The current output related option(s):

+-h  height                   integer  value only
+-w  width                    integer  value only
+-mi max_image_buffer_memory  integer  value only

become(s):

+-ih  image_height            integer  value only
+-iw  image_width             integer  value only
+-ibm image_buffer_max        integer  value only

Parsing code mostly there, nothing else.

Some ideas floating around about about fewer / different internal 
storage formats which would likely drive new options in this 
family/grouping.

One idea - mentioned a few places in the code by prior coders too - is 
having r,g,b plus rt gt bt channels internally rather than only one 
universal transmit (alpha channel) value as a result. It would likely 
make easier limited compositing of individual features in a way which 
would not fly well beyond the zero to one (represent-able) range. The 
phong, specular features - for example - would use up each of the 
individual transmit channels as they maxed out. (Yuqk has that 
experimental 'intensity_max' finish feature which comes at the problem 
in other ways).

An internal format which aligns more directly with our perceptual range 
for intensities and/or display/print gamuts is attractive (on the 
surface at least) too - especially when looking at some of the output 
curves we might support with the avif image format for example.

Bill P.

Post a reply to this message