very long xgterm window
Petr Skoda wrote on Jun 14, 2006
Hi all,
I wanted to work with a long part of merged echelle spectra (having about 5000 points) not loosing the resolution by calling
xgterm.bin -G 7000x400+10-10
It produces long graphics window (after starting the plot commands)
hopefully 7000 points long on the X11 screen. I wanted to run
splot longfile.fit to see the whole spectrum with the pixel resolution and pan it by using mouse wheel (using window manager setup).
The plot is, however, very coarse resembling histogram with box-like features or some comb-like structures in place of water lines.
But there are no details - only when zooming using "a-a" or "we-e"
on part of plot I can see the detailed line profiles. But after "wa" the boxed pattern shows again. So what is wrong ?
Why the plotting does not use every point of window available ?
I have cmbuflen set to large number (4512000, but I tried larger or smaller - no change)
Moreover the axis lables are very deformed and the figure frame seems to be deforming as well for various lengths of window - not expanding with the same rate as window itself when dragging the window edge.
Best regards,
Petr Skoda
Astronomical Institute Ondrejov
Czech Republic
I wanted to work with a long part of merged echelle spectra (having about 5000 points) not loosing the resolution by calling
xgterm.bin -G 7000x400+10-10
It produces long graphics window (after starting the plot commands)
hopefully 7000 points long on the X11 screen. I wanted to run
splot longfile.fit to see the whole spectrum with the pixel resolution and pan it by using mouse wheel (using window manager setup).
The plot is, however, very coarse resembling histogram with box-like features or some comb-like structures in place of water lines.
But there are no details - only when zooming using "a-a" or "we-e"
on part of plot I can see the detailed line profiles. But after "wa" the boxed pattern shows again. So what is wrong ?
Why the plotting does not use every point of window available ?
I have cmbuflen set to large number (4512000, but I tried larger or smaller - no change)
Moreover the axis lables are very deformed and the figure frame seems to be deforming as well for various lengths of window - not expanding with the same rate as window itself when dragging the window edge.
Best regards,
Petr Skoda
Astronomical Institute Ondrejov
Czech Republic
Mike Fitzpatrick wrote on Jun 14, 2006
Hi Petr,
The physical size of your window is a matter between your X server and your window manager, as far as IRAF is concerned the window has whatever resolution is specified for an xgterm in the dev$graphcap file (800x630). The relevant fields are :xr and :yr to set the GKI resolution and X1/X2 and Y1/Y2 to specify the physical coord transformation. I think what your seeing is the 7000 points being mapped to this lower resolution but when you zoom in there are fewer physical points to map to the screen so you see more of them.
Note some window managers won't allow you to create a window that large, but assuming you've gotten past that if you want to change this you'll need to create a private graphcap file with these changed parameters. For example,
You can put the last 3 lines in your login.cl/loginuser.cl and experiment with the values to get a better result. Note also that there are device resolutions specified for printers as well, hardcopy may look just as funny. You might also have a look at SPECPLOT to stack the spectra rather than look at them completely horizontally. Hope this helps.
Cheers,
-Mike
The physical size of your window is a matter between your X server and your window manager, as far as IRAF is concerned the window has whatever resolution is specified for an xgterm in the dev$graphcap file (800x630). The relevant fields are :xr and :yr to set the GKI resolution and X1/X2 and Y1/Y2 to specify the physical coord transformation. I think what your seeing is the 7000 points being mapped to this lower resolution but when you zoom in there are fewer physical points to map to the screen so you see more of them.
Note some window managers won't allow you to create a window that large, but assuming you've gotten past that if you want to change this you'll need to create a private graphcap file with these changed parameters. For example,
cl> copy dev$graphcap home$graphcap
cl> edit home$graphcap
.....add an entry such as
wxgterm| :xr#4096:yr#512:X2#4096:Y2#512:tc=xgterm:
cl> reset graphcap = home$graphcap
cl> reset stdgraph = wxgterm
cl> gflush
You can put the last 3 lines in your login.cl/loginuser.cl and experiment with the values to get a better result. Note also that there are device resolutions specified for printers as well, hardcopy may look just as funny. You might also have a look at SPECPLOT to stack the spectra rather than look at them completely horizontally. Hope this helps.
Cheers,
-Mike
Last post on Jun 14, 2006