When running CMC, the toes on my model become extremely mangled. They appear ok all the way through RRA. I've tried locking the subtalar and mtp joints, but this causes an unexpected shutdown. When I don't lock them, CMC finishes, but with mangled bones. This is the only property I am varying between working/not working runs.
Any ideas why this causes a crash?
The log file starts out like this:
#
# An unexpected error has been detected by HotSpot Virtual Machine:
#
# Internal Error (0xe06d7363), pid=3148, tid=3564
#
# Java VM: Java HotSpot(TM) Client VM (1.5.0_06-b05 mixed mode)
# Problematic frame:
# C [kernel32.dll+0x1aafc]
#
--------------- T H R E A D ---------------
Current thread (0x0ad24310): JavaThread "Thread-9" [_thread_in_native, id=3564]
siginfo: ExceptionCode=0xe06d7363, ExceptionInformation=0x19930520 0x0cacc020 0x0ba00a0c
Registers:
EAX=0x0caca930, EBX=0x7c3ac1d1, ECX=0x00000000, EDX=0x0cacc020
ESP=0x0caca92c, EBP=0x0caca980, ESI=0x0caca9c0, EDI=0x0caca9c0
EIP=0x77e7aafc, EFLAGS=0x00000206
CMC/Locking Joints
-
Kathleen Reilly
- Posts: 30
- Joined: Fri Jan 11, 2008 6:25 am
RE: CMC/Locking Joints
By the way, locking subtalar and mtp after RRA for CMC doesn't actually seem to do anything-CMC gets through ~60% before shutting down, and I can see that the toes are still mangled.
-
Sam Hamner
- Posts: 58
- Joined: Tue Jun 26, 2007 5:03 pm
RE: CMC/Locking Joints
Hi Kathleen,
If you want to lock the joints, I think that it's best if you go back and re-do IK, then do RRA and CMC with the new IK solution. If not, you will be giving these analyses data for coordinates (i.e., joints) that no longer exist. CMC uses a "hard constraint" in it's optimizer to track the joint coordinates that you calculated in RRA.
Additionally, you are going to have to change the "Actuators" XML files for RRA and CMC, as well as all the "Task" XML files for IK, RRA, and CMC (as well as Scale if you re-do that) and remove everything associated with the joints that you lock (i.e., subtalar and mtp).
Hope this helps. Let me know if it works out.
Cheers,
Sam
If you want to lock the joints, I think that it's best if you go back and re-do IK, then do RRA and CMC with the new IK solution. If not, you will be giving these analyses data for coordinates (i.e., joints) that no longer exist. CMC uses a "hard constraint" in it's optimizer to track the joint coordinates that you calculated in RRA.
Additionally, you are going to have to change the "Actuators" XML files for RRA and CMC, as well as all the "Task" XML files for IK, RRA, and CMC (as well as Scale if you re-do that) and remove everything associated with the joints that you lock (i.e., subtalar and mtp).
Hope this helps. Let me know if it works out.
Cheers,
Sam
-
Kathleen Reilly
- Posts: 30
- Joined: Fri Jan 11, 2008 6:25 am
RE: CMC/Locking Joints
Hi Sam-
Thanks, I locked mtp and that joint no longer becomes distorted.
Do you have any insight in to why this may have happened in the first place? Like I said, it looks ok all the way through RRA, and then mtp becomes distorted during CMC. I don't want to lock this joint if I don't have to, of course, so if you have any ideas, I'd appreciate the help.
Thanks!
Thanks, I locked mtp and that joint no longer becomes distorted.
Do you have any insight in to why this may have happened in the first place? Like I said, it looks ok all the way through RRA, and then mtp becomes distorted during CMC. I don't want to lock this joint if I don't have to, of course, so if you have any ideas, I'd appreciate the help.
Thanks!