Tue, Nov 26, 12:46 AM CST

Renderosity Forums / Poser - OFFICIAL



Welcome to the Poser - OFFICIAL Forum

Forum Coordinators: RedPhantom

Poser - OFFICIAL F.A.Q (Last Updated: 2024 Nov 25 12:38 pm)



Subject: My home-made morphs break up


wotsupdoc ( ) posted Wed, 21 August 2002 at 10:01 AM · edited Sat, 03 August 2024 at 12:22 AM

I've made a simple object in Rhino, saved it as an .obj, pulled a few ponts around, saved it as an .obj, loaded it into Poser, saved it as a .obj and with the morphing target thingie checked (at least now it stays in one place) I've loaded the main object in Poser and installed the morph objects for it The object consists of 4 parts and has 2 morphs. Now when I move the morph slider some parts morph ok, some break up. I don't know what I did differently for the other parts. It's driving me nuts! Can someone give me a few hints????


ronmolina ( ) posted Wed, 21 August 2002 at 10:15 AM

How about a screed shot. Ron


ronmolina ( ) posted Wed, 21 August 2002 at 10:16 AM

screen


terminusnord ( ) posted Wed, 21 August 2002 at 11:59 AM

It would help to see the problem. I've had a problem with C4D doing something similar, where it exports the same model after a few morph operations, but it gets the vertices in a different order. This happened when I accidentally re-ordered groups in C4D's object manager prior to exporting the model as a single piece. Take a peek in your OBJ file with a text editor and see if you can see any obvious problems in the ordering of the vertices. -Adam


wotsupdoc ( ) posted Wed, 21 August 2002 at 2:00 PM

file_20958.jpg

Thank god I have a kevlar screen. Here it is, I hope you can help me. The top left is the original. The top second left is morph 3 at .1 The top third left is morph 3 at .2 The top forth left is morph 3 at 1 The bottom first left is morph 2 at .1 The bottom second left is morph 2 at .2 The bottom third left is morph 2 at 1


terminusnord ( ) posted Wed, 21 August 2002 at 2:13 PM

I guess now all we need to know is what is supposed to be happening :-) It does still look to me like vertices got reordered in the morph somehow. The model is not 'exploding' the way it would if the scale or 3d-space positioning were screwed up. It's mostly staying put, just the vertices are getting the wrong deltas. -Adam


wotsupdoc ( ) posted Wed, 21 August 2002 at 2:24 PM

This is how I created the objects: I made three illustrator files, imported the first one into Rhino, extruded the first curve and made a mesh out of it (mesh command), deleted the extruded surfaces. I saved it as an .obj file, I imported the second illustrator file, placed the mesh over it, tuned on the control points and went to the "front" view window, selected all the points in one line ("z" dimension) and moved them to fit the new (illustrator) curve. I wanted to be sure that I had the same amount of points in all files. I repeated the same procedure for the last (illustrator) curve.


wotsupdoc ( ) posted Wed, 21 August 2002 at 2:35 PM

file_20960.jpg

THIS IS WHAT SHOULD HAVE HAPPENED.


wotsupdoc ( ) posted Wed, 21 August 2002 at 3:28 PM

The files aren't that large (100 k each) I can put it on the net and post the link here. I looked at the files in a text editor. I mean, all the points have 3 coordinates. What am I supposed to look for? There is a part where I find : vn -0.934054 0.357133 0 vn -0.885113 -0.464729 -0.0245207 vn 0 0 -1 vn 0 0 -1 vn 0 0 -1 vn 0 0 -1 vn 0 0 -1 vn 0 0 -1 vn 0 0 -1 vn 0 0 -1 vn 0 0 -1 vn 0 0 -1 vn 0 0 -1 vn 0 0 -1 vn 0 0 -1 vn 0 0 -1 vn 0 0 -1 vn 0 0 -1 vn 0 0 -1 vn 0 0 -1 vn 0 0 -1 vn 0 0 -1 vn 0 0 -1 vn 0 0 -1 vn 0 0 -1 Is this normal? f 624/624/624 629/629/629 610/610/610 f 620/620/620 638/638/638 657/657/657 f 696/696/696 684/684/684 697/697/697 698/698/698 f 698/698/698 697/697/697 687/687/687 692/692/692 f 684/684/684 699/699/699 701/701/701 697/697/697 f 697/697/697 701/701/701 700/700/700 687/687/687 f 687/687/687 700/700/700 703/703/703 693/693/693 f 693/693/693 703/703/703 702/702/702 686/686/686 f 692/692/692 687/687/687 693/693/693 695/695/695 f 695/695/695 693/693/693 686/686/686 694/694/694 f 682/682/682 688/688/688 696/696/696 f 688/688/688 689/689/689 684/684/684 f 689/689/689 670/670/670 699/699/699


terminusnord ( ) posted Wed, 21 August 2002 at 3:30 PM

if you could zip up the original, morph 2 and 3, I would be happy to take a look at them and see if I can verify the exact nature of the problem. -Adam


wotsupdoc ( ) posted Wed, 21 August 2002 at 5:03 PM

Attached Link: http://members.tripodnet.nl/wotsupdoc/Flame.zip

I tried so much combinations that I can't find back the old files I showed here. Anyway, these files only have 2 parts that morph right. However, they do have materials (the ones pictured here only had the preview material in render materials)

I appreciate all the help I get from you people!
Great to be helped so much!


terminusnord ( ) posted Wed, 21 August 2002 at 6:38 PM

Attached Link: http://www.airraidsirens.com/flamemorphs.zip

Try these. I reordered the points of the morphs to match the originals. In morph2 I was able to do it with a text editor alone, since there were materials groups to show me which faces were part of which flame part. For Morph3 I had to use cinema 4d's object manager to view the list of vertices visually, because this morph had no materials to guide me. Let me know if these work for you. I can explain in more detail what i did if you think you'll ever have to do it. Really i only wanted to diagnose the problem and I hope there is a way to fix it in Rhino prior to export. In morph2 the left 2 flame parts were reveresed. In morph3, the pieces were all moved except the middle one (a real mess!). -Adam


terminusnord ( ) posted Wed, 21 August 2002 at 7:10 PM

note that these new versions are "squished" morphs, in that they contain only vertex data (all a morph target needs anyways). That's why they are so small. Also, after I manual reordered the vertices, the face and normals are no longer valid anyways. -Adam


wotsupdoc ( ) posted Wed, 21 August 2002 at 10:10 PM

Yess, it works!!!!! (Ofcourse only if I keep the settings between -0.5 and 1) Thanks for the help Now the big question: how can I prevent this next time? Did I do something wrong or did I simply choose the wrong application (Rhino)? Somehow I don't think that I can blame it on Rhino, it is probably my inexperience that caused mesh to breakup, but how do I tell Rhino to keep it in the same place? I have started again working with Lightwave (Last time was 10 years ago on the Amiga computer) but have not mastered. the Lightwave modeller yet. You talked about the C4D's object manager. Cinema 4d was originally developed on the Amiga, but when Commodore went bankrupt, things went downill. When cinema 4d decided to stop support for the Amiga they gave away their Amiga version to the Amiga community. So I still should have an old copy of C4D somewhere. My Amiga was a really fast machine in those days with the PowerProcessor, I could fire up that old beast....


terminusnord ( ) posted Wed, 21 August 2002 at 10:32 PM

Unfortunately, I don't know Rhino well enough to say how this happened. I'm not sure if it is something you did when arranging groups, or something Rhino chose to do on it's own. I know in some programs OBJ groups are exported alphabetically how they are named, in other programs they are exported in the order they are placed in the programs hierarchy editor (as is true with C4D), and some other programs have other criteria. If you can figure out how Rhino is choosing to order the groups in the OBJ file, then you will know the answer. I think you need to take what we've learned here, and ask this more detailed question in the Rhino 3D forum. In the meantime, try re-exporting your original flame, then making as few changes as possible to create a morph, and then export again. Don't change any materials or groups or anything other than the placement of a few vertices in each piece of the flame. I would expect this to work. If it doesn't, Rhino is really got a mysterious export scheme!! -Adam


terminusnord ( ) posted Wed, 21 August 2002 at 10:35 PM

Could it possible be ordering the groups chronologically, last modified to most recently modified?? I hadn't thought of that one. Maybe you could try tweeking the groups in a different order each time, and seeing if this changes the order of the vertices in the file. Still though, someone in the Rhino group should have figured this out by now.


terminusnord ( ) posted Wed, 21 August 2002 at 10:39 PM

Could it possible be ordering the groups chronologically, last modified to most recently modified?? I hadn't thought of that one. Maybe you could try tweeking the groups in a different order each time, and seeing if this changes the order of the vertices in the file. Still though, someone in the Rhino group should have figured this out by now.


terminusnord ( ) posted Wed, 21 August 2002 at 10:41 PM

What made me think of this last one was the fact that the Morph3 you show in the picture above has a different arrangement than the one you included in the zip file. The one in the picture has the lower right piece not moving, and the one in the file has the center piece not moving (not as much anyhow, it gets a few vertices robbed away from it). -Adam


wotsupdoc ( ) posted Thu, 22 August 2002 at 7:37 AM

I don't think it is the chronological order, because after saving the file from Rhino the first time, I didn't change anything in the meshes, only in the layers. Yet the first time there was one part in morph 3 that morphed ok, and later [b]everything [/b] messed up while I had not changed anything in the mesh, only in the Rhino layers.

However, you are right, somebody else must have come aross this problem and solved it, I will look in the Rhino forum.

And again: thanks for all your help!


wotsupdoc ( ) posted Thu, 22 August 2002 at 7:38 AM

I don't think it is the chronological order, because after saving the file from Rhino the first time, I didn't change anything in the meshes, only in the layers. Yet the first time there was one part in morph 3 that morphed ok, and later everything messed up while I had not changed anything in the mesh, only in the Rhino layers.

However, you are right, somebody else must have come aross this problem and solved it, I will look in the Rhino forum.

And again: thanks for all your help!


terminusnord ( ) posted Thu, 22 August 2002 at 9:41 AM

Layers?? are the layers arranged or ordered in any way?


Privacy Notice

This site uses cookies to deliver the best experience. Our own cookies make user accounts and other features possible. Third-party cookies are used to display relevant ads and to analyze how Renderosity is used. By using our site, you acknowledge that you have read and understood our Terms of Service, including our Cookie Policy and our Privacy Policy.