@Karam_Baki has just written an incredible post on the “Al Wasel” tower in Dubai that delivers a state of the art building that focuses on sustainability, beauty, aesthetics and energy efficient 
The delivery of this tower leaned heavily on Dynamo’s ability to do the grunt work and automate away tedious tasks, as well as some bonkers awesome physics and math based solvers to achieve remarkable outcomes.
Come take a read and let him know what you think! 
Me too! The geometrical outputs are simply stunning in all cases.
The Freeform stuff is amazing
I’m really happy to hear Marcel, and I really hope you’ve enjoyed the Al-Waser tower blog post.
Most of architects i think would try to resolve with Rhino Grasshopper and the geometry not to be parametric families of Revit so resolved in Rhino and geometry converted to revit components. Happy to see someone uses parametric families with formulas and options to toogle with Dynamo to achieve goals and study behaviours
Thanks Ruben,
One of the lessons I’ve learned overtime is that it is actually more practical to use Adaptive components, even for at least hosting information and data.
If the geometry was more complex, then it would indeed require “Programmatic” geometry, but still over an adaptive component instance, as you can host any sort of fabrication related or information related data per element with ease, and communicate these data with any other project party along the line.
A very good example is this video:
where I showcased how we utilized adaptive components as placeholders, while Dynamo did the geometrical heavy lifting, and at the same time we injected all information required to proceed for further steps in the process and communicate all the data easily with anyone requesting it.
I stopped using Adaptive families because the revit categories of MEP are not compatible, cannot change the Adaptive family category to others that i need, for example I cannot nest MEP families or nest adaptive family on MEP loadable family, and I work with MEP for fabrication models. I think with structural families also happen same.
So if using Adaptive families, all nested families must be made with adaptive families as well. But obviously this is not going to happen because certain elements would lose their behavior like pipe fittings inserted in host of pipe.
Revit has limitations and 0.8 mm line length which is fun sometimes. I suppose the facade panels can contain nested families of thinner plates or frame or other components. But if inserting the geometry as CAD import doesn’t bother this limitation to Revit.
Loadable families like generic model template is okay but would be better to be the adaptive families template by default in Revit because more geometrical capabilities.
Seems you got the wrong idea; a placeholder means a merely information holder and has consistent XYZ positioning reference that exist in space.
Anything else can be dealt with programmatically regardless of the adaptive component geometrical and category changing capabilities.
I have never seen creating MEP family programatically with Dynamo with same capabilities of adding MEP connectors. If that is possible let us know please.
I had the bad Revit experience to model a parametric chain with adaptive generic model family and wanted to nest it in a pipe fitting family and Revit doesn’t allow it, appears a warning window explaining you can’t mix adaptive family with non-adaptive family. Changing category of Generic model family to MEP category is not allowed too is what i see using Revit.
I created an adaptive family with 8 points which looks like a conduit fitting but not been able to have connectors.
Yes it is possible, though I have not done so in Dynamo.
There is this class which I believe has the required constructors: ConnectorElement Class
Or if you start with the right family template there is the initial connector which can be duplicated. Or you can copy a connection into the document from another family document with a preconfigured connector.
You are still limited by the same limitations of Revit, such as not being able to put a connector into an adaptive family.
As @jacob.small and @sovitek stated, you may add the connections programmatically.
Also, looking at the things from a broader point of view, let me explain more about the approach I talked about.
Once you have adaptive components placed, they can act as a spatial element that can coordinate any piece of other geometry/element no matter what category, what feature, and whatever behavior it does.
Imagine yourself just designing and swimming inside the spatial code of each adaptive component, each adaptive point represents a coordinate, each multiple adaptive points represents a line or a curve, each surface can be lofted, have precise coordinates, can be offset inside, outside, any direction you want. Can be utilized to place any elements above, besides, bellow, simply anywhere relative to that element, and best of all. Everything can be stored, explored and shared via those adaptive component’s parameters.
It’s a matter of thinking differently and how to utilize every possible exploit to achieve the goal. and everything is achievable.
However, I would indeed encourage Autodesk’s Revit team to invest more in the adaptive components, as they have HUGE potential to further benefit from their natural behavior on the long run in multiple other projects.
I hope this opens up a possibility for many others to also think in similar approach.
I have chatted with several developers who would love to do so. However the bulk of the user base doesn’t see the potential or understand how to utilize what we have now, and as a result there are many voices asking for ‘make rhino imports better’, not caring that imports and links result in data destructive workflows due to the nature of the incoming content. 
As a result it is likely that more advocacy and education on what is there now has to happen first, and stuff like the AU session you linked and the blog post which started the convo go a long way. 
It is also worth noting that you don’t necessarily need adaptive components to utilize families as a placeholder. They certainly allow a greater degree of geometric flexibility, but data and references can be held in empty generic model families just as well for the bulk of content.
Well, let’s hope these case studies does indeed push the development of such valuable aspect in the Revit platform, in fact, the Al-Wasel Tower utilized the full extent of the adaptive components, which I will talk about in the next part of the blog hopefully soon,
This project in specific had all it’s geometry utilized through the adaptive component instances. On the other hand, the other Airport project utilized them as placeholders…
But to be honest if you asked me what mistake I did in the Al-Wasel Tower, it was actually utilizing ALL the geometry in the panel family. This lead to huge loading times, sometime 6 or 9 hours of just reloading an updated family, to separate zones, each zone is around 4 levels only.
This lesson lead me to utilize this concept differently in the Airport, and allowed us to cut the loading time to less than 15 mimutes, using the approach I described in the video.
The developments I’m really eager to see in regards to the adaptive components is related to performance primarily, then everything else can gradually be enhanced… That’s my take on it
Oh, and in regards to the enhancements done for better rhino imports, actually they are affecting even Dynamo imports, everything is getting better for the geometry interoperability, let’s just hope they don’t kill the dwg branch that heals the geometry, as we figured out a way to utilize that along with the new shapeimporter that recognizes textures and understands multiple formats. I’m excited to talk about this later, remember the node called KFamilyInsert? it is now full C# with 4k+ optimised lines of code that handles everything and heals the geometry perfectly, @solamour have seen it before and feel free to ask him more about it, until then, stay tuned for more.
