Can we have adjustable thickness to a composite?
For example, I want to have a composite of "concrete-insulation-stone", instead of:
I believe this is crucial.
How would you deal with the polygonal nature of the slab tool? It seems to me your case / example is limited in use to being a single rectangular shape, while the slab can have very complex shapes, where offsets won't make sense. What happens when you manipulate the borders of the slab?
If it is just a 'box' shape, why not use the beam tool with a complex profile with the modifiers?
The name of a tool is just a starting point. A roof can be a floor, a beam can be floor (great for making pre-cast concrete slabs), a slab can be a flat roof or a table top, a column can be a bit of wall, etc etc
I would suggest updating the OP with this image and perhaps a better description of the wanted functionality. As it stands now it's hard to understand the wish which makes it less likely to be acknowledged by GS.
So the wish is not to make any fundamental changes to how composites work but enhance how composites are managed and defined.
What happens when you change the thickness of a composites skin in model space?
Should it add the new configuration as a new composite?
Should it create subtypes of the composite with different thicknesses?
Should it become undefined in terms of composites?
I think the subtype approach has some appeal especially as it gives a better organisation of composites. Perhaps with the functionality to define different thicknesses in the Composite Structures Dialog Box that then works as snap points in model space.
The question is then what happens with composites with the same skin configuration in terms of thicknesses but differing in skin type and graphics?
It is easy my friend. It will work the same way as the profiles. I'll answer each one of your questions:
What happens when you change the thickness of a composites skin in model space? - The same as profile modifiers (i.e. columns in model space). It would update all relevant drawings. You should be able to change the thickness of a composite when it is cut (i.e. walls in plans and sections, slab only in section etc.)
Should it add the new configuration as a new composite? - No. Just like profiles. When you add a column and change its modifiers, AC does not create a new profile.
Should it create subtypes of the composite with different thicknesses? - No. No subtypes. Just the composite itself.
Should it become undefined in terms of composites? - No. It will use the same name.
I attach a picture of some columns that use the same profile. Everything stays under the main profile.
The problem with the profile modifier approach is indexing of elements. Today element listing parameters regarding structure are limited to name of material, composite, and profile. So if you make a list for the two elements in you example there's no way to tell if they have the same cross-section or not.
We can't have that for composites so modifying the skin of a composite would either have to make it undefined in terms of structure or it would have to create a new (sub-type) composite. Thinking of it - complex profiles should be amended with regards to this.
There are two sides to this, Art & Science if you will. Stretchy skins for modelling would be great. Tracking & documenting those stretches could be problematic. I don't think that is a reason not to have it, the responsibility just needs to fall firmly on the shoulders of the end user and the stretching needs to be a conscious effort that can't accidentally happen. This isn't really any different to introducing morphs into the model, they are equally difficult to quantify, but it is possible for the end user to document them and nobody is saying we shouldn't have morphs are they?
I'm not opposed to the wish (as i understand it). I agree that things can be done about how composites are defined and I do se some value in being able to define it graphically.
And my questions about how it should be implemented was not posed as obstacles to the wish but to clarify it with the goal to increase the likelihood of it being acknowledged by GS.
I don't really see any problem reconciling the two sides you mention. For example, when stretching the user can be prompted to choose to define a new sub-composite or to change the elements structure to "free-stretch composite".
But I can't really see how it could be implemented as described (like profile modifiers). I would even say that it would compromise the functionality of composites due to the indexing issue.
I agree to both of you @thesleepofreason and @DGSketcher. This may have pros and cons. I do not use listing-indexing elements in my workflow, unfortunately, since architects in my county are faaaar away from using BIM. I solely design and produce drawings. This is a model issue from me. I believe there should be an indexing solution to this as well, as @thesleepofreason said.
@Erwin Edel I think "offset" is probably a misinterpreted terminology. What I believe @Vasileios S_ is looking for is the application of "Offset modifiers", like we have in Complex profiles, to allow you to adjust a skin thickness in a placed element which is using Composite profiles.
So you would still have seperate slabs, but with a complex profile instead of a composite.
I see. You would be limited to only 'box' shapes in the profile, I guess, since it can only extrude up and down.
Wall skins can have modifiers for height and/or width - But having seen more of the slabs you want to model now, why would you not want to model all of the components (concrete and insulation) as individual elements of some kind for the sake of convenient and simple Interactive Schedules and easy adjustments, and Labeling in the Details ? A lot of slabs have uniform thickness insulation, but the concrete is sloped. I model sloped concrete as roofs, insulation as slabs, and I usually model the gravel and sand below them as well for the sake of scheduled volumes. I usually model the excavated earth as well as the backfill areas too since I need to know the volumes when I am making Excavation Plans. I do a lot of modeling of things that have multiple layers that vary in many ways, and because of this, I model each element with what ever works best for the sort of data I need to extract from it, and for the sake of the Drawings that are generated from the model. What I like about Archicad is that there is always some way to get what I want 🙂
I think there is some merit to the wish but I'm not really sure exactly what the wish is and I think it has to be clarified if it is to be considered by GS.
There is quite a difference between