Hi Michael,
I had a couple questions I’ll ask in the forum.. (I had a couple models I've sent you in email, but the answers may be relevant to the forum)…..

Could I get some brief info regarding this?

So in the model *******_draftB77 3dm file, I’m getting ready to look at how I will provide it without errors…. (I started with simply opening the model in rhino, which is what my questions are related to)

I get different results with different formats and would appreciate if you can speak to some of the questions to help me.

1. opening the 3dm directly into rhino and using “select bad objects”, highlights invalid areas….
(a. can this be a particular setting/setup issue with my rhino to make it match what MoI has written?
(b. Could just be rhinos system has an issue with how I have defined and created these surfaces?

IGES is worse….

A STEP export (model_test_2.stp) yields much better results and rhino appears to be reading and stitching entities on import. Importing this stp (into rhino) file yields only 3 things that need addressed. 2 objects which are selected with “select bad objects” can just be deleted and re-mirrored from their counterparts (a plank on the right side and a purchase block at the back)

At the front, the “net” has some bad surface in the import and I can just delete it, export that solid from MoI as a 3dm and replace it (front_rope_fixer.3dm)

This has rhino have a file with “no bad objects”….


2. Is this just a function of data exchange? Maybe you could speak just a little as to “why those particular objects are “bad” in rhino, though just a mirrored part from the MoI file?
3. would trying to export more “single items” help, as opposed to a “319 solid” file export….

Any info regarding this "main" 3dm file and having it be at it's best (with regard to translations/other apps) is greatly appreciated..

Thanks.

[EDIT] There is no "model fixing" needed or being requested here... It is more looking for observations on the model and process described...

Hi Burr,

> (a. can this be a particular setting/setup issue with my rhino to make it match what MoI has written?

No there isn't any kind of settings or setup type things that I am aware of that would have any effect on this.


> (b. Could just be rhinos system has an issue with how I have defined and created these surfaces?

Well, it depends on the particular case - Rhino flags an object as "bad" if it the Rhino check function complains about something in the model structure. Not all of these complaints are equal in severity, some are easily ignored and others can be more significant problems.

A whole bunch of them in your case here, those rope ties, just have some very slightly wrong flag on one trimming curve: "ON_Brep.m_T[11].m_type = ON_BrepTrim::seam but m_iso is not N/E/W/S_is" - means something like the edge has a flag indicating it's a seam of a closed surface but another flag on it is not set saying that it's an isoparm curve like it should probably be. It's unlikely that will cause you any actual problem.

I don't know why that particular trim flag is not set, it could be a bug in MoI's exporter or it could be a bug in the OpenNURBS 3DM writing library that MoI uses to write 3DM files, it would unfortunately take a lot of investigation work to find out more about it.

For those particular errors it's probably best to just ignore them, it's kind of more likely that you will end up with worse results on those doing a STEP import instead because that will involve a lot of recalculation of data - basically all the UV trim curves will be thrown out and they will all be recalculated by pulling down the 3D edge curves onto the surface. That's a lot of stuff happening just to get a single not really very meaningful edge flag to get set...


> 2. Is this just a function of data exchange?

Yeah I guess so - different programs can be finicky about different particular things.


> Maybe you could speak just a little as to “why those particular objects are “bad” in rhino, though just a mirrored part from the MoI file?

Well, by definition they're bad because they don't pass the "Check" function. You can get the specific error by selecting the particular object and running the Check command on it.


> 3. would trying to export more “single items” help, as opposed to a “319 solid” file export….

I wouldn't think it would make any difference...

To me it makes more sense to look at the specific problems that Check reports.

Not sure if that helps you any... ?

- Michael

"""""""""" just have some very slightly wrong flag on one trimming curve: "ON_Brep.m_T[11].m_type = ON_BrepTrim::seam but m_iso is not N/E/W/S_is" - means something like the edge has a flag indicating it's a seam of a closed surface but another flag on it is not set saying that it's an isoparm curve like it should probably be. It's unlikely that will cause you any actual problem.""""""""

Perfect!

""""""""I don't know why that particular trim flag is not set, it could be a bug in MoI's exporter or it could be a bug in the OpenNURBS 3DM writing library that MoI uses to write 3DM files, it would unfortunately take a lot of investigation work to find out more about it."""""""""""""

And again, perfect!

Thanks for taking a look at it for me....

Hi Burr, well those weren't the only problems that was just the majority of them about that wrong edge flag thing.

There are a couple more ones like a couple edges out of tolerance (a gap like 0.01 units instead of at least 0.001 units as the edge was marked), and a couple of edge endpoints that are not totally lined up with each other. There aren't as many of those but a couple of those are potentially worse problems than that edge flag thing, like there may be one slightly messed up edge in the one that has the endpoints not matching by a fair amount... Some of them are errors in the edge structure in the UV part of the trim edge where it's hard to judge just from a numeric value how much the error actually is.

But out of like 70 objects flagged, 58 of them are just that "iso flag not as expected" error, those are the ones named block_rope_tie, block_rope_tie_2, block_rope_tie_3, and tail_net.

- Michael

And if you want to get a more narrowed down view of the specific area of the model it's complaining about, select the object and run Explode to break it into individual surfaces and then run SelBadObjects, that will then show you the particular surfaces that have the "bad" stuff in them and you might look over those to see if there is a noticeable problem in their trim edges.

I think there used to be a command called ShowBrokenEdges or ebroken but it seems to have gone away when some different commands were merged into one "ShowEdges" command.

- Michael

Hi Michael,
Thanks again. I'm going to mark this thread and go back to things at finalization time and look at cleaning things up for a polish.....

I think the "rope and netting", I'll just redo, as it is just slapped out early in the model... You can see it is just not really "well formed" netting.

But some of the surfaces on the hulls and masts etc, I want to be sure are quality...

I'll use your input to inspect closely and rebuild when I find those bad trims....

Another way to inspect a surface that says it has bad trims on it is to select the surface and then run CreateUVCrv - when it asks you to select any additional curves just push enter. That will stick the UV space boundary curves into the Top view, use ZSA (Zoom selected all) to zoom to them. There will be a rectangle for the surface's outer boundary and then all the UV space versions of the trimming boundary inside of that. If some of those looks weird and chaotic like squiggling around or shooting off strangely off the boundary then that's a damaged UV curve, those can be repaired by untrim + retrim.

- Michael

Aha! :)

Invaluable Michael. Thanks!

Thought I made some beautiful blends on the hulls... Turns out there are some horrible self intersecting/twisted trims... :o

Repaired! probably spend a few hours going over everything in another week or so when it's done....

Thanks again....

No problem Burr! The model is really cool!

- Michael

Thanks michael. I will make a post about it when im done. Im making it for polynesian voyaging society and their UH education program.