Oriented Bounding Boxes

[m.araujo]

"Hello World"

I'm hoping to write a program in .net that will generate a lumber schedule from a 3D model of a post and beam house. First I need to figure out how to extract that data. I have the feeling that the answer may lie in the bounding box. The problem is that I specifically would need an oriented minimum bounding box for each timber. The bounding box info in MASSPROP is given based on the current ucs and is not accurate when 3d objects are not aligned with the ucs. Anyone have any ideas?

[Anonymous]

Matt, You can explode the solids and examine the resulting entities or you'll need to use the brep library which is only availible
to .net in 09. Send me an email.

wrote in message news:5985825@discussion.autodesk.com...
"Hello World"

I'm hoping to write a program in .net that will generate a lumber schedule from a 3D model of a post and beam house. First I need
to figure out how to extract that data. I have the feeling that the answer may lie in the bounding box. The problem is that I
specifically would need an oriented minimum bounding box for each timber. The bounding box info in MASSPROP is given based on the
current ucs and is not accurate when 3d objects are not aligned with the ucs. Anyone have any ideas?

[dmarcotte4]

I would be interested in what you guys come up with, I know it’s possible as these guys can do it
http://milllister.com/

[SEANT]

I’ve have explored this issue with VBA (see http://www.theswamp.org/index.php?topic=21524.0).

The general methodology of the routine is to WCS orient a copy of the 3DSolid via PrincipalDirection and cross section analysis. The analysis also includes some rudimentary cross section recognition – if a library of parts is pre-processed.

As mentioned by Mr. Richardson, the new 2009 .NET capabilities will no doubt make this analysis and manipulation more robust. I will likely explore that same venue myself in the coming months.

I was, however, pleased with the general performance of the VBA example as posted at The Swamp.

[Anonymous]

Hey Daniel - I assume they must be using native ARX and the brep library. Other than that the only way would
be to explode the solids as far as I know... btw I wasn't trying to hide anything with the OP by having him email
me - I just know him. I will gladly post anything else we come up with.


wrote in message news:5985901@discussion.autodesk.com...
I would be interested in what you guys come up with, I know it's possible as these guys can do it
http://milllister.com/

[Anonymous]

I've use the slice method in VBA - the issue for me was it was a bit slow and did not handle compound angles.
The nice thing in .net is you don't need to add the resulting entities from slice, explode, ... to the database so it
will be quicker. If you're using 09 I would examine brep - Autodesk did a webcast recently you might want to
check it out.

wrote in message news:5985920@discussion.autodesk.com...
I've have explored this issue with VBA (see http://www.theswamp.org/index.php?topic=21524.0).

The general methodology of the routine is to WCS orient a copy of the 3DSolid via PrincipalDirection and cross section analysis.
The analysis also includes some rudimentary cross section recognition - if a library of parts is pre-processed.

As mentioned by Mr. Richardson, the new 2009 .NET capabilities will no doubt make this analysis and manipulation more robust. I
will likely explore that same venue myself in the coming months.

I was, however, pleased with the general performance of the VBA example as posted at The Swamp.

[Anonymous]

I've use the slice method in VBA - the issue for me was it was a bit slow and did not handle compound angles.
The nice thing in .net is you don't need to add the resulting entities from slice, explode, ... to the database so it
will be quicker. If you're using 09 I would examine brep - Autodesk did a webcast recently that you might want to
check it out.

wrote in message news:5985920@discussion.autodesk.com...
I've have explored this issue with VBA (see http://www.theswamp.org/index.php?topic=21524.0).

The general methodology of the routine is to WCS orient a copy of the 3DSolid via PrincipalDirection and cross section analysis.
The analysis also includes some rudimentary cross section recognition - if a library of parts is pre-processed.

As mentioned by Mr. Richardson, the new 2009 .NET capabilities will no doubt make this analysis and manipulation more robust. I
will likely explore that same venue myself in the coming months.

I was, however, pleased with the general performance of the VBA example as posted at The Swamp.

[SEANT]

It would probably be slow with numerous Solids, not so bad, perhaps, with dimensioned lumber. Can you describe the "Compound angles" issue a bit further?

I had actually signed up for that webcast but was forced away from the desk that whole day. I don't suppose you've heard how it can be retrieved.

[Anonymous]

I wasn't doing stright dimensional lumber so any complex solids rotated at multiple angles from the wcs were tough. I don't know
were they keep the Webcasts for download - I did a quick google and didn't come up with anything...


wrote in message news:5985957@discussion.autodesk.com...
It would probably be slow with numerous Solids, not so bad, perhaps, with dimensioned lumber. Can you describe the "Compound
angles" issue a bit further?

I had actually signed up for that webcast but was forced away from the desk that whole day. I don't suppose you've heard how it can
be retrieved.

[SEANT]

I was able to locate a lot of webcast, except that one. Curse the luck.

The routine at The Swamp does a pretty good job with solids at complex orientations. It can also handle a reasonable amount of end treatments and mid-span features. That capability, unfortunately, makes the routine even slower than you may have originally imagined. 😞

Thank goodness for speedy computers. 🙂

[Luis Esquivel]

It is here:

http://www.adskconsulting.com/adn/cs/api_course_sched.php

Look for:
AutoCAD : Brep .NET

[SEANT]

Luis, I'm very much obliged!

[dmarcotte4]

Well I started a kludge fest over at theSwamp if anyone wants to jump in
http://www.theswamp.org/index.php?topic=24045.msg290829#msg290829