IMD RELEASE NOTES

------------------------------------------------------------------------
Version 4.1, released 1-December-1998
------------------------------------------------------------------------

Aside from the parameter estimation enhancements described in the
What's New In IMD 4.1 hypertext documentation, and along with a number
of bug fixes, users should note the following changes to IMD 4.1:

* Support for IDL 5.0/5.1/5.2.

* Measured data can now be saved along with IMD calculations, by
choosing this option in the File->Preferences->Auto Save... menu
option.

* New X-ray optical constants have been generated for TaN and Cu4Si.

* The GAUS_CONVOL algorithm, used to account for finite instrumental
angular and/or spectral resolution, has been changed slightly (by
Roger Dejus) as described in the gaus_convol.pro file in the 'windt'
idl directory.

------------------------------------------------------------------------
Version 4.0, released June-1998
------------------------------------------------------------------------

In addition to the major new features described in the hypertext IMD
documentation, a number of bugs have been fixed (though there are
certain to be new ones.)  In addition, users should note the
following:

* IDL V4.0x is no longer supported.  Indeed, users are strongly
encouraged to upgrade to IDL 5.1: in addition to the numerous bug
fixes on Windows and Macintosh platforms, opening IMD files under IDL
5.1 is hugely faster.

* A change has been made in the way IMD_F1F2TONK works, in accord with
the conventions adopted by Eric Gullikson at the CXRO. In previous
versions of IMD, the optical constants n,k were computed from the
atomic scattering factors f1,f2 using the exact expression relating n
and k to the square root of the dielectric function epsilon.  In IMD
V4.0, n,k are computed using the 'approximate' form of this
relationship, where the square root has been expanded in a first order
Taylor series.  This change will result in a slight difference in
optical constants only at the long-wavelength end of the X-ray
spectrum (i.e., for wavelengths longer than about 100 A.)  Note also
that new optical constants files have been generated in accord with
this change.

* New optical constants (which I recently measured) have been included
for Y, from 50-240 A.

* A change has been made in the implementation of the modified Fresnel
coefficient formalism: the choice of interface profile function is now
completely separate from the choice of which type of modification is
made (as described in Section 2.2 of the IMD documentation.)  Users
should note that when opening IMD files created with previous versions
of IMD, this change implies that the data contained in the older IMD
files may not necessarily correspond to the interface parameters
indicated.

* A known limitation exists when computing optical functions as a
function of wavelength or energy.  If you perform the computation as a
function of wavelength, contour + image plots will not be displayed
correctly if you change the wavelength/energy units from wavelength to
energy.  Similarly, if you perform the computation as a function of
energy, these plots will be incorrect if you change to wavelengths.

* Note that the IMDOPLOT procedure has been renamed to IMDSPLOT.

* Note that the chi_sqr variable has been renamed to chi_2.


