New Browndye release

[Gary Huber]

Dear Users,

A new release is now on the website.  I have added a new option to the born_integral program - the option to sum over atoms, just as SDA does it, rather than approximating an integral by using grid points.  The polar desolvation forces have always been rather tricky, and I'm now wondering whether doing an integral rather than a sum over atoms was a good course of action.  Namely, the integral is singular, and perhaps does not behave well near surfaces.

The integral version was behaving as advertised, although I had misprinted the equation in the documentation (now fixed).  For now, the integral version is still the default as called by bd_top, but I might change that.  The version with atoms also uses multipoles to speed up the calculations, and compares very well with a brute-force summation.

Many thanks to Jakob Niessner in the Wade group for extensively studying this issue and calling my attention to it.  

Gary