[GRASSLIST:426] RE: Maximum number of colors in rasters

Patton, Eric epatton at nrcan.gc.ca
Mon Mar 27 11:06:05 EST 2006


Hamish, 

Thanks for the info. 

>Wild guess as to why you are asking this:
>your output looks like it has like 256 contoured bands when you think it
>should be (perceivably) continuous. try this:

>r.mapcalc 'elev_int=int(elev)'
>r.colors elev_int rast=elev   # maybe redundant
>d.rast elev_int

>[mult the DCELL map by 10000 before int() if decimal places are
>significant]

>For some reason I don't understand, floating point maps don't do smooth
>color transitions. On the other hand I have used this "feature" on
>occasion to enhance a raster map hardcopy (sometimes you want the subtle
>contour lines).

Yes, this was the issue that a colleague was having. I was at a loss to
explain why it seemed there were explicit breaks occurring in the bathymetry
data.

>I tried to document this ~1-2 years ago I think. If interested search
>through the mailing list archive or the bug tracker for the post.
>commonly seen: scale a lat/lon SRTM map to 1:1:1 XYZ for NVIZ by
>dividing elev map by 60*1852. set some colors with r.colors and then d.rast
both 
>original CELL and new DCELL maps.

Great, thanks, I'll pore through the archives for this thread; this stuff is
good to know.

Cheers,

~ Eric.




More information about the grass-user mailing list