source: trunk/source/geometry/solids/test/fred/README.test2

README.test2          DCW    10/9/98

Test 2 sends many geantinos at once through a volume and displays which tracks
intersect a volume. Human judgement is used to decide if there are errors.
To make this easier, the origins of the tracks are evenly placed on a grid
that is large enough to cover the cross section of the volume. All of the
tracks are sent in the same direction. In the display, the starting points
of the tracks are drawn as blue if there is no intersection with the volume,
red if they intersect correctly, and white if they intersect incorrectly.
If the angle of the view is selected such that the tracks move exactly
in or out of the view plane, these points will form a silhouette of the 
cross section of the volume if tracking is working properly.

Since it is assumed that all tracks start from outside the test volume,
an incorrect intersecting track is one that does not have the same
number of entrances into the volume as exits.

For this test, the grid gun should be selected:

/fred/gun GRID

And the "shadow" display mode should be selected:

/fred/draw SHADOW

Before run initialization, the test volume should be selected. See
README.volumes for more information.

This test does not work for 3D views with perspective. Instead, the OpenGL
interface is a good choice for this test. It can be invoked with
the script "draw.script".

The following commands control the grid gun:

/gridGun/n1 <n1>
/gridGun/n2 <n2>
/gridGun/position <x> <y> <z> [<units>]
/gridGun/grid1 <g1x> <g1y> <g1z> [<units>]
/gridGun/grid2 <g2x> <g2y> <g2z> [<units>]
/gridGun/direction <vx> <vy> <vz>

n1 = the number of grid points along the first axis (default 10)
n2 = the number of grid points along the second axis (default 10)

The total number of tracks = (n1+1)*(n2+1)
            
x,y,z   = the origin of the grid
g1x,y,z = the direction and length of the first grid axis
g2x,y,z = the direction and length of the second grid axis

The x origin of track i,j (0 <= i <= n1, 0 <= j <= n2) is:

           x(i,j) = x + g1x*i/n1 + g2x*i/n2
           
and similar for y and z.

vx,y,z  = the direction of the tracks.

If n1=0 and n2=0, only one track is generated, starting at (x,y,z)
and in direction (vx,vy,vz). This can be convenient for debugging.

The somewhat awkward view port controls of GEANT4's visualization
can make it difficult to adjust the viewport angle to line up with
the direction of the tracks. To simplify this, and generate many
different test angles, the scripts "grids1.script", "grids2.script",
and "grids3.script" have been supplied. These scripts have been
generated using the FORTRAN program "makegridscripts.f".

An example test session:

# select GRID gun
/fred/gun GRID
# select SHADOW draw mode
/fred/draw SHADOW
# select BOX test volume
/fred/volume BOX
# initialize
/run/initialize
# initialize x-windows OpenGL viewer
/control/execute draw.script
# select a large but not too time consuming number of tracks
/gridGun/n1 40
/gridGun/n2 40
# run script 1
/control/execute grids1.script
# run script 2
/control/execute grids2.script
# run script 3
/control/execute grids3.script