Next Previous Up Contents
Next: Comparison with TOPCAT plotting
Up: Plotting
Previous: auto
Several of the plot output modes
write the plot in some graphics format or other.
When selecting an output format it is important to understand the
distinction between bitmapped and vector formats;
basically bitmapped formats represent the image as a grid of finite-sized
pixels while vector formats notionally draw smooth lines.
Bitmapped formats are fine for a computer screen, but for high quality
paper printouts you will want a vector format.
You can convert from vector to bitmapped but not (usefully) in the
other direction. There are a couple of subtleties to this distinction
specific to STILTS graphical output as discussed below.
The following formats are the available values for the
ofmt
parameter of the various plot commands:
-
png
- PNG format.
This is a flexible bitmapped format providing transparency
and an unlimited number of colours with good compression.
It is fairly widely supported by browsers and other image viewers,
but perhaps not as widely as GIF.
-
gif
- GIF format.
This is a very widely-supported bitmapped format providing transparency.
The number of colours is limited to 255 however, so if you are using
auxiliary axes (colour variation to represent higher dimensionality)
or other plot features which use a wide range of colours you may see
image degradation.
-
jpeg
- JPEG format.
This is a bitmapped format intended primarily for photographs.
Transparency is not supported, and although there is no limit on the
maximum number of colours, its lossiness means that plots generated
using it generally look a bit smudged.
-
eps
- Encapsulated Postscript.
This is a vector format which is suitable for printing at high resolution
either standalone or imported into some other presentation format
(you may need to convert it via PDF depending on the intended destination).
However, there are a couple of caveats when it comes to using it with
STILTS plots.
- Unfortunately the postscript driver used by STILTS is not very
efficient and can result in large, sometimes very large, postscript
output files. This is likely to be a problem for plots with a large
number of non-transparent points. See
eps-gzip
.
- Postscript has no support for partial transparency, so if plots
are drawn with partially transparent points (common for very large
data sets) the only way they can be rendered is by drawing the body
of the plot as a bitmap rather than as vector graphics.
This is probably a blessing in disguise since with very large numbers
of points a vector postscript file would likely be unmanageably
large in any case.
So if there is any transparency in the plot, the interior of the
plot will be pixellated. The axes and annotations outside of the
plot will still be drawn in vector format however.
-
eps-gzip
- Just like the
eps
format above except that the output
is automatically compressed using the GZIP format as it is written.
Postscript compresses well (typically a factor of 5-10).
Next Previous Up Contents
Next: Comparison with TOPCAT plotting
Up: Plotting
Previous: auto
STILTS - Starlink Tables Infrastructure Library Tool Set
Starlink User Note256
STILTS web page:
http://www.starlink.ac.uk/stilts/
Author email:
m.b.taylor@bristol.ac.uk