Original Message
> Rick, another thought....
> rspinney wrote:
>>
>> A slice through a cube would have one value defined by the user (i.e. a
>> slice at X=0). The surface could then be color coded by the value of the
>> grid point from the cube file.
>>
>
> Alternatively, one could see "x=0" as defining that surface onto which
> the desired cube file is mapped. In that case, any plane  diagonal
> or not  could be defined. All you would need is a linear expression
> in x, y, and z, and test for > or < 0. So:
>
> x the yz plane
> y the xz plane
> z the xy plane
>
> x3 the plane parallel to the yz plane, though (3,0,0)
> xy the plane perpendicular to the xy plane going through x = y.
> x+y the plane perpendicular to the xy plane going through x = y.
> x+y+z the oblique plane through (0,0,0), (1,1,0), (1,0,1)
>
> That is, the plane would be defined by a set of four numbers:
>
> {x,y,z,w}
>
> {1,0,0,0} the yz plane
> {0,1,0,0} the xz plane
> {0,0,1,0} the xy plane
>
> (1,0,0,3} the plane parallel to the yz plane, though (3,0,0)
> {1,1,0,0} the plane perpendicular to the xy plane going through x=y
> {1,1,0,0) the plane perpendicular to the xy plane going through x=y
> {1,1,1,0}the oblique plane through (0,0,0), (1,1,0), (1,0,1)
I like this, very flexible! So the command would be something like
isoslice [name] [color range] slicedefinition "cubefilename";
> Is there some standard notation for this sort of generic plane that we
> should be using?
Not that I am aware of. The only place I have seen this implemented is in
Spartan where you can generate a mobile slice of an orbital, electron
density, ESP ... It is a nice feature as the display can be in contours,
solid or translucent planes. The slice can be rotated or translated to
highlight different areas of the structure. This is likely not very easy to
do with Jmol, but it would be nice. A static slice would be fine.
> The easy construction of a simple data set along these lines, where
> each point is measured relative to its distance from this plane and
> having the same dimensions as the designated cube file would do the
> job. The JVXL format would allow for any number of predefined planes;
> one would still have to use the cube file for generating the JVXL
> data, though.
True, but hopefully we end up with a much smaller file.
> So I propose that this be part of the mapping business. We get that
> going, then implement this as a simple addition to the JVXL format.
>
>
> Bob
>
> 
Sounds great.
Rick
