In Architecture modeling, we will often encounter lots of simple variations based on the same block.
Say I made a curtain wall block for my building, the same block will likely covers 80% of the building façade, but there will always be 20% that will be covered by some simple variation of this standard block, it might be different level height, or extra shadow box for overhang, or raised shadow box for parapet, or smaller modules at around corner. In AutoCAD I can simply make them all essentially one block, but controls each variations using dynamic block functions, therefore control all of them at the same time; but in Rhino I had to make different blocks for each variations, and make things much more complicated than it needs to be.
Revits solves this problem with it’s family system and parameters. It’s such a basic and important feature in architecture modeling that it baffles me how Rhino never implemented it and haven’t even mentioned it in RH7 as well.
(And yes, I know how to use Grasshopper, but dynamic block is much more simple and elegant way to organize this kind of facade modeling, you don’t want to change and bake everything in Grasshopper every time when you make a simple change)
PS: While we are at this, might as well get the XCLIP function in AutoCAD as well.
There are different ways people are doing this for facades. There are a number of workflow issues that tend to influence which system is used:
Grasshopper Styles is one way to implement this. These are Grasshopper driven blocks: https://youtu.be/6jgzsxPxRyA . For that matter Visualarq will handle many standard wall and window conditions with the Styles.
There is also Elefront that can be combined to create facades and replace existing geometry on the fly with all the metadata also.
dynamic blocks require constraints which is not supported in rhino at all so dont expect anything like it. i just wish rhino implemented constraints but it seems it will never happen
We are always discussing new features and projects for Rhino. And in this case a mechanical style parametrics is looked at many times (straight line, arcs and extrudes). Yet, there are different parametric engines for Rhino for other geometries:
It is important to note there is not just one parametric system, but many. Getting the right one to fit the correct modeling is the trick.
Do Grasshopper based Dynamic blocks work? Give them a try in VisualARQ, thy can even be part of a windows definition driving a dynamic opening in a wall.
Because there are many so it naturally loses “elegance”. There should be optimally one robust parametric and constraint system inside Rhino and other plugins above it could make use of it. Lets say you could have parametric block inside Rhino which could be loaded into GH and parameters could be adjusted inside grasshopper which would unlock another universe of possibilities.
Thats why i am supporter of my own idea “Rhino NextGen” written from scratch, with new core, looking and feeling just like Rhino but it would be nextgen in its guts. With combination of grasshopper and resolving all the issues you have piled up it would be killer cad soft without competition. you would definately take over sketchup userbase, autocad userbase and many more. right now drafting capabilities are bit limited compared to obsolete autocad. BIM capabilities are only through 3rd party. Non existent constraints, non existent dynamic drawigns etc. With all of this implemented there would be only one choice for majority firms in AEC sector. Due to affordability it would be Rhino for sure.
Hi Scott, thank you for pointing out that VisualARQ has made progress in simple Dynamic Blocks. But if I may ask, apart from parametrics, has Mcneel seriously looked at the idea of real dynamic deformable Blocks? By dynamic deformable blocks I mean being able to apply bending, morphing, cage edit etc. while maintaining instancing. From what I’ve read in the forums it’s an extremely difficult thing to do programming-wise, but since you mentioned that you’re trying to add new features, could this be a possibility in the future release or is it really that monumental a task? I also have experience with max and other softwares and from what I’ve read none have done this. In my opinion this would be such a huge thing because it gives so much flexibility in the way we model and considering the memory constraints.
Yes, varying and instance beyond scale and rotation is a challenge. The way blocks are setup mixed with the complexity of the morph, they are currently incompatible. Some of that has to do with the actually math of the geometry changing once morphed. Using a group would get around the limitation, but then it is not a true instance.
Ok,scott. What about Block boolean? Is there a chance? I know that boolean operations on blocks are impossible right now but what about some type of clipping operation that mimics it? If I’m not mistaken clipping planes work on blocks also Vray does this. Can mcneel come up with something that clips/conceals a block based on a geometry say a curve or brep?With parametric extents. This is along the lines of what the maker of this thread said about dynamic blocks. A direct architecture example of this would be if I have many blocks of standard ply wood and I want to scale differently and scatter it but i need to cut holes on it to place lighting fixtures. I wouldn’t want the holes to scale as well so maybe a clipping volume could help.
