PDA

View Full Version : Nevron Genoma Rig in Layout is creating a Gimble Lock on Ankles - help



snsmoore
11-12-2013, 02:35 PM
Anyone know how to prevent (or correct) gimble lock on the ankle control when you set up a Nevron-Genoma motion capture rig in Lightwave. In Layout, I want to rotate my character's feet downward to match the kinect skeleton's feet orientation, but I can't because the pitch (I think) is rotated 90 degrees on the Genoma control.

I'll upload a screenshot later, but hopefully the question is straightforward enough...

snsmoore
11-12-2013, 11:37 PM
Here's the screenshot. The heading and bank of the control are practically on the same axis.

118159

Not sure why Genoma would set up a rig where the control is Gimble locked, unless there's something I'm missing in modeler (like rotating the Genoma "skelegons")

erikals
11-13-2013, 05:11 AM
even though Genoma is great, there are pitfalls one must be aware of (it being Genoma 1.0)

Rebel Hill covers some of them here >


http://www.youtube.com/watch?v=RlFlbU9I6Fs

RebelHill
11-13-2013, 06:25 PM
Yeah... orientations are a complete mess in genoma.

if its just base pose you're lining up, switch to local, or world axis mode and rotate on those handles... it *should* be ok. Do note that whilst you can get away with such in situations like these, you wont be able to most of the time whilst animating. Check out my video on rotations for more details of what works, why, etc.

erikals
11-13-2013, 07:27 PM
hi, did you cover this specific gimbal problem in this series? >
http://rebelhill.net/html/rhr

RebelHill
11-13-2013, 08:46 PM
ofc...


http://www.youtube.com/watch?v=3BeDUqjD1Oc&list=PLTds3QePYrWEWipwKkLmyNT4Tf_JTigM2&index=5

Though its not really a "specific" gimbal problem, its the self same gimbal problem you get in any euler system in any 3d app. The fact that genoma rigs are contain such problem controllers is simply down to poor planning and design, they're just bad rigs. In the image, it looks like the item selected is the handle null which controls the bone directly (whicll be done via SAI, probably in world mode)... there's no way to fix that directly as any alteration to the pivot of the controller would screw up the constraint... If a "controllable" solution were wanted, then you can patch it (sort of) by inserting an intermediate item as the problem items parent (pre-oriented as desired).

Couldnt tell you if that'd mess up the functionality of nevron itself though... it may be transparent to it, or it may error out because of it.

snsmoore
11-14-2013, 02:13 PM
Yeah... orientations are a complete mess in genoma.

if its just base pose you're lining up, switch to local, or world axis mode and rotate on those handles... it *should* be ok. Do note that whilst you can get away with such in situations like these, you wont be able to most of the time whilst animating. Check out my video on rotations for more details of what works, why, etc.

Cool, thanks RebelHill, I'll try local/world axis modes to see if they help.

snsmoore
11-16-2013, 06:46 PM
yep, using world axis mode enables me to correct the feet. It doesn't seem to present any problems with the motion capture. I still need to get it to be a bit more accurate, but this at least solves the foot problem.

Ryan Roye
11-18-2013, 05:13 PM
yep, using world axis mode enables me to correct the feet. It doesn't seem to present any problems with the motion capture. I still need to get it to be a bit more accurate, but this at least solves the foot problem.

Try this out and see if it works:

IKBooster + Quaternion mode on the problem item + FK control.

The rotation will happen relative to your viewport orientation.

erikals
11-18-2013, 07:52 PM
kinda related,..
any of the following IKB tutorials of yours that covers how to set up a rig in Genoma for then to alter it with IKB ?

snsmoore
11-18-2013, 09:52 PM
Try this out and see if it works:

IKBooster + Quaternion mode on the problem item + FK control.

The rotation will happen relative to your viewport orientation.

I tried it out and I could apply Quaternion and FK, but when adjusting the HPB controls(values), I couldn't resolve the gimble locking - they did not act relative to the viewport. This of course is on the control not the bone (the bone looks to have expressions on it)

I can adjust the control using World axes as a workaround though.

- - - Updated - - -



kinda related,..
any of the following IKB tutorials of yours that covers how to set up a rig in Genoma for then to alter it with IKB ?

I've had better luck using pure skelegons and then converting them to bones in Layout. Cleaner result that works well with IKB for me at least.

Ryan Roye
11-19-2013, 05:00 AM
kinda related,..
any of the following IKB tutorials of yours that covers how to set up a rig in Genoma for then to alter it with IKB ?

Not the commercial ones, but I'd be willing to put out a free one as there are some gotchas to using Genoma in general that people really should know about (IE: Demonstrating how to preserve a facial rig setup between updates, or how to prevent Genoma from destroying your IKB setup among other things).


I tried it out and I could apply Quaternion and FK, but when adjusting the HPB controls(values), I couldn't resolve the gimble locking - they did not act relative to the viewport

Yeah, the UI value controllers themselves won't fix the gimal issue, the relative viewport rotation happens only when click-dragging the selected item.

RebelHill
11-19-2013, 03:29 PM
Quaternion and FK, but when adjusting the HPB controls(values), I couldn't resolve the gimble locking

Quaternion rotation in IKB is no different to applying the quaternion booster motion modifier (infact its exactly the same since when you select quat in IKB, quat booster gets added to your item atomatically)... This motion modifier doesnt give you a TRUE quaternion (as LW doesnt support quaternion co-ordinates for rotation, only euler)... what it does instead is to "flip" channels when you rotate "cross axis" (using either world or local manipulation, or IKBs screen space manipulator, or quattro) to prevent the "euler drift" that occurs between keyframes otehrwise.

But for correcting an initial pose as you are (not animating between positions) its ineffectual.

Ryan Roye
11-19-2013, 04:34 PM
https://www.youtube.com/watch?v=ujWL6QN_BkU&feature=youtu.be

snsmoore
11-19-2013, 08:13 PM
Quaternion mode indeed works as well. I must have had some user error the 1st time I tried it. Thx, chazriker and RebelHill!

erikals
11-19-2013, 09:40 PM
that's great! http://erikalstad.com/backup/misc.php_files/smile.gif

Ryan Roye
11-20-2013, 05:35 AM
Let's take this one step further for the sake of completeness... just to give you an idea of possible solutions and alternatives:


https://www.youtube.com/watch?v=N6kx5GNdliI&feature=youtu.be