In #comfyui, the proportions of the interior and its original details are, of course, in place. But there is no speed advantage here, because you have to insert items one by one, not all at once as in ChatGPT, and wait for rendering each time. Besides, such workflow can be built only on Flux, and it is not fast by itself.

There are a lot of “Photoshop is dead” posts on the internet right now - you can edit images in LLM using just words. The story here is that even if the accuracy will be corrected in the future, the input of commands using text or voice is not suitable for any complex editing. Actually, people in the office do not communicate with each other in this way: to explain something properly you need sketches, masks, marks... in short, the functionality of Photoshop, otherwise no one will understand the task, neither a human nor AI (even if the AI will be smarter than a human, the problem here is not in intelligence, but in the fact that not everything can be formulated in words in a reasonable time). So AI model is not enough, you also need an interface with tools for graphical editing, that is AI inside the analogs of Photoshop, Revit, Rhino and everything else.

But, overall, sure cool, I still have some ideas on what can be done with these new ChatGPT functions

made a video about the current stage of 3D models generation with #comfyui

link to the custom nodes I used (there are also examples of workflow on the same github):
https://github.com/kijai/ComfyUI-Hunyuan3DWrapper

an overview of what you can do with #comfyui

How is the workflow of Rhino+Revit via Rhino.Inside organized?

In general, #revit uses standard families, both loadable and system, to represent geometry from #rhino, and they can be easily edited with the normal tools. In cases where standard families cannot be used (due to Revit limitations, as not every shape can be created in the family editor), the geometry is translated through DirectShape. DirectShape edit in Revit is no longer possible, but it is not a simple import, there are more possibilities: you can assign category, parameters, material from which you can then calculate volumes and areas.

On the example in the attached video DirectShape geometry used in a very minimal way. But this building is quite simple in shape, in other projects the situation may be different.

It is important that on any project, even a simple one, there are parts of the facade that cannot be done with standard Revit tools. And there are two ways out: to significantly simplify geometry or to keep details, but together with Rhino. Both options have the right to life depending on the project objectives and client requirements.

Project: Office building in Stuttgart, developed by Oliv and WENZEL + WENZEL
My part: developing facade geometry and BIM model for detail design phase

Why it can't be done in Revit + Dynamo(?):

1. Dynamo is a node-based programming language, i.e. you can create there only what obeys logic, the project may not be logical (in fact, it is never logical), elements may be located on the facade with unequal spacing, in random positions from the point of view of logic. That is, you need an interface that simultaneously allows you to program complex geometry and model it manually. And it is desirable that the script can create some parts, then allow you to manually change something, and pick up the modified result for further processing. In Rhino+Grasshopper you can do this. And then Rhino.Inside comes in to translate the result into Revit.

2. A complex project does not consist of a single script, different script parts are responsible for different tasks, which must be run separately, and sometimes it is useful to run scripts in a certain sequence. In Dynamo it is not convenient to organize such workflow, while in Grasshopper it is possible not only to keep several scripts open at the same time and transfer information between them, but also through one script pick which of the Revit files to work with (it is important when the project is divided into links). Also one script can distribute information to several files at once.

3. In Dynamo+Revit there are simply no functions to create many, even quite trivial geometric bodies or their creation takes a lot of time, for example, on the screenshot I highlighted a curtain wall panel, which is simultaneously curved along the radius in the plan and the bottom of the panel is trimmed along the radius on the front. If you use only Revit, then you have to build two shapes and with one volume cut another. Whereas in Rhino you model an object in the correct position with a couple of buttons, and then the script places it in Revit in the same coordinates. If there are many such objects, then the modeling in Revit will take forever. And it is only the panel, its shape repeats the bottom mullion and the wall under it, in general there is a whole set of elements stretches along the chain.

4. In Dynamo there are no ways to optimize and analyze the geometry, while in Grasshopper you can calculate options and select the best, including automatically by criteria.

5. It is impossible to create complex mesh geometry in Dynamo, and many things in architecture can only be created via mesh

6. From Rhino you send geometry to several addresses, not only to Revit, but also to the rendering department to create images and animation, engineers and LEED consultants for calculations, to workshop for 3D printing models, to manufacturers for creating mockups and at the next stages directly for production.... And although some of the cases mentioned above are applicable to Revit as well, in general case, a facade assembled in Rhino will be a more convenient material for processing and exporting in most related tasks.
....
#revit #rhinoinside