[GRASS-dev] [GRASS GIS] #1427: strange, low res display output behavior

GRASS GIS trac at osgeo.org
Mon Aug 22 14:01:13 EDT 2011


#1427: strange, low res display output behavior
-------------------------+--------------------------------------------------
 Reporter:  cmbarton     |       Owner:  grass-dev@…              
     Type:  defect       |      Status:  new                      
 Priority:  normal       |   Milestone:  7.0.0                    
Component:  Display      |     Version:  svn-trunk                
 Keywords:               |    Platform:  Unspecified              
      Cpu:  Unspecified  |  
-------------------------+--------------------------------------------------
 By chance I discovered a weird behavior of the display output. Here is how
 to reproduce it.

 1. Open a raster map to full extent. If you save the display to a graphics
 file, it will look like the display. See attached file:
 DEM_fullextents_saved.pnb

 2. Zoom the map to a small area. A saved display will again look like the
 screen. See attached files: DEM_zoomedarea_screenshot.png and
 DEM_zoomedin_saved.png.

 3. Set the computational region to match the display and zoom back to the
 full map extent. The on-screen display will look the same as in (1). But
 if you save this to a graphics file, the saved image will have the extents
 of (1), but the resolution of the number of x and y map cells in (2). See
 attached file: DEM_zoomedout_saved.png.

 The constrain resolution to computational settings option is off. The
 display resolution is supposed to stay constant (i.e., number of x and y
 pixels in a display window) regardless of the zoom level and computational
 region settings if the constrain resolution option is off. For the
 display, this seems to indeed be the case.

 But saving the output to a graphic file is somehow getting this mixed up.

 I've only seen this in the most recent svn trunk, but have not yet tested
 in other branches.

-- 
Ticket URL: <http://trac.osgeo.org/grass/ticket/1427>
GRASS GIS <http://grass.osgeo.org>



More information about the grass-dev mailing list