Improved flat roof drawing: mesh with composite structure and variable or constant thickness

[Angel Berruezo]

Drawing a flat roof in Archicad can be tedious, often requiring numerous coping elements and solid operations. In contrast, Revit offers a more elegant and straightforward approach. This process could be significantly improved by:

- Allowing mesh elements to support composite structures

-Enabling the definition of composite layers with either variable or constant thickness

These enhancements would allow any flat roof to be modeled using a single mesh element, with subsequent modifications reduced to a matter of seconds. The same applies to inclined garage floors and other complex geometries, where current workflows demand excessive manual adjustments. Introducing these improvements would streamline the design process and significantly increase modeling efficiency.

[Laszlo Nagy]

I have a question about this wish:

Why Mesh, why not Slab?

Slabs are more suitable for this; the only thing needed is to have non-constant thickness skins in Slab composites.

[pmelo]

Hi László!
Ángel mentioned roofs as an example, but if you think about roads or pavements in an urban-scale project, Meshes become essential. You often need to control the Z-height of multiple points individually, which is exactly what Meshes are designed for. That level of flexibility wouldn’t be practical with Slabs alone.

[Laszlo Nagy]

@pmelo 

Let us please stick with the use case of the original poster's wish, which is

Flat Roofs modeled by composites with variable-thickness skins

If you want to make a wish about roads and pavements, please create a new wish. By the way, there is already a wish about Meshes with constant thickness, and it has already reached Upvoted status:

https://community.graphisoft.com/t5/Wishlist/Mesh-with-constant-thickness/idi-p/601155

[thesleepofreason]

---

@Angel Berruezo wrote:

The same applies to inclined garage floors and other complex geometries, where current workflows demand excessive manual adjustments.

---

@Laszlo Nagy  - GS's inability to deliver modelling tools that are general and efficient, instead focusing on specific use cases is certainly a cause for ACs catatonic development. The original poster obviously see a wider use here so let's not limit the wish to flat roofs.

[pmelo]

@Laszlo Nagy  
you asked why meshes. I just answered your question.
But again, even for flat roof, sometimes, controlling multiple points z height is much more interesting than calculating slopes.

[pmelo]

@thesleepofreason 

I completely agree with your point, and I'd like to broaden the conversation a bit.

It’s not that Archicad’s existing tools are inherently bad — they have served us well for a long time. As someone who has been a faithful Archicad user for over 20 years, I still truly enjoy working with it and appreciate its strengths. However, as users, our expectations naturally evolve. Today, we expect our models to deliver even more valuable, precise geometry and to better support increasingly complex design and documentation demands. It's about keeping pace with the needs of modern practice.

Looking back at Archicad’s development history (https://community.graphisoft.com/t5/Getting-started/Archicad-versions/ta-p/304207), it’s notable that the last fundamental improvements to core modeling tools came with Archicad 23 (released in 2019), when the Column and Beam tools were re-engineered and the Opening tool was introduced. Since then, updates have been valuable — some more than others — with features like Distance Guides, Design Options, and improvements to MEP modeling. However, they have largely focused on workflow, collaboration, and documentation, rather than significantly advancing the core modeling capabilities.

The issue @Angel Berruezo  pointed out with sloped flat roofs is a good example, but it reflects a broader trend. Many fundamental modeling workflows still rely too heavily on workarounds.

At the same time, it’s hard not to notice that other platforms have been pushing forward in modeling innovation. I say this with concern, not criticism: I would love to see Archicad take bold steps again in strengthening its modeling tools. Modeling is still the heart of BIM, and I believe Graphisoft has the foundation — and the loyal user base — to lead in this area once more.

[Laszlo Nagy]

Point 4. of the Wishlist Guidelines:

https://community.graphisoft.com/t5/Our-Community/Wishlist-guidelines/ta-p/631148

"4. One topic per wish per thread.
When one thread includes several feature requests, it is hard to keep track of and update them. If you have more than one feature idea, you can submit each one as a new wish. This way we’ll know exactly which wish community members are voting for.

Tip: Make sure that each wish has a clear, concise title to make it easier for viewer to understand what they may be voting on."

[Angel Berruezo]

Well, personally, I find the Slab/Roof/Mesh modelling situation in Archicad to be total nonsense. But to follow the guidelines, I would just submit another wish and comment there. 🙂

[pmelo]

@Laszlo Nagy  

Thanks for pointing that out.

Just to clarify — my intention wasn’t to introduce a different wish, but to expand on Ángel’s original example by showing how the same modeling limitations also affect other scenarios, like roads and pavements in urban-scale projects, where Meshes are crucial.

It’s really the same underlying issue: the need for more flexible, geometry-driven modeling tools that can handle complex surfaces without relying on heavy workarounds.

That said, I understand the importance of keeping the Wishlist organized, and if needed, I'm happy to reframe or separate it into a more focused wish.

[Laszlo Nagy]

@pmelo 

Yes, that would be preferable for the reasons above.