diff options
| author | Thomas White <taw@physics.org> | 2021-08-13 22:03:51 +0200 |
|---|---|---|
| committer | Thomas White <taw@physics.org> | 2021-08-13 22:06:17 +0200 |
| commit | 20a9a974474d0c620604a851e680f628f2b61aa6 (patch) | |
| tree | a5009d94eace87cc367d0ff37570e79de4a076cf | |
| parent | 545dc94f7c681dcebb58d5de8521578342e2002a (diff) | |
partialator: Relax rejection criterion for profile radius
The previous cutoff of 0.2*reflection_separation seems to have been too
conservative. It cuts out over 1000 crystals from the 8000 crystals in
one of the regression tests. Examining the values shows that 0.5 is
better value. It's also a realistic physical limit - more than this,
and the reflections should be overlapping.
| -rw-r--r-- | src/partialator.c | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/src/partialator.c b/src/partialator.c index 35d1ff95..7d2cfd7b 100644 --- a/src/partialator.c +++ b/src/partialator.c @@ -1530,7 +1530,7 @@ int main(int argc, char *argv[]) if ( crystal_get_resolution_limit(image->crystals[i]) < min_res ) continue; lowest_r = lowest_reflection(crystal_get_cell(image->crystals[i])); - if ( crystal_get_profile_radius(image->crystals[i]) > 0.2*lowest_r ) { + if ( crystal_get_profile_radius(image->crystals[i]) > 0.5*lowest_r ) { ERROR("Rejecting %s %s crystal %i because " "profile radius is obviously too big (%e %e).\n", image->filename, image->ev, i, |