IFC to citygml

[Pdongjaelee]

hello.
Thank you for telling me about the ifc and citygml conversion tools last time
I succeeded in converting two ifc files to citygml.

But I have two problems.

First, large ifc files cannot be converted.

When using ifcConvert0.8, is there a capacity limit or an object that prevents conversion?

Second, I don't know how to compare with existing ifc objects after converting to citygml.

How do I confirm that this wall from ifc is this after converting it to citygml?

When converting existing ifc to glb, objects were checked using guid.

[Moult]

There isn't any capacity limit, it's probably just due to the complexity of the file that it takes a long time. Have you tried converting to other formats to rule out whether it's an IFC issue or a CityGML issue?

You will also need to monitor your system resources, are you swapping? Is it crashing? What does memory and CPU look like?

GUIDs are a good way to correlate the objects.

[Pdongjaelee]

Oh, I haven't tried it in any other format.

Oh, so the conversion of a slightly complex IFC file stops but is that because my computer's performance is not good?

The CPU has 8 cores.

Will this computer be enough?

[Moult]

There are a lot of factors, without knowing the file contents we can't help that much.

[Pdongjaelee]

Would it be helpful if I could give you the IFC file?

[Moult]

Yes? dion@thinkmoult.com

[Pdongjaelee]

Please check your email.

[Moult]

There seems to be a bit of confusion, there is no such thing as "LOD" in IFC. LOD is purely a CityGML concept. You cannot detect whether or not an IFC has a particular LOD. You can use heuristics, like if it has a bunch of walls and roofs it's probably not LOD0. If it has vertices more than X it's probably not LOD2, and so on.

The model geometry is pretty small and loads quickly and looks like a standard building. I suspect where it's starting to take forever is in the CGAL processing (closing gaps, padding, etc). @aothms will know the details of exactly what is computationally expensive and what isn't, but I think unless you're willing to take a deep dive into the code and learn about what algorithms are being used, you'll be hitting a dead end: this is academically fascinating, but it isn't off-the-shelf ready to throw any BIM model at it and expect results immediately.

In the meantime you might be better off in loading it in Blender and custom modeling the geometry you need for CityGML using OBBs and various modeling and retopology tricks rather than relying on a one-size-fits-all algorithm.

[Pdongjaelee]

but it isn't off-the-shelf ready to throw any BIM model at it and expect results immediately.
-> Are you referring to the Subway.ifc file I gave you?

[Moult]

I haven't run it using this particular repo, but yes. I only opened it to make sure there was nothing inherently complex in the IFC itself.

[Pdongjaelee]

Oh, we tried 3 files and they didn't work. Could you please check one more file?
I wonder if it's the same problem with the SUBWAY.IFC file.

[Moult]

I'm not saying the file works using the conversion, I'm saying the file's geometry can be processed from IFC ... whether it can be auto simplified is another matter.

[Pdongjaelee]

This file has been successfully converted. What is the difference in the conversion? 2024년 5월 8일 (수) 오후 2:35, Dion Moult ***@***.***>님이 작성:

…

I haven't run it using this particular repo, but yes. I only opened it to make sure there was nothing inherently complex in the IFC itself. — Reply to this email directly, view it on GitHub <#2 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/A2VN66LZZ7HD63G2Q62SZ2LZBG2S7AVCNFSM6AAAAABGSJHKTOVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDAOJZG43TMNRXHA> . You are receiving this because you authored the thread.Message ID: ***@***.***>