This may be a lengthy summary of the response I got and there is one decent workaround finally.
You can skip to the end of this email if you just want to know the solution.
The question I raised is:
" I just wonder anybody has a workaround for displaying VNMR on a Linux computer.
" I'm using RedHat Linux 9.0, Gnome and connect to a Sun-Blade/SunOS via ssh.
When I tried to run vnmr(6.1c) and display remotely, everything is ok except the
display in the graphics window is abnormal: for fid and spectrum it's black
on black except when you move the mouse over the window the screen will
flash and the normal color of the curves will appear for less than one second.
" The remote display works well on Cygwin on Windows, but the Acqi button is disabled
in that case. "
The responses I got fall into 3 groups:
1. Solutions on linux box:
John M. Wright suggested:
I have found that if you run Blackbox
(recommended) or WindowMaker instead of Gnome or KDE, the spectrum
display is usable. When you mouseover (or click on) the display
window, the normal spectrum colors will appear. The colors in the
rest of the windows turn funky, but text is still readable.
The one thing that doesn't work normally is that the middle mouse
button will not change the colormap levels in the "dconi" display.
I tried twm which comes with RedHat Linux 9 and it works similar
to what John described.
2. Use an alternative solutions:
Sara Kunz suggested XWin32 from Starnet. The one I'm using Cygwin
also works except it's slow for X display.
David Redwine from the Dow Chemical suggested VNC instead of a X server
and but he didn't state how to setup the things. At the beginning I thought
I have to install vncserver on the spectrometer computer. But that would
require root privilege on the server side. The advantage of running a vncserver
on the spectrometer computer is that you can access the vnc screen from any platforms
that support vncviewer, even in a web browser. I saw people use this way to
control spectrometers thousands of miles away, so I take it for granted
that Dave means this way.
Later on it suddenly occured to me that VNC can be used in another way around,
which I'll state at the end and which is a decent solution actually.
Maybe Dave actually mean this way around.
3. Pinpointing the reason of funky display:
As pointed out by Steve Philson of Univ. Minnesota:
The problem seems to have something to do with
the palettes that XFree or Vnmr uses. You might try starting your
Xserver with a depth of 8 instead of 16 or 24 or 32; that seemed to
work for me. Edit /etc/X11/XF86Config and set DefaultDepth to 8 and
start things up again.
Sara Kunz also raised this point:
I know I can get it to work with 8 bit graphics but that is no fun.
Yes it is no fun to use 8 bit X-window manager, so the problem seems to
do with the color depth. I know there is one way to configure overlay
of two color depths for some graphics cards that's capable of dual display.
And I think that should be the real solution.
Well, then I found modification of XF86Config is not an easy work actually
for a new video card, and the solution can vary according to the video card...
Finally.....
What I finally find out is that I can start VNCserver and VNCviewer both
on local linux computer, and ssh to remote remote spectrometer computer
on a term in the local vncserver. Surprisingly, the color depth of the vncserver
can be 8, 16 or 24 bits, and the X-server in the vnc can be gnome. Others
should also work but I didn't test.
Steps:
1. on a local RH Linux 9.0 term, run:
vncserver -depth 24
yes the color depth can be 24 here!
remember the X desktop like "mycomputer:1"
2. on the local term, run:
vncviewer mycomputer:1
3. in the vnc display, run in a term:
ssh user_at_remotespect
continue as if you were sitting before the spectrometer computer.
the display should be ok
4. shutdown vncserver at local computer terminal using:
vncserver -kill :1
Tested on RH Linux 9.0 vs SunOS 5.8 _at_ Sparc. for vnc-3.3.3r2-47.
BTW: someone suggested using vnmr macro "color" and that just doesn't work.
Cheers,
Yufeng
Yufeng Tong
Dept. Physiol. & Biophys.
Case Western Reserve Univ.
Cleveland, Ohio 44106
Tel: (216)368-8654 (O)
_______________________________________________________________________________
Life is an everlasting game of Weiqi(igo, baduk).
2004-06-09
Received on Wed Jun 23 2004 - 18:37:54 MST