Hi, everybody,
I had some difficulty to understand what the term "mask" means with
the ToneEQ module. Only after watching Aurelien's tutorial
(https://www.youtube.com/watch?v=kzACn3l49HM), in particular the part
starting at 20:13, I grasped the idea (which is actually not hard to
understand) but which does for me not match what a mask is.
For me, a mask is always something masking an effect. For simple pixel
ops like tone effects, for each pixel the output O is O=M*F(I)+(1-M)*I
(where '*' denotes an operation which ensures M*I+(1-M)*I=I). I is
the input value, F the function (e.g. a tone curve) and M is a mask
value.
However, the so-called mask in ToneEQ is part of the function F :
O=Msk*F(Map,I)+(1-Msk)*I where Msk is a usual mask and Map is what's
currently called mask in ToneEQ. Hence, for me, and I think others as
well, the usage of the term mask in the ToneEQ module is confusing. It
is rather a parameter map than a mask as it determines what's
happening to a value but not to which extent (which leads to the
effect that, in addition to the ToneEQ mask, one can apply real masks
to restrict the effect to parts of the image). Now, the term has been
introduced in the manual, but still I'd find it disturbing. Maybe one
could change this over time so to keep the meaning of mask clear, and
call it a parameter map?
Uwe
still a DT beginner but slowly getting to it.
____________________________________________________________________________
darktable user mailing list
to unsubscribe send a mail to [email protected]
