PDA

View Full Version : This must be a JOKE..



hypriit
02-03-2009, 11:56 PM
Make Joint chain just don't work right in 9.6. Goals Don't hold foot still.I need to go back to 9.5 to fix these problems where all these features are working correctly. Why mess up already working features in next release ? How is that every next release is getting buggier ?
Why put new features inside when they are not working correctly ?

Some people need to work with that app....Im looking for some stability and fixed bugs in next release but it seems its getting worse every time.

When the release is in beta version .. please let it be there until its fixed.

Its more like a toy not working tool
frustrated LW user

Frank_Geppert
02-04-2009, 01:52 AM
Did you create a bug report? And don't forget to send your scene to NT.

I experienced that they react very fast but don't read every forum post.

hypriit
02-04-2009, 02:39 AM
What i m saying is that i m not in beta test team to search bugs and report i dont have time for that and wait that mait be in next version they will fix it.And there is big chance that in next release i need to do it again with same bug because its back again.Even beta version is more stable than final.Im not talking about some minor bugs.In version 9.5 fiberFX was one big BUG Lw mostly crashed when you used that.
All the time you need to search ways around to avoid bugs.
When i bought that soft i thougt its fully functioning and i can work with all the features in it.I didnt expect that some of them even don't need to work and are in beta phase.

They should make that beta phase much more longer. And give fully functioning product to customer.
Maybe they should remove some of the new fatures and put more effort to make lightwave more stable.

adamredwoods
02-04-2009, 12:33 PM
Make Joint chain just don't work right in 9.6. Goals Don't hold foot still.

Make joint chain is particular. I generally dont use it, and just create my own.

Goals aren't guaranteed to hold the foot still, depends on the situation.

If you have screenshots or even the scene file, we can help you out.
Rigging is challenging on most 3d apps.

RonB
02-04-2009, 12:46 PM
As far as I am concerned LW 8.5 was the last workable version...I have stayed with that...

jm gruninger
02-10-2009, 03:55 PM
Hi,
I had the same sort of problem , I had a goal for the ankle and a goal for the knee. The ik engine in 9.5 resolved everything perfectly, but now in 9.6 the ankle was no longer fixed by its goal , no matter how high the goal strength. The fix was to get rid of the knee goal and eveything worked great, apparently more than 1 goal in a ik chain confuses the issue since lw doesn't know which goal to resolve.
Marc

jaxtone
02-11-2009, 05:09 AM
Make Joint chain just don't work right in 9.6. Goals Don't hold foot still.I need to go back to 9.5 to fix these problems where all these features are working correctly. Why mess up already working features in next release ? How is that every next release is getting buggier ?
Why put new features inside when they are not working correctly ?

Some people need to work with that app....Im looking for some stability and fixed bugs in next release but it seems its getting worse every time.

When the release is in beta version .. please let it be there until its fixed.

Its more like a toy not working tool
frustrated LW user

I haven´t had time to play around with the CA part of LW9.6 and maybe that´s not a bad solution while it sounds like problems like this occure! Have you ever tried the Maestro CA plugin from Stillwater Pictures. Eric at Stillwater Pictures told me a while ago that his latest update of Maestro had fixed problems like this!

Another suggestion is that NT with the purchase of the License for 9.6 also attach a package of "nails and a big hammer" to fix the foot problem of their undeveloped part of Lightwave´s Character Animation feature! :)

adamredwoods
02-11-2009, 11:17 AM
Hi,
I had the same sort of problem , I had a goal for the ankle and a goal for the knee. The ik engine in 9.5 resolved everything perfectly, but now in 9.6 the ankle was no longer fixed by its goal , no matter how high the goal strength. The fix was to get rid of the knee goal and eveything worked great, apparently more than 1 goal in a ik chain confuses the issue since lw doesn't know which goal to resolve.
Marc

DId you try different goal strengths? I haven't had time to test multiple goals lately, but I do remember having problems.

ericsmith
02-11-2009, 10:16 PM
I haven´t had time to play around with the CA part of LW9.6 and maybe that´s not a bad solution while it sounds like problems like this occure! Have you ever tried the Maestro CA plugin from Stillwater Pictures. Eric at Stillwater Pictures told me a while ago that his latest update of Maestro had fixed problems like this!

I don't recall discussing this with you, but I'm afraid you may have misunderstood something I said.

The sad truth is that in my opinion, 9.5 and 9.6 are both big dissapointments in the rigging department. Sure, there are some new features, but the IK solver is much worse than it was in 9.0, and in general, scenes in 9.5/9.6 with rigged characters get about half the FPS interaction speed as in 9.0.

As much as I hate to say this, my recommendation is to do your rigging and animating in 9.0, and then import MDD scanned objects into 9.6 if you want to take advantage of the new features. Of course this would exclude and of the new rigging features, but I would rather work with less tools that work right rather than more tools that don't.

Eric

JBT27
02-12-2009, 02:46 AM
Just to be completely precise about this - do you mean the actual 9.0 release, or any 9.x prior to 9.5/9.6?

I ask because someone here is getting into rigging, but going in at 9.6.....which is starting to sound like a bad idea, from this thread.

We do have a license of Maestro as well.....

We tend to keep 9.3.1 and 9.6 on the boxes, but nothing earlier.

Julian.

Jim M
02-12-2009, 05:15 AM
No problems with 9.6 on the rigging front here

ericsmith
02-12-2009, 07:37 AM
Just to be completely precise about this - do you mean the actual 9.0 release, or any 9.x prior to 9.5/9.6?

I meant 9.0.

9.2 and 9.3 had significant interactivity problems with Maestro, which were fixed by 9.5. but 9.5 was when the new IK solver was implemented, which has a tendancy for the end of the chain to slip away from the goal even when the bones are not fully extended.

But if you're not using Maestro to animate with, 9.3 will probably not present any major issues.

Eric

adamredwoods
02-12-2009, 11:24 AM
I meant 9.0.

9.2 and 9.3 had significant interactivity problems with Maestro, which were fixed by 9.5. but 9.5 was when the new IK solver was implemented, which has a tendancy for the end of the chain to slip away from the goal even when the bones are not fully extended.


Does this problem happen only with Maestro, or IK rigs in general?

jasonwestmas
02-12-2009, 04:12 PM
To those having problems with IK goals, are you just trying to do a simple walk cycle or something? It just sounds odd in my experience. If we were talking about LW 7 yeah then I could understand the issues with IK.

ericsmith
02-13-2009, 09:17 AM
Does this problem happen only with Maestro, or IK rigs in general?

The rigs that Maestro creates use nothing but standard LW rigging tools, so the problem can only be with LW's IK solver.

Eric

DeepThroat
02-13-2009, 03:48 PM
There must be something wrong here.

LW IK works great in LW9.6....better than ever.

Take a look:

http://www.youtube.com/watch?v=zGWeCrU6RNo

http://www.youtube.com/watch?v=r72qocWVfEg

Both characters have been rigged in LW9.6. I had no problems doing that.

IK and Vectors are now absolutely reliable.

The Dragon is part of LW9.6 content...so anybody could load the scene and take a look.

I never use Make Joint Chain or Make Bone Chain while rigging.

I always build the rig structure from scratch or starting from an IK Chain already created by me (now is very simple to transfer parts of a rig to another character since you can easily transfer Bones from an object to another using parenting).

Please if you can show us your rig, I'm sure we can solve any problem.

(and I really have to quickly complete my 9.6 rigging tutorial....had to much work to do lately...mainly characters!)

ericsmith
02-13-2009, 04:53 PM
Here's a video showing the drift in action:

http://www.stillwaterpictures.com/IKdrift.avi

Here's the deal. Newtek tried to "soften" the IK solving so you wouldn't get pops when the IK chain reaches it's extent.

Personally, I think this was a really bad solution to a problem that is much easier solved by setting up the bones properly. The whole point of IK is to keep the end of the chain constrained to the goal. If you can't trust that the end of the chain will be FIXED to the goal, then the rig could very possibly be unuseable.

I know that the end of the chain will pull away if the bone chain straightens completly out, and there can appear to be a "snap" as you approach the extent, but that's a physical reality, not something that can or should be circumvented by software.

On the other hand, if you have any insights that might point to something I could do differently, I'm interested in hearing them.

Eric

DeepThroat
02-13-2009, 05:24 PM
Here's a video showing the drift in action:

http://www.stillwaterpictures.com/IKdrift.avi

Here's the deal. Newtek tried to "soften" the IK solving so you wouldn't get pops when the IK chain reaches it's extent.

LW9.6 ik works perfectly in my rigs.

SoftIK is an option you can simply turn off. It works the way it works....if you don't like it you can really just ignore it (and there's no need of using it....and if you want you have to use the new Same As Item options or a Constraint to evitate the sliding).

I repeat it again...9.6 IK is absolutely reliable. Take a look at the mov...no sliding at all (and we're talking about more complex legs....).

Rigging complex characters in LW9.6 has never been easier.

IK , IK Vectors are and Same as Item works very well and are absolutely reliable.

What I see in your video is something that doesn't happen in my rigs.

So maybe there's something we can talk about.

DeepThroat
02-13-2009, 05:38 PM
The rigs that Maestro creates use nothing but standard LW rigging tools, so the problem can only be with LW's IK solver.

Eric

....or in the way the Maestro rig is built.

DeepThroat
02-13-2009, 05:51 PM
I haven´t had time to play around with the CA part of LW9.6 and maybe that´s not a bad solution while it sounds like problems like this occure!

Well, I think you should take a closer look. False informations don't help users to work better while they really could.




Another suggestion is that NT with the purchase of the License for 9.6 also attach a package of "nails and a big hammer" to fix the foot problem of their undeveloped part of Lightwave´s Character Animation feature! :)

When I say "false informations"....well....that's what I'm talking about.

ericsmith
02-13-2009, 06:54 PM
SoftIK is an option you can simply turn off. It works the way it works....if you don't like it you can really just ignore it (and there's no need of using it....and if you want you have to use the new Same As Item options or a Constraint to evitate the sliding).

I repeat it again...9.6 IK is absolutely reliable. Take a look at the mov...no sliding at all (and we're talking about more complex legs....).

I think you misunderstood what I said. I'm not using SoftIK in this rig. What I said was that the new IK solver implemented in 9.6 is dampened to alleviate the "pop" that happens when you extend an IK chain to it's length.

Looking at your movie, The foot does look to be fixed, but I'd like to see what happens when you pull the pelvis away to the point where the legs straighten out.

The thing is, my rig is a very standard, tried and true "reverse foot" method. It worked flawlessly before 9.6's new IK solver. So if it doesn't work now, I have a bit of trouble accepting that it's due to user error.

I did try the new "same as Item" instead of Match goal orientation, and didn't get any better results. Also, it's worth noting that I was able to lock the foot down a bit better by setting the goal strength to 100,000,000, but it still pulled away from the goal when I pulled the hips to a point where the legs started to straighten out. Note that I did not pull so far away that the chain couldn't reach the goal. I know that would pull the foot away from the goal.

Still, I would love to get ahold of your rig so I can see if there's any differences between the two that could explain the descrepancy.

Eric

geothefaust
02-14-2009, 12:24 AM
I did try the new "same as Item" instead of Match goal orientation, and didn't get any better results.

Hey, I know you've been doing this a while... but MGO compared to SAI is no match. MGO is really faulty. SAI works every time for me.

I've had no problems with it at all. Did you tick 'World' when you used SAI? I'm sure you know this, but, if it's not ticked when using bones, it wont function right, just like any other motion controller in LW.


Just my 2 cents.

DeepThroat
02-14-2009, 02:26 AM
I think you misunderstood what I said. I'm not using SoftIK in this rig. What I said was that the new IK solver implemented in 9.6 is dampened to alleviate the "pop" that happens when you extend an IK chain to it's length.

Yes, now I better understand what do you mean.
Trying to completely straight out an IK chain tends to make the goal not follow its target. But that's a normal behavior (try to do the same in Maya....you get the same problem).

But watching your video I can clearly see the foot sliding even when you lower the root so the leg is bent.

So there must be something strange in the rig.




Looking at your movie, The foot does look to be fixed, but I'd like to see what happens when you pull the pelvis away to the point where the legs straighten out.

If I do that, the feet stay fixed. Using Same As Item option makes this very easy and reliable.




The thing is, my rig is a very standard, tried and true "reverse foot" method. It worked flawlessly before 9.6's new IK solver. So if it doesn't work now, I have a bit of trouble accepting that it's due to user error.


Maybe the new 9.6 IK solver may work a little differently...but it works fine and accordly to the industry standards.

Maybe there's something in your rig that broke since now we can finally do things the right way (we have a real pole vector and working constraints in 9.6).

I think the Maestro rig structure should be revised and adapted to the new IK tools.

Attached you can find a tipical IK leg rig.

I'd like to take a closer look at your rig, I hope I can help solving any problem with it.

dballesg
02-14-2009, 03:06 AM
Hi

I am happy I found this thread! :)

I use quite a lot Maestro and its rig, because it simplifies the trial and error of rigging. In fact Eric's rig is very good and easy to modify once you take your time to dissect it.

I've been looking the new 9.6 tools, and I think they can simplify Eric's rig as well. How? Thats the problem for me, no idea! :)

On the new 9.6 content there is a demon rig made with joints, nice rig, but there is no explanation on how was done, i.e: starting from how you setup the joints, to the ik controllers.

I do not want hijack the thread but I would live to know a few things, so maybe I can simplify Maestro's rig and do a new one with the new tools. And Joints! :)

One thing I noticed on that rig DeepThroat has posted and is completely different in Maestro's rig. In Maestro you can move the foot goal as much like and the foot doesn't separate from the leg. On DeepThroat rig it does.

Question:

1.- Do you use the draw joint tools to setup the joints on Layout? Because Proton's video said that you can move the joint after draw it to adjust it. I tried it and then you broke the rotation on the joint, so you need to use the Bane Twist to align the rotation axis. I found this other thread talking about it:

Joint rotation messed up (http://www.newtek.com/forums/showthread.php?t=93958)

Skelegons would alleviate the problem, but then you have the "old" trial error way to adjust the joints (or bones) and repeat the process of add the IK controllers every time you change something on the skeleton.

So basically what it is the best way to create the skeleton for later apply the IK controllers and motion controllers?

If I solve that, I know how later modify the scripts that Eric uses on Maestro to automate the rigging process once you have a working rig.

2.- It is possible to do a reverse foot setup that avoids the foot separate from the leg when you move the goal. On Proton's videos the foot separates itself from the leg.

I am sure that more questions will arise. And sorry for such a long post! :D

David

DeepThroat
02-14-2009, 06:36 AM
One thing I noticed on that rig DeepThroat has posted and is completely different in Maestro's rig. In Maestro you can move the foot goal as much like and the foot doesn't separate from the leg. On DeepThroat rig it does.


You can easily get the same behavior of Maestro simply animating the weight of the Same As Item position item.



Question:

1.- Do you use the draw joint tools to setup the joints on Layout?

So basically what it is the best way to create the skeleton for later apply the IK controllers and motion controllers?

I always use to draw BONES directly in Layout when building my rigs.

It's always best to use bone because you can be always aware of their real orientation. Once you have built the rig, you can transform the Bones in Joints, if needed.

That's how I work, and for sure is not the only way.



2.- It is possible to do a reverse foot setup that avoids the foot separate from the leg when you move the goal. On Proton's videos the foot separates itself from the leg.[/QUOTE]

Sure it can be done!



I am sure that more questions will arise. And sorry for such a long post! :D

David

I'll start making a tutorial in about two weeks...I really hope to cover all aspects of LW9.6 rigging tools.

dballesg
02-14-2009, 06:39 AM
I'll start making a tutorial in about two weeks...I really hope to cover all aspects of LW9.6 rigging tools.

Thanks for all your comments! :)

And I am going to look forward to that tutorial more that the CORE announcement! :D

David

allabulle
02-14-2009, 07:54 AM
In a while I'll need to take care of the rigging process myself so I am too looking forward for the DeepThroat tutorial.

Thank you in advance!

ericsmith
02-14-2009, 05:48 PM
You can easily get the same behavior of Maestro simply animating the weight of the Same As Item position item.

http://www.stillwaterpictures.com/IKdrift2.avi

Sorry, but that won't accomplish anything good. The foot will swish as you straighten out the chain just like my rig does.

The fact is, the rig you posted simply exhibits a work-around to the problem we're discussing here. If you look closely, the goal bone in your leg setup doesn't stay fixed to the goal any better than my setup. You just keep the foot locked by seperating it from the chain via a positional constraint. I could easily do the same thing with my rig, but in my opinion, having the foot pull away from the leg bones is not an ideal solution when working with realistic characters, and while this may be nitpicky, the problem you're trying to cover up here could cause weird deformation in the ankle as well.

I haven't worked personally with Maya, but I seriously doubt it's IK solver is this sloppy. I know I have seen enough video tutorials that didn't require jumping through all these hoops to set up a basic leg rig to justify this opinion.

Honestly, if this new behavior is industry standard, then congrats to the old Lightwave team for writing the only accurate IK solver in the industry.

I hate to be so blunt about all this, but I think the point needs to be clear. The new IK solver does not perform as a constraint should. The end bone of an IK chain should stay perfectly fixed to the goal unless the root is moved far enough away that the bones can no longer reach the goal. THAT is industry standard.

On the other hand, you've immediately jumped to the conclusion that there's something "weird" about my rig, when from a technical point of view, the IK chain in the setup you posted does the exact same thing.

Eric

Castius
02-15-2009, 02:21 AM
First let me say i do know that LW 9.6 IK is slower. I'm guessing that it because it's taking into account more information. Like target items controllers and new options. And The IK has to be set up differently because it's a top down system instead of bottom up. I'm not sure that is limiting any rigs. But it does take a a different setup. Basically a lot more attention to Unaffected bu IK descendants.

Now with that fun out of the way. I looked at DeepThroat's rig. He is in need of some good prebend angles. When that is done the IK works as i would expect it to. I have no idea what is going on with your rig. Is there any chance i could take a look at just the leg setup. I know you have been rigging in LW for some time. So I'm just really curious what is causing that sloppy IK.

DeepThroat
02-15-2009, 03:27 AM
http://www.stillwaterpictures.com/IKdrift2.avi

Sorry, but that won't accomplish anything good. The foot will swish as you straighten out the chain just like my rig does.

Here again....that's a normal behavior.

And again, you should notice that in your rig the feet are ALWAYS sliding, not only when you straight out the chain.





The fact is, the rig you posted simply exhibits a work-around to the problem we're discussing here. If you look closely, the goal bone in your leg setup doesn't stay fixed to the goal any better than my setup.

Here again....my rig works as it should (read up).

Do you notice any slide in the leg when mine is bent? Don't think so.

Yours is always sliding. So saying that my rig has the same problems of yours is simply not true.




You just keep the foot locked by seperating it from the chain via a positional constraint. I could easily do the same thing with my rig, but in my opinion, having the foot pull away from the leg bones is not an ideal solution when working with realistic characters, and while this may be nitpicky, the problem you're trying to cover up here could cause weird deformation in the ankle as well.


It's quite a standard way of rigging a leg.

The constraint could be easily disable, and that's what I've done in my scene you find attached (and I set my preferred angle better....thank you Scott).





I haven't worked personally with Maya, but I seriously doubt it's IK solver is this sloppy. I know I have seen enough video tutorials that didn't require jumping through all these hoops to set up a basic leg rig to justify this opinion.


I've rigged in Maya for 3D features. You haven't worked with it. Well you should try so you can solve any doubt.





Honestly, if this new behavior is industry standard, then congrats to the old Lightwave team for writing the only accurate IK solver in the industry.


The only difference with the previous solver is that the new one introduces a sort of soft ik, and looks like it's always active.
But that's not what is causing problems to your rig. Too much sliding there, anyway you move.




I hate to be so blunt about all this, but I think the point needs to be clear. The new IK solver does not perform as a constraint should. The end bone of an IK chain should stay perfectly fixed to the goal unless the root is moved far enough away that the bones can no longer reach the goal. THAT is industry standard.


That is what you think...but it doesn't work that way. Could be more or less accentuated, but it's there.





On the other hand, you've immediately jumped to the conclusion that there's something "weird" about my rig, when from a technical point of view, the IK chain in the setup you posted does the exact same thing.

Eric

Don't think so. Test my rig again, and you'll see it working as it should.
I'm quite tired to repeat myself, but in your video is clear how the feet are always sliding.

So I ask for it again: could you post your rig so I can take a look a it?

DeepThroat
02-15-2009, 03:29 AM
Now with that fun out of the way. I looked at DeepThroat's rig. He is in need of some good prebend angles. When that is done the IK works as i would expect it to. I have no idea what is going on with your rig. Is there any chance i could take a look at just the leg setup. I know you have been rigging in LW for some time. So I'm just really curious what is causing that sloppy IK.

Thank you Scott, my bend angles were really too little!

mav3rick
02-15-2009, 04:01 AM
eric cmon show yourself .... :)

jaxtone
02-15-2009, 06:59 AM
Deep Throat!

Sorry for not replying until now but I´ve been busy with an animation case for a customer!

By the confidence you announce I will for sure look through your videos and check out your rig right on! I hope I will not be dissapointed as earlier since LW always seems to had ugly tradition of complicate routines in their CA! I admit that I more or less always have given up the struggle because the final result seemed to be too weak when it comes to rig and animate human or animal characters!

I am not picking a fight but to me you definitely sounds like a rare species when it comes to boundless hail and congatulate Lightwave´s CA! I have never heard any other user that have had 100% control over LW´s CA before, so this sounds very promising to me.

jaxtone
02-15-2009, 07:04 AM
Maybe I misunderstood you but here´s the mail correspondance anyway!

Sent from jaxtone to Eric: Monday, December 08, 2008 2:10 PM

1. I wonder what I can do to keep the characters feet fixed at a zero-Y level when he jumps, kneel or create any other move? As it is now the feet will crossover the ground level when he gets down from a jump or likely.

2. I also wonder whats the trick to keep any of his feet in locked positions when he is moving forward, backward or in any other position? As it is now the feet keep sliding and follow the body in its movements. Looks really wacky!


Received from Eric: Monday, December 08, 2008 10:21 PM

Okay, I can see what's going on now.

You need the latest version of Maestro. Go to http://www.stillwaterpictures.com/mtwudp and download version 2.35. This will solve two problems.

1. The reason your arms and legs weren't working properly in 9.5 is that fulltime IK was not turned on for the IK chains. This is because in 9.5, Newtek changed the default of this switch from off to on. And since the command is a toggle, the rigging process actually turns it off. The latest version of Maestro detects if you're running 9.5, and ignores the command if you are.

2. The interaction will also greatly improve, as Newtek created a fix in 9.5 that solved a problem introduced in 9.2 where dragging on the hotspot would prevent layout from updating until the mouse stopped moving. This fix from Newtek required us to add a new command to Maestro.


I don't recall discussing this with you, but I'm afraid you may have misunderstood something I said.

The sad truth is that in my opinion, 9.5 and 9.6 are both big dissapointments in the rigging department. Sure, there are some new features, but the IK solver is much worse than it was in 9.0, and in general, scenes in 9.5/9.6 with rigged characters get about half the FPS interaction speed as in 9.0.

As much as I hate to say this, my recommendation is to do your rigging and animating in 9.0, and then import MDD scanned objects into 9.6 if you want to take advantage of the new features. Of course this would exclude and of the new rigging features, but I would rather work with less tools that work right rather than more tools that don't.

Eric

jaxtone
02-15-2009, 07:07 AM
Two separate rivers can fill the lake!

ericsmith
02-15-2009, 10:56 AM
Here again....that's a normal behavior.

I guess that's a matter of opinion. To me, normal behavior is when the end of an IK chain doesn't drift from the goal.


And again, you should notice that in your rig the feet are ALWAYS sliding, not only when you straight out the chain.

Yes, but in the bigger picture, both rigs are exhibiting drift from the IK goal. Mine is just worse, possibly due to the fact that I have a more complex setup (ie. the extra bone in the knee).


Here again....my rig works as it should (read up).

Do you notice any slide in the leg when mine is bent? Don't think so.

Yours is always sliding. So saying that my rig has the same problems of yours is simply not true.

Here's the bottom line: If your position is that it's perfectly normal to have an IK chain drift from the goal even when the chain is bent (which was the original point of the thread), then there's not much point in arguing the point.

Now what I do find very interesting is that with the v2 setup you posted, with more pre-bending, the problem actually seems gone. So how much more pre-bending did you do? I tried doing a massive amount of prebending (like 60 degrees) and it made no difference.

Anyways, for anyone interested, I've attached my rig.


Maybe I misunderstood you but here´s the mail correspondance anyway!

Sent from jaxtone to Eric: Monday, December 08, 2008 2:10 PM

1. I wonder what I can do to keep the characters feet fixed at a zero-Y level when he jumps, kneel or create any other move? As it is now the feet will crossover the ground level when he gets down from a jump or likely.

2. I also wonder whats the trick to keep any of his feet in locked positions when he is moving forward, backward or in any other position? As it is now the feet keep sliding and follow the body in its movements. Looks really wacky!

Okay, I think I missunderstood your original question because I wasn't aware of the severity of this problem at that point in time. I thought that when you rigged the character, FullTimeIK was not turned on so you were basically getting FK limbs. I can see now that your question was directly related to what we're discussing now.

As a side note, you posted the private download link for Maestro users on a public forum. I would have appreciated it if you had not done that.

Eric

Castius
02-15-2009, 11:26 AM
Thank you very much for posting the rig.

Turn on Unaffected by IK Decendants on rfoot. Your ankle will snap right to the goal.

I hope this is a simple solution to your sliding. I'm not sure if it changes the behavior or your rig. But it does fix the sliding. This is because IK is calculated from the top of the hip first then the toes. In previous versions I was calculated from the toes first. I hope this helps.

ericsmith
02-15-2009, 12:03 PM
Hey Castius,

Thanks for the tip. That does make a big improvement.

When it was mentioned to check "unnaffected by IK of decendants", I was thinking upstream rather than downstream, considering the word "decendants".

There's still a bit of drift if you pull the root backwards as the legs straighten out (and this is not exactly a common thing to do), but overall, I would consider this close enough to be acceptable.

The good news is, I can incorporate this into the Maestro rig without any detrimental effect to those still working in a pre-9.6 version (like myself).

Eric

Castius
02-15-2009, 01:19 PM
You need more pre-bend on the knee bone. It's not flipping because you have good limits. But it's making it slide of it's goal. I would also put your shin bone Min limit to -25. To allow a little bit of hyper extend.

I'm glad it's not a larger problem with 9.6. But I do wish the speed was back to were it was. But i wouldn't trade it for the extra feature we have now. I see a bunch of rigs that are effected by the top down IK change. But i haven't seen one that couldn't be updated to work in 9.6. But there could be one. I just haven't seen one yet.

DeepThroat
02-15-2009, 03:07 PM
The good news is, I can incorporate this into the Maestro rig without any detrimental effect to those still working in a pre-9.6 version (like myself).

Eric

That's simply great.

ericsmith
02-15-2009, 03:42 PM
You need more pre-bend on the knee bone. It's not flipping because you have good limits. But it's making it slide of it's goal. I would also put your shin bone Min limit to -25. To allow a little bit of hyper extend.

Good points. And thanks once again for your help on this.

Eric

Castius
02-15-2009, 04:40 PM
My pleasure, there aren't many of us rigging in LW. We need to look out for each other. :thumbsup:

LW doesn't make these problems very apparent either. These types of transparent problems are what make learning LWs IK difficult.

dballesg
03-04-2009, 05:51 AM
I'll start making a tutorial in about two weeks...I really hope to cover all aspects of LW9.6 rigging tools.

Hi,

Any news on your rigging tutorial?

David