Think there may be a number of fixes here, but from the changelog:
Found an issue finding 3rd bounding galaxy for a hole when galaxies are colinear with the
hole center as in some synthetic datasets.
Changed Period Mode strategy to add an offset to existing galaxies instead of allocating
additional memory as new cells. May want to revisit strategy, but appears to be working well,
in non-periodic mode this offset should always be 0.0. VF on SDSS DR7 finds 900 voids now using the
example script.
Think there may be a number of fixes here, but from the changelog:
Found an issue finding 3rd bounding galaxy for a hole when galaxies are colinear with the hole center as in some synthetic datasets. Changed Period Mode strategy to add an offset to existing galaxies instead of allocating additional memory as new cells. May want to revisit strategy, but appears to be working well, in non-periodic mode this offset should always be 0.0. VF on SDSS DR7 finds 900 voids now using the example script.