IFL and Morph animations at once
by John Cooney · in Artist Corner · 05/16/2006 (5:44 pm) · 13 replies
Has anybody experieced crashing in MAX 8 when trying to export an object utilizing the above animation techinques at once? I'm trying to export this tower collapsing and after about a half hour of trying to export, MAX 8 just crashes. The animations work fine individually, but not when the're together. Thanks in advance.
#2
05/17/2006 (1:41 pm)
I'd personally stay away from morph animation if you can help it. If its possible to create a skeletal rig to create your animation, that's always the better option as far as Torque is concerned.
#3
The skelatal rig could be simple point helpers. You could even use Boxs then do some reactor stuff to key in a good animation. At minimum you could animate the thing falling down or whatever. I believe a morph animation just pops from one shape to the target shape. More of a switch than a morph.
Matt
05/17/2006 (9:12 pm)
Ya you would be better off using a skelatal type rig. As long as you don't need any LOD. The ifl won't go through the Unmess. You already know that.The skelatal rig could be simple point helpers. You could even use Boxs then do some reactor stuff to key in a good animation. At minimum you could animate the thing falling down or whatever. I believe a morph animation just pops from one shape to the target shape. More of a switch than a morph.
Matt
#4
05/19/2006 (10:43 am)
Thanks for the replies. I'm trying to use bones right now, expecially since Morph animations take FOREVER to export. I did get an error though when trying to export the bones - " Error: bone missing from shape ". And whether I try using bones or point helpers, as you suggested Mathew, I get that same error. I'll take a look at the Character Pack to see how bones get exported correctly, cause I just can't export using any sort of skinned mesh. Again, thanks for your feedback though. 
#5
Logan
05/19/2006 (3:13 pm)
A bone error like that is due to the fact that you have a bone listed in your Skin modifier that is being culled or ignored by the CFG (or vice versa), this obviously creates a loop that the exporter does not like and you need to fix in your files.Logan
#6
05/19/2006 (4:33 pm)
Hmm, so just for testing sake, I exported a character that was bought from GG that was skinned to a bone rig and it exported fine. I unskinned the character and deleted all but one bone. I then skinned a cube to that one bone, weighted all verts to that one bone, made a .cfg file that included the bone and mesh for exporting, and saved my MAX scene and CFG file in the same directory. And upon exporting just a skinned simple cube, I get that same error ( Error: bone missing from shape ). Should I not be deleting bones from a whole skeleton heirarchy? I even tried just creating a new bone and skinned that, and I still got that error. I believe things worked fine in this situation when I used MAX 7, so could this just be a problem with MAX 8? Although I tried using MAX 4 and I got the same error.
#7
there is an issue with the skin modifier that might be worth mentioning.. in older versions of max (not sure about the new ones as I never tested).. if you delete a bone from the skin list in the skin modifier, it removes in from the interface, but did not rebuild the skin list array.. so when the exporter asks the modifier for the bone list, it finds a bone in the list which is not in the shape.
also, the skin will send a list of bones to the exporter, and the exporter will look for a bone in the 'shape'.. the shape being the heirarchy defined as a valid DTS shape (as per the docs) if it cannot find the bone in what it thinks is the shape (can't find the proper node tree or detail markers) it will let you know it cannot find it.
hope this helps a bit.
05/19/2006 (6:24 pm)
Post some screenshots of the heirarchy and the names of the nodes and meshes.there is an issue with the skin modifier that might be worth mentioning.. in older versions of max (not sure about the new ones as I never tested).. if you delete a bone from the skin list in the skin modifier, it removes in from the interface, but did not rebuild the skin list array.. so when the exporter asks the modifier for the bone list, it finds a bone in the list which is not in the shape.
also, the skin will send a list of bones to the exporter, and the exporter will look for a bone in the 'shape'.. the shape being the heirarchy defined as a valid DTS shape (as per the docs) if it cannot find the bone in what it thinks is the shape (can't find the proper node tree or detail markers) it will let you know it cannot find it.
hope this helps a bit.
#8
05/22/2006 (9:23 am)
Well, I've been trying to post a screenshot of my heirarchy setup within MAX but I can't figure out how to post images on these forums ( and yes, I have looked at the help link below ). So, I'll at least post how I potentially fixed this issue. Instead of leaving my bones outside of the heirarchy, I grouped my bones beneath my "start" node where all the other nodes are grouped while leaving the mesh that is skinned to these bones grouped outside of the heirarchy. Now things seem to export and work fine in Torque, although I'm not sure my visibility animation is working on my skinned mesh but thats minor compared to getting this thing to export. : ) Thanks for all the feedback though
#9
05/22/2006 (10:15 am)
Oh, and to fix the visibility problem, I guess it would help if I had the visibility box checked in my sequence helper.
#10
this is good.. the 'shape', as in the vaild DTS shape, is what is beneath the start node in the shape tree of what is considered a 'valid' DTS shape. the exporter looks for the shape to be setup this way to determine what is 'in' the shape, and what is not. the skinned meshes are a special case. It sees that there is one, and if it has a trailing number that matches a detail marker (or has mutlires set up appropriately) it says, aha, this mesh is skinned to my bones, and looks to 'the shape' to see how the bones are linked.
the bones in the skin list in the skin modifier have to exist in the shape, and the skinned meshes have to be outside that heirarchy.
this is different than a normal non-skinned mesh, which has to exist inside the heirarchy. seems like you are getting it to work though.
05/22/2006 (11:12 am)
@John, this is good.. the 'shape', as in the vaild DTS shape, is what is beneath the start node in the shape tree of what is considered a 'valid' DTS shape. the exporter looks for the shape to be setup this way to determine what is 'in' the shape, and what is not. the skinned meshes are a special case. It sees that there is one, and if it has a trailing number that matches a detail marker (or has mutlires set up appropriately) it says, aha, this mesh is skinned to my bones, and looks to 'the shape' to see how the bones are linked.
the bones in the skin list in the skin modifier have to exist in the shape, and the skinned meshes have to be outside that heirarchy.
this is different than a normal non-skinned mesh, which has to exist inside the heirarchy. seems like you are getting it to work though.
#11
this is good.. the 'shape', as in the vaild DTS shape, is what is beneath the start node in the shape tree of what is considered a 'valid' DTS shape. the exporter looks for the shape to be setup this way to determine what is 'in' the shape, and what is not. the skinned meshes are a special case. It sees that there is one, and if it has a trailing number that matches a detail marker (or has mutlires set up appropriately) it says, aha, this mesh is skinned to my bones, and looks to 'the shape' to see how the bones are linked.
the bones in the skin list in the skin modifier have to exist in the shape, and the skinned meshes have to be outside that heirarchy.
this is different than a normal non-skinned mesh, which has to exist inside the heirarchy. seems like you are getting it to work though.
05/22/2006 (11:39 am)
@John, this is good.. the 'shape', as in the vaild DTS shape, is what is beneath the start node in the shape tree of what is considered a 'valid' DTS shape. the exporter looks for the shape to be setup this way to determine what is 'in' the shape, and what is not. the skinned meshes are a special case. It sees that there is one, and if it has a trailing number that matches a detail marker (or has mutlires set up appropriately) it says, aha, this mesh is skinned to my bones, and looks to 'the shape' to see how the bones are linked.
the bones in the skin list in the skin modifier have to exist in the shape, and the skinned meshes have to be outside that heirarchy.
this is different than a normal non-skinned mesh, which has to exist inside the heirarchy. seems like you are getting it to work though.
#12
05/22/2006 (2:37 pm)
Well, apparently, when a mesh that is skinned to bones is placed outside the heirarchy, the exporter doesn't process the visibility animation that I have assigned to it, since I want the skinned mesh to turn on/ off at a specific frame. I guess this is because its not placed beneath the start node, hence the exporter doesn't read this as being a valid shape. I guess the workaround would be to assign an IFL material that has an alpha that displayed the mesh at a specific frame and use that as the texture for the skinned mesh ( assuming that IFLs with skinned meshes works ).
#13
don't know if either will work as I have never tried either. They may work, they may not. If you are having issues with either working, let me know and I will see if it is an exporter side thing. I would imagine that if it does not work, it is because we never really anticipated anyone amimating the visibility of a mesh that was deformable.
The IFL animatiom with a skinned mesh would be a good one to test as well.
If you can give me an idea of what you are trying to accomplish, I might be able to offer some suggestions of how I would approach it.
05/22/2006 (8:08 pm)
@John, don't know if either will work as I have never tried either. They may work, they may not. If you are having issues with either working, let me know and I will see if it is an exporter side thing. I would imagine that if it does not work, it is because we never really anticipated anyone amimating the visibility of a mesh that was deformable.
The IFL animatiom with a skinned mesh would be a good one to test as well.
If you can give me an idea of what you are trying to accomplish, I might be able to offer some suggestions of how I would approach it.
Torque Owner John Cooney