Jump to content
Sign in to follow this  
Pamela Galli

LL: Mesh is useless?

Recommended Posts


Da5id Weatherwax wrote:


Pamela Galli wrote:

I got this error:

 
Screen shot 2011-12-05 at 11.54.06 PM.png

 

BTW, all the models had the same one material assigned. 

 

I've seen that infuriating X appear in the loader a few times when I know everything has only the one material assigned too... The green "ship it" appears when I load the highest LOD and then changes to that error as soon as I select files for any of the lower ones. After selecting all the LOD files, try hitting the "reset form" button - I found that flipped all the X's back to nice green checkmarks and I was once again invited to ship it
:)

Yes, I think I did something like that. Loading a lower LOD model triggered it, and I think what I did to fix it was just reload the higher one. I can't remember exactly.

Sure look forward to having a nice Owner's Manual with Troubleshooting Guide!

Share this post


Link to post
Share on other sites

OK. I had to reinvestigate. So I downloaded the latest release (3.2.1(244864)Nov 10 2011) and beta (3.2.4(245931)Nov 30 2011) viewers and did some experiments. They have indeed reverted to requiring the same number of materials in the LOD meshes as well as requiring the names* to be the same. :matte-motes-agape: :matte-motes-crying:

This is now the case in both release and beta. In an older beta, the only requirement was that the lower LODs had to be a subset - it could be fewer than all of them. In the beta, but not in the release, you still get the "not a subset" error message, which makes me think the reversion is unintentional - no point in talking about subsets if they all have to be there! In the release viewer, you just get the greying out of the calculate button.

As far as the physics mesh is concerned, the release viewer is still applying the same requirements as the LOD meshes, that is both number and names of materials have to be the same. This is corrected in the beta viewer where there are no restrictions. You can even have more materials in the physics mesh than in the LOD meshes. This is as it should be.

I will table a question for the mesh/content meeting to ask if this reversion was deliberate. I might also glance at the source to see how it came about, if I can find the time.

ETA: checked dev viewer 246174. It's the same as the beta.

* "name" is used here as abbreviation for the material attribute of the <triangles> or <polys> tags.

Share this post


Link to post
Share on other sites

Braydon you wrote: "I have found that you can sort of reuse your UV map on lower LOD items by pinning the areas that define its size and location,  like around the edges in the uv editor,  and then asking it to unwrap again  and most of the time it will move the internal vertices of the UV map to compensate for the holes generated by removing ring loops."

I too hastily dismissed this workflow deciding the process of selecting the edges of the discontinuous lower lods would as much work as just moving things around.  I discovered that you can just select the entire UV set up, set the entire UV set to "pin" and re-unwrap.  It all reconnects nicely without having to pick and choose borders.  All the verts remain in place where they were on the higher LOD UV set.

Thanks so much for this tip. It's a huge timesaver when making LODs.

  • Like 1

Share this post


Link to post
Share on other sites

So I make a whole new bedpost (with Nacy's help). I make two more LOD models. All three models have the same material (one material each). I upload one and get this message:

 

Screen shot 2011-12-08 at 7.15.24 PM.png

 

Is there any information anywhere that can tell me what the problem is this time?  Because I don't know what this means. If I knew where to look it up, I would.

I have yet to be able to upload anything but the first LOD model.

 

Share this post


Link to post
Share on other sites

Probably the problem is that you made the LOD models, and added a material afterwards to each. This way the material names will be different for each LOD. A lately added requirement is that the material names of the lower LODs have to match the material name in the Hi LOD.

That's pretty annoying, I know. To fix that, you can edit the dae files itself within a text editor. Or you make your HI LOD mesh, add the material(s), and copy that mesh to do the lower LODs from it.

Share this post


Link to post
Share on other sites

Thank you -- do you have any idea why this is a requirement?  It will probably take several hours for me to figure this out and fix it (yes, I know, it is a snap for other people, but this is me).

Share this post


Link to post
Share on other sites

That means that your high LOD model uses another material. It may be possible that you have defined multiple materials all with the same settings, but assigned different materials to each LOD version. But i it also can mean :

 

  • Bug in the importer
  • The .slm file got in the way (again)
  • You unintentionally have exported more than you thought in to one of the .dae files
  • maybe you did not unwrap all model faces in one of your LOD models. Theoretically this could confuse the importer and let it assume, there is another material in your object.

 

Share this post


Link to post
Share on other sites

I believe the same name requirement is to make sure that all LODs show the same textures, on all faces, over all levels.

Assuming it's the material name problem;

You can try to select/apply the HI LOD material to the other meshes within Blender, from the material selection list, and export them again. That might be the easiest  way to fix it.

Share this post


Link to post
Share on other sites

Yes. The change was in respopnse to peoples' complaints that the wrong materials were appearing on the wrong faces in the lower LODs. This was a result of the earlier situation where face numbers were determined simply by the order, in the Collada file, of the <triangles> or <polys> sections defining the faces the materials were applied to. Most exporters did not provide an explicit way of controlling that order. Now the sequence of materials for the lower LODs are re-ordered to match the order of the high LOD, based on the "material" attribute of the <triangles> tag. In Blender (at least 4.9) this is copied from the name of the material. You can edit the attributes by hand, if you know which triangle list is which, but it's much easier to simply make sure you apply the same materials to all the LOD meshes, as Arton said.

Share this post


Link to post
Share on other sites

Okay, to make a long story short: I got my LOD models to load.  But the Medium one was not very good so I redid it. And now when I try to upload,  the Medium shows up as having the very same poly data as the high.  So, thinking I did something wrong when saving it, I saved it again.  Same thing -- it shows as the same as the HI.  So I deleted the .slm files and tried again. Same thing. 

Of course, this almost doubles the PE. Anyone have any suggestions why the uploader thinks the models are the same?

(Feeling more like a character in a Kafka novel every day. )

Share this post


Link to post
Share on other sites


Charlar Linden wrote:

If you're willing, send me the collada files and I'll take a look.

 

Charlar

charlar@lindenlab.com

Thank you Charlar - I kept restarting the viewer and Blender and re-uploading the models until I finally got them uploaded correctly somehow, but I don't know if it had to do with the Collada export or the Collada upload.

 

ETA:  Okay I screwed it up again so I am sending it :-)

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Sign in to follow this  

×
×
  • Create New...