cover up those chars, or add new ones. If I run my hbar test (which
just runs an hbar from 0-100 on line 2, and 100-0 on line 4), whenever
the downward scrolling hbar goes from a sliver to what should be a
blank, I see the garbage char. They accumulate as the bar shrinks,
and if you kill the client and return to the server screen, the
garbage remains on the 4th line (most likely because the server screen
doesn't write there and the backing store doesn't know there are non
blank chars there).

At this point, I'm suspecting a fence-post error in the code that
determines how much of a string to send to the display after assessing
the state of the backing store. If anyone else is using a Matrix
Orbital display and sees odd chars with LCD 1.5.2, please speak up.
The more examples of this we have, the faster we should be able to put
a finger on the solution.

For the record, 1.5.2 looks flawless on my 40x4 lcd2usb display, so
I'm fairly sure the problem is localized to the Matrix Orbital driver.

-ethan



  • [Lcdproc] Strangeness with 1.5.2 and Matrix Orbital VFD, Ethan Dicks, 05/03/2007

Archive powered by MHonArc 2.6.18.

Top of page