Finally figure out how 'Move with Host' works - need to take into account previous comment somewhere that 'Move with Host' needs 'Draft Move'
Now the difference in concept is more apparent:-
Expected 'SuperImposed' Placement
The Additions / Subtraction's Placements are 'superimposed' over the parent ArchWall's Placement:
Move with Host - Command-wise (need Draft Move)
- It seems with other Part Object, after fusion etc, the resultant object is a 'coherent' composition / compound.
- Similar with Part Compound or Sketches in Mapmode (further w/ Attachment Offset settings available)
- Also, I always have Autocad (10 years ago knowledge) block concept in mind: it is a compound of 'parts' with a distinguishable Insertion Point
- So it is my (own?) expectation to have the merged ArchWall (w/Additions / Subtraction) works like a compound / block
- Then, it goes like Additions / Subtraction's Placements are 'superimposed' over the parent ArchWall's Placement
- So, it doesn't relay on Draft Move to have the whole merged walls to move it/them altogether
- The 'Move with Host' to my understanding only have Draft Move command acknowledge it
- So that's why changing / inputting figures in the Host ArchWall's Placement won't have the whole merged Walls move together
So could 'Move with Host' at least be interpreted as 'Superimposed' children's placement? - This may affect existing models.
Or simply added an option ' Superimposed Parent/ Children's Placement'. This is a matter of intuition, convention, productivity and common usage of the current users I think.
Because of my intuition or preconception w/ (acad's) block concept, it seems I am experimenting with expressions, usage of sketch, codes etc. to 'imitate' those 'superimposition' behaviour. So, a facade I experiment > 1 year ago with many parts merged in 1 ArchWall may more easily like one object. Better still, I always want to find a workflow: Model a prototype building, move the whole building in actual 'true world coordinate' / an actual site, then every building corners' true coordinates could be read (screencaptures attached again below). I have that difficulty with current Arch object's properties as experimented and commented in other posts.