Age | Commit message (Collapse) | Author |
|
|
|
|
|
|
|
Previously, we were using random(), which is really really bad.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
There are no side-effects to setting it either way, and the caller can just set saturated=NULL
or ignore the result.
|
|
|
|
|
|
This fixes a speed regression introduced by dca1938a.
|
|
|
|
|
|
Saturated pixels seem to be OK during initial peak search
|
|
|
|
|
|
|
|
|
|
|
|
Conflicts:
tests/integration_check.c
|
|
|
|
Conflicts:
src/get_hkl.c
tests/morphology_check.c
|
|
|
|
It was totally broken and wrong, and just added noise to the data, and it
should be done at the scaling stage instead of now.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|