Any Solution For Droopy SIK Yet?

[BESTrin]

try selecting sik tag on the hiest bone in the hierarchy and turny off dynamics. Then you can edit the bones position with bone tool or what ever you like without having saggy bones. You might also try increasing drag and lowering gravity next time so the bones dont just drop. SIK is a mixed blessing in my mind. Its great idea and very usefull but it also causes alot of headaches. Good luck.

[LucentDreams]

check the computer arts projects may issue, the file that comes in that issue has Droopy rabbit ears that work great. takes mere seconds to setup the SIK part. THe tutorial itself doesn't cover too wel what was done for that, but the file makes it pretty easy to look at the settings. Key thing for droopyness is to delete the goal made when using the setup SIK command.

[BESTrin]

I think his poblem though is to droopy not cant do droopy. If he had the target it seems to me he would be in buisines for animating a tail. Just my thoughts.

[sanciok]

actually i made a rig in the 8.2 where the tails was controlled in SIK. To control the bending of it I had added an expression with a slider driving the P rotation channel in the SIK tag. In the 8.2 versions it was working perfectly; now, in the 8.5 it doesn't work. In fact, if i go over than 8 degrees on the control, the tail drops down, deadlike.

I think too that there's something strange between the MOCCA and the 8.5

Cheers,

Sanciok :bounce:

[abstrax]

Hi folks,

unfortunately I can't reproduce your problems. Did you take a look at the tutorial scene) You can find it in ...\Cinema_4D_R8\Tutorials\Mocca\Scenes\Bones\Tail Final.c4d and it works perfectly here.

So can you reproduce your problems on that rig? If yes, it would be geat, if you would post it here.

Cheers,
Marcus

[bobzilla]

Well, yes, the problem is I DON'T want droopiness. And, yes, this is a bug. I think I found a work around though. If I rotate the anchor bone until the tail straightens out, it will work. The bones will have to be reset, and fixed again, then reset the Chain Rest Rotation.

Try doing the dino tail tut from scratch in 8.5 and you should see the problem.

[abstrax]

Hi Bobzilla,

ok I did the dino tail tut on R8.2 and on R8.5. And there's no difference in the behaviour. After deleting the goal null object and activating dynamics in the anchor tag, the dino tail is hanging down like a rope. Then you increase the strength of the rest rotation to 30% and the tail is hanging down more like a tail than a rope.

If you don't want the dino tail to hang down you can either increase the rotational rest strength of all soft ik tags to about 70% or you set the dynamics speed to 0% or you set the dynamics gravity to 0.

So you have plenty of possibilities, but I don't see a bug here. So perhaps you can send me/us a example file for further investigation.

Cheers,
Marcus

[bobzilla]

I've attached an example made in a previous version of C4D. It's been set up to work fine (as expected, I should say). Now, move the Peveis around and the tail will droop, and none of the SIK settings will get it to go back up.

If you don't get the same problem, please let me know. I understand what everyone is saying about the settings. I know how they work. I've gotten them to work before, but since I've been in 8.5 I had this problem.

[flyingP]

try this

reset your bones, turn off SIK, select the tail bone, in the bone tool, click on "add update null" give that null then an SIK anchor and try moving the Pelvis again. :D

EDIT: Oh and don't forget to reset the bones after adding the null :wise:

[flyingP]

.............:)

[bobzilla]

Yeah, I got some weird results. Actually, the easiest way to fix it seems to be to rotate the first bone in the chain (the one with the anchor) 1 degree on the Z (B) axis.

[flyingP]

yeah, just wait a bit, I just found out it craps out in the file that I uploaded when the pelvis gets rotated......there seems to be something a bit strange with that soft IK chain, I'll get back in a few minutes my machine just crashed :banghead:

[wesware]

Hey,

I hope I am examining the correct issue here, but I created a tail with the Bone tool and "add child bone" button. Ceated a soft IK chain with the soft IK button. Deleted the goal object... increased the rotation strength and got a very nice rigid tail that whips when I animate it. No problems here in 8.5

Am I in the right ballpark here?

[flyingP]

Hey,

I hope I am examining the correct issue here, but I created a tail with the Bone tool and "add child bone" button. Ceated a soft IK chain with the soft IK button. Deleted the goal object... increased the rotation strength and got a very nice rigid tail that whips when I animate it. No problems here in 8.5

Am I in the right ballpark here?

The problem here Wes seems to be in the mix of the 2 IK systems in this case SIK and an IK expression.....just out of curiosity bobzilla why are you wanting to use both systems here ??

[abstrax]

Hi Bobzilla,

thanks for your file. Now I see the problem. Until now I don't know the root cause for this behaviour (it looks as if the chain is too long/too heavy and breaks) so I have to investigate. When I know the reason I will reply again.

Cheers,
Marcus

[flyingP]

Yeah, I got some weird results. Actually, the easiest way to fix it seems to be to rotate the first bone in the chain (the one with the anchor) 1 degree on the Z (B) axis.


Wierd, seems rotating does fix it, with exception of Pitch to start with (at least but me) :surprised

[abstrax]

Hi Bobzilla,

it smells like a bug. It seems that the problem only occurs, when the anchor bone has small bank or heading angles (<5°, depending on the chain length). That is the reason why it doesn't appear in the dino tail tutorial, because there you have to rotate the bone to heading=270°. Looks like some kind of gimbal lock happens in the IK solver. I will forward the problem to the developers and we'll see.

I have attached the changed rig. I added a null bone so that the anchor bone can have a 45° bank angle without changing its general direction.

Cheers,
Marcus

[LucentDreams]

Yes seems almost impossible to prevent the SIK chain from breaking when its this long. This is a curious one.

[bobzilla]

Thanks for investigating and spending the time on this, gang.

It figures. The only thing I use SIK on in my rig, and it's broke! Sheesh!

Just rotating the bone 1 degree seemed to fix it for now. 'tis strange, though...

I thought maybe the gravity was stronger here in New Jersey :eek:

[Cactus Dan]

Howdy,

Here's something else weird: if you rotate the bank angle slightly as suggested, it works, but then if you also rotate the pitch angle so that the tail is pointing straight up (with animation on), it starts thrashing about wildly. It's a cool looking effect, but definitely seems like some kind of bug.

Adios,
Cactus Dan

[LucentDreams]

gimble *cough* lock *cough*

[abstrax]

Howdy,

Here's something else weird: if you rotate the bank angle slightly as suggested, it works, but then if you also rotate the pitch angle so that the tail is pointing straight up (with animation on), it starts thrashing about wildly. It's a cool looking effect, but definitely seems like some kind of bug.

Adios,
Cactus Dan
Hi Cactus Dan,

I don't think that is a bug, because soft IK is a dynamic system with masses and energy. And by rotating the anchor you put energy into the system, that has to be decreased again, if it shouldn't jitter too much. The problem here is that for speed reasons only a very simplified model stands behind soft IK that due resonance increases the internal energy when put to its extremes. So you either have to set the rest rotation strength to ~70% or you dampen the system with damping>=60%.

Cheers,
Marcus

[wesware]

OK,

Thanks for setting me straight FlyingP.
Got the file and saw what was happening.
Thanks for the insight as well, abstrax.

[flyingP]

Thanks for setting me straight FlyingP.

Nahhhh to be honest I still find it a bit confusing :blush: I've played around quite a bit with soft IK chains but haven't actually run into this problem myself until now :shrug:

[bobzilla]

Well, I think the original problem is definitely a bug in 8.5. So rather than wracking our brains (which I already did for quite a while) hopefully it will be fixed by Maxon. I had e-mailed tech support, but they didn't say it was a bug or say anything about fixing it.

We'll see...

[abstrax]

Well, I think the original problem is definitely a bug in 8.5. So rather than wracking our brains (which I already did for quite a while) hopefully it will be fixed by Maxon. I had e-mailed tech support, but they didn't say it was a bug or say anything about fixing it.

We'll see...
Hi Bobzilla,

as I told you before, I will ask the developer, why soft IK behaves the way it does. When I know something new, I will tell you. But even if it's a bug, it is not sure that there is an easy solution. So it might take a while. But thanks for your hint.

Cheers,
Marcus

[CGTalk Moderation]

This thread has been automatically closed as it remained inactive for 12 months. If you wish to continue the discussion, please create a new thread in the appropriate forum.