I do not use the LC factor, except on wooden structures to integrate certain regulatory factors or for a vibration analysis.
I always use the second option, because on large models, the relationships between the vertical load (applied on the roof for example) and the multiple cases of wind (applied on the facades) can completely change the distribution of forces in the model, and therefore the local calculation of each bars.
It is therefore never easy to know which load case is main or not.
For optimisation i prefer on this case use Galapagos, Goat or Octopus.
It takes a long time to do and program. However, the computation time of the Karamba solver is much faster than most civil engineering software that I use (without mentioning them )
Especially when you have to be flexible and have a parametrical design
i think you are approaching the LC Factors wrong. At this stage there is no possibility to do load combinations. The values that you are inputting into the model view are just simple multiplications of the view output of the model, they do not affect the actual calculation.
At this stage only load cases are available, so if u wish to really make a load combination, you have to set multiple types of loads for each load ‘case’. For example, LC0 gravity, LC1 deadload, LC2 1.35.gravity+1.35.deadload. Karamba will not combine these for you automatically.
The cross section optimisation component automatically optimises for all load cases applied.
in Karamba3D 1.3.2 the LC-factors provided at th ‘ModelView’-component influence the output of those components which take the load-case to be displayed from the selection of the ‘ModelView’-component. These are e.g. the ‘BeamView’-, ‘ShellView’- and ‘ShellLineResults’-component.
At the moment one has to apply the load-factors when defining the loads and load-cases.