On this page, you will find examples of some plots and drawings that were created
using the SLxfig module. Click on a figure below to see the code
that generated it.
This is a simple plot of a damped exponential. |  |
This plot illustrates the use of color.
Note the ``small-caps'' font used in the caption. |  |
A shaded histogram. |  |
Here are two stacked plots sharing a common X axis. |  |
This plot contains both X and Y errorbars |  |
This image plot uses S-Lang's drywet colormap. |  |
This plot shows the ratios of the speed of the
python Numeric and numarray extensions to speed of S-Lang as a
function of array size. As the plot shows, S-Lang is about 10 times
faster than numarray is for small arrays, and almost twice as fast
for large arrays. It also shows the well-known fact that numarray
is worse than Numeric for small arrays. The dramatic decrease in
speed of numarray for arrays larger than 10000 elements may be due
to cache misses. |  |
This so-called order-sorting plot shows an image
of a binned Chandra level 1.5 event list. The X axis corresponds
to the Chandra MEG grating wavelength and the Y axis corresponds to
an effective grating order given as the product of the grating
wavelength and the ACIS CCD energy. Such a plot is extremely
useful for diagnosing problems with the order-sorting process as
well as the gain calibration of the CCDs. |  |
I created this figure to illustrate the features
that one might see in an order-sorting plot. |  |
Plots of this type were created for a paper on the
Con-X off-plane gratings. This example provides a simple
illustration of the vbox/hbox statcking functions. |  |
This figure was created to graphically illustrate
the meaning of some of the maplib module's projections. Did I
mention that SLxfig can do 3-d projections? |  |
The bottom plot shows, among other things, the
effect of a proposed sub-pixel algorithm for ACIS data. |  |