General Intersection Rule (SOL/1) - false positives with "Duplicates"

[Stefan L.]

Hi,
I get some weird results with the General Intersection Rule (SOL/1) with only “duplicates” activated:
The two pipes are declared as duplicates, although they don’t even touch each other.
Different GUIDs, (slightly) different sizes, different systems…

Information says (my own translation de-en):
Depth, width, height and volume of intersection is:
(A) Belag.-2.2047, (A) Belag.-2.1747, 18 mm, 8 mm, 28 mm, 0 l
Geometry of components is equal, but types are different

There are some more cases like this. Am I missing something here?

–
Stefan

[Stefan L.]

… still helpless…

[agroni]

Have these two elements been maybe colliding in previos versions of your IFC model?

[Stefan L.]

Thanks for your reply, and sorry it took so long.
I checked it in a new smc file with newly loaded IFCs, they - and others - are still marked as duplicates.

–
Stefan

[agroni]

strange.
and do they have different GUIDs?

Also, I don’t think you need to exclude Architecture and other disciplines if you are Including only the HKLS discipline

[Stefan L.]

Yes, different GUIDs, different geometry, different system, different name…

Regarding the include-exclude-thingy:
Yes, it looks a bit weird, but it’s a saved filter I use in different places to switch back and forth.
Does no harm I suppose…

–
Stefan

[Matti]

@Stefan-L, could we have a small example model for this case? You could remove all other components than these pipes. We could investigate what is going wrong here?

[Stefan L.]

Sure, but not before next week, we have a deadly deadline here…

–
Stefan

[Stefan L.]

Sorry it took so long, had a few days off…

OK, here you go:
I had to recreate the file with those IFCs where the error occured, but this time, different duplicate objects were found. Obviously this is not a constant error. I found like 15 obvoius false positives in a 300mb smc file with about 358k elements.
I then looked for a simple example and removed all other objects. You still can see all checking results, the false positive is the only “accepted” one, and there’s an issue too.
If you check the model in this state, everything is fine, so this really seems to be context specific.
The 300mb file is a bit more private, so I won’t post it here.
[Edit: still trying to upload the file ]
[Edit 2: OK, I’m obviously too dumb to upload smc or zip files (1mb). Can someone give a hint? ]

–
Stefan

[JSN]

@Stefan-L said in General Intersection Rule (SOL/1) - false positives with "Duplicates":

h those IFCs where the error occured, but this time, different duplicate objects were found. Obviously this is not a constant error. I found like 15 obvoius false positives in a 300mb smc file with about 358k elements.
I then looked for a simple example and removed all other objects. You still can see all checking results, the false positive is the only “accepted” one, and there’s an issue too.
If you check the model in this state, everything is fine, so this really seems to be context specific.
The 300mb file is a bit more private, so I won’t post it here.
[Edit: still trying to upload the file ]
[Edit 2: OK, I’m obviously too dumb to upload smc or zip files (1mb). Can someone give a hint? ]

Origin Marker.ifc

[Stefan L.]

Thanks for replying, I tried this button several times and got this message. Maybe I’m too ‘young’ here?

–
Stefan

[Eric]

@Stefan-L You’re simply hitting the limit for attachments, anything larger than 10Mb won’t pass the upload here. If you need to share a larger file, please use another service, thanks!

[Stefan L.]

But the file in question is only 1mb

–
Stefan

[Stefan L.]

OK, maybe this way:
https://gerberarchitekten.sharefile.com/d-s0ca86c497574136b
Link is valid for one week…

–
Stefan

[Eric]

@Stefan-L I take the liberty to attach your file here:

200515-FalseDuplicates07.smc

Apologies since there was actually a misconfiguration in our reverse proxy to effectively allow file uploads (images weren’t affected though…).

[Stefan L.]

Thank you!

[Stefan L.]

Am I the only one with this issue?
It’s still happening here…

–
Stefan

[JSN]

I just took the liberty to check the previous uploaded file as I was curious what you are facing and it seems to be ok.

Are you still having the same issue in that file as well? Or do you have a new example file containing the recently depicted elements?

[Stefan L.]

Hi, thanks for replying.

The last two pics are from “the same” file - with updated models of course.
Some more things:
It only occurs within those HVAC files generated with “mh-BIM” (“mh-Software” is a german company), files from other packages (Archicad, Revit, Trimble Nova) are not affected.
While preparing the file for upload some weeks ago, I realized that the false positives are changing or vanishing depending on the models loaded by the time of checking.
Currently there are 9 files from mh-BIM loaded. One of these checked in a separate file probably wouldn’t show the error.
So a file with only those two elements wouldn’t show this error too.

–
Stefan