Closed LinhWuerzburger closed 2 years ago
Fixed in this PR:
Noisemaker stays the same as it is. But the temperature source functions now differ between absolute
and relative
noise. This enables the option (relative/absolute) in the XML. Now the XML part for the random parameters for the temperature source is the same as in the initial conditions.
all values should be positive. or at least close to zero
Noisemaker produces values has the parameters
range
andstep_size
, which produce values in[-range, range]
with the givenstep size
.In the case of using an absolute value, the noisemaker has to be used in this way:
any_field += noise_maker->random_field(any_field.get_size());
in the case of using a relative value, the noisemaker has to be used in this way:but currently, it is used in this way:
any_field *= noise_maker->random_field(any_field.get_size());
Therefore I would expect negative values as well, resulting in a huge difference between the original and the distorted value, which is definitely not how it should be.