[GRASSLIST:4283] Re: more gdal help!

[Glynn Clements]

Kevin Slover wrote:

> Thanks to everyone with the previous response for creating a composite
> image.  i.composite worked the best, however is still not completely
> working for me.  I would say there was a problem with the file, but we
> are able to process it with another program, just takes hours to
> manipulate.
> 
> The problem that I am having is that there doesn't seem to be much color
> variation.  How do I explain this.  Example, the United States
> background color is a bright yellow.  It should be a softer yellow, with
> different shades.  Not happening.  I have tried a variety of color
> levels, from 2 (min) to 6/7 (for 256 colors) and all the way to 500.
> None have worked.

First, check the range of the actual data, with r.info.

Second, check the images' colour tables, either visually with
d.legend, or by examining the file in the mapset's "colr" directory. 

This should be a grey scale, where black (0 or 0:0:0) corresponds to
zero and white (255 or 255:255:255) corresponds to the maximum
possible value (e.g. 255 for 8-bit data).

Then, try displaying the data with d.rgb, or exporting it to a PPM
file (with r.out.ppm3) and viewing the PPM file.

Also, I'm curious as to your reason for generating a composite map. 
I've yet to hear of a good reason for using them (I'm not saying that
there aren't reasons, just that I don't currently know of any).

-- 
Glynn Clements <glynn.clements at virgin.net>