All,
Recently, while I was modelling the creation of some kind of wind power substructure, I decided to do it as always: first customising my flowitem and, later on, creating the 3D object through a"Create Object" activity in m Process Flow. All of a sudden, an issue came up here: the 3D model of the flowitem was extremely heavy, so every time there was a token in the aforementioned activity about to create another item, the model got stuck for a split second.
Afterwards, in order to solve it, I decided to first create a Box flowitem and change its shape right after with a "Change Visual" block, which surprisingly didn't take any noticeable extra-time for the simulation already running. At that point, I was just glad at having it solved so I no longer wonder about it.
However, I have just come across the fact that, if you recycle any flowitem in a 3D sink, the process flow "Create Object" activity does not reuse it to optimise the running. Put differently, I guess the "Create Object" PF activity works in a different way compare to 3D Source when creating.
Thus, some questions come to my mind regarding this issue:
- What was the cause of the drag on the model performance in my case? In other words, what's exactly the difference between creating a non-optimised customised flowitem or a default softer one (i.e. box) and changing its shape afterwards? I suspect has something to do with splitting the operation (first create, then set 3D model) but, if the 3D model is equally heavy, same amount of memory is allocated, isn't it?
- Why PF "Create Object" activity does not take recycled flowitems as source does?
- And, a little bit beyond this, is there any difference between recycle flowitems once the process is finished or keeping them hidden in the model in terms of performance hit? Just in the case they are being recycled but not reused.
Thanks in advance,
Javi
