PDA

View Full Version : N8Girard Huntress Torso Maya 8.0


N8Girard
12-27-2006, 06:43 PM
This is my first attempt at one of the challenges. I was intrigued by the model and the feedback that people were submitting so I thought I'd post my own rigging solution on here. I'm working in Maya 8.0. I took the liberty of splitting out the armor pieces and shading them so that I could tell them apart from each other. I also shaded a few extra areas on the Huntress skin since it looks like it has been modeled for such things as boots and underwear. I may or may not work with the existing cloth mesh. It looks nice in T-pose but I'm sure it won't deform pretty in other poses, especially side to side stuff. The armor looks to be fairly well thought out but I'll find out it's intricasies when I go to rig it. I'm excited to try out some stuff with the armor instead of just parenting it to bones. We'll see how it goes. Good luck everyone! ;)

http://emanateimaging.com/images/Huntress_01_sm.jpg

N8Girard
12-27-2006, 08:31 PM
Ok, I have placed the bones now and used jsOrient to orient them all correctly. I tested all of the axis and made sure that they behave properly. The only ones here that might catch your eye as being off is the end effectors which don't matter. I'm doing the cross clavical setup because I generally have better deformations with that one. I may put more bones in simply for deformation purposes, but this is the set I'm primarily binding to and testing with and I'll add others as needed later. Here is the binding skeleton with local rotation axis visible.

http://emanateimaging.com/images/Huntress_02.jpg

I opted for a separated hip/back setup. Selected is where the lower bones start. Here is just the bones from a front and side view:

http://emanateimaging.com/images/Huntress_03.jpg

backstreet-boys
01-02-2007, 10:20 PM
wow
amazing
keep going man

stewartjones
01-03-2007, 12:30 AM
Hmmm, interesting! I've never used a cross-clavicle setup before. I think I'll have to give that a try sometime to see what I get with it. To be honest I've never even thought about doing it that way before, and it seems pretty clever!

Good stuff dude!

Cheers

-Stu

N8Girard
01-03-2007, 04:10 PM
The cross clavical setup is a setup that was taught to us at Vancouver Film School (VFS) by a modeling instructor. I had never used it before he taught us as well, but after seeing the results with just plain old weighted deformations, I've been using it ever since. See, the pivot point is better situated to give you a good shrug as well as pulling your arms forwards or backwards. In the backwards position, you usually have to use some kind of additional deformation to get it to look like a shoulder blade is there. I'm going to try using a pose deformer since I haven't used them yet and see what I can get. Right now though, the torso is fully weighted. I'll post a screen capture movie later on today or tomorrow to show the testing proceedure. Posted are stills of what the cross clavical setup looks like and the deformations I am getting out of it. Currently these have no additional deformers applied outside of a smoothBind.

Currently the pictures depict the following: Cross clavical setup from above, wireframe and shaded view with and without armor of the torso controls. Below that are the clavical range tests, up and down, forward and back. The area with a red glow is the problem area that I will be fixing with pose deformers in the back.

http://emanateimaging.com/images/Huntress_04.jpg

xenoid
01-04-2007, 05:04 AM
Nice idea N8Girard. Thanks for sharing! I'll definitely check this out. :)

N8Girard
01-04-2007, 05:11 PM
I know that we are not to that section of the challenge yet, but I would like to start discussing the automatic controls for the armor now so that I can prepare the setups in the future.

At first I attempted to place the pivot point for the shoulder armor group where the shoulder bone is. Then I parent constrained it to the shoulder bone. That works fine for moving the armor with the bone, but of course it collides with the body at certain points. So I attempted to add limits to the shoulder armor so that it would stop rotating at a certain point and the arm continues to move. That sortof works except for certain angles of the arm where it tries to solve the angle but in the wrong way. So limits won't work.

I sortof think that a pose deformer would work on it, except instead of actually modelling a change in the pose, just select all of the points and translate or rotate them one way or another so that it looks correct for the shoulder in that position. I haven't tried it yet but it makes sense.

Does anyone else have some cool automation ideas? Blend shapes? Set driven keys? Wrap deformers? Lattices?

Also, I am playing with the idea of using jiggle deformers applied to a null for each piece of armor. The armor is point constrained to the null. The jiggle deformer is set really low so that it behaves very slightly when moved.:bounce: This way, if the character jumps or lands or gets hit hard, all of the armor vibrates a little. Combine that with a collision object setup and the armor won't pierce the skin. It's just an idea, but the problem is, it requires dynamics which will slow down the animation process if it isn't turned off. Of course I'll have to figure out a way to do that so that it doesn't inhibit the animators and then when the animation is done, just turn it back on and have it solve. Of course, if I incorporate dynamics there, I'll have to use it elsewhere too...:eek:

N8Girard
01-04-2007, 07:31 PM
Ok, here is my first torso rig test video. http://www.youtube.com/watch?v=He-hPRDJ8ts

This test is without any pose deformers, wrap deformers, or deformers of any kind. Simply a smooth bind test with and without armor. I'm pretty happy with the deformations so far. The shoulders could still use some pose deformers I think, and I know that the shoulder armor is crashing. It's only point and orient constrained at this stage. Currently there are only FK controls that are being tested but I have the rig set up to incorporate IK next.

xenoid
01-04-2007, 11:32 PM
Really nice deformations there N8Girard! With some poseDeformer corrections and I think you'll be good to go. No crits from me for now, but think I'll be able to give more comments once I get hold of your scene file. :)

Haven't really thought of the armor part yet but just some ideas:
1. Idea would be similar to the ribbon spine. Construct a mesh(which will be hidden) similar to the part of the skin where the armor will be. Attach a joint to the surface using the method I laid out in my post. Parent the armor piece to it.
2. Transfer skin weights from the character to that poly mesh using Maya's Copy Skin Weights tool. Or you can even use wrap deformer for it to deform with the skin totally.
3. So the idea will be that the armor will follow how the skin will be deformed. Yet you can create an extra controller that controls the U and V values of the pointOnSurfaceInfo/pointOnMeshInfo node. So you can slide the armor around!

Just some ideas, but I think I'll try this first when we start the armor part. :bounce:

N8Girard
01-06-2007, 07:58 PM
Here is my second and probably final submission for the torso challenge. http://www.youtube.com/watch?v=eOsZY0ymK2I

As I've been discussing, I'm trying out Michael Comet's Pose Deformers on this rig. It is a free plugin and is available on his homepage: http://www.cometcartoons.com The idea behind it is that after skinning and weighting the object as best you can, you can sculpt corrective shapes for problem areas on the rig where collisions, pinching, and the deformations are problems. It gives you control over what bones are rotated to accomplish the corrected deformation, allows you to adjust the angles at which the deformation begins to affect the geometry, allows for twist, etc. All of this without affecting your skin weights. As an artist, it is a very visual way to tell what the skin will look like when the bone is at a certain position.

For my rig, I used 12 pose deformers on the shoulders (6 on each). I know that some people would classify the shoulders as part of the Arm Challenge portion, but in my experience they affect the torso way too much in deformations to be classified only as part of the arms. Most of the pinching and collisions from shoulder rotations are actually on the torso itself and therefore affect torso deformations.

The only things that I was unsure of incorporating into the torso/head rig was an IK/stretchy spine and jaw controller. I've weighted for a jaw bone but haven't actually put a controller on it because it was specified not to deal with face shapes. Since the jaw bone affects the shape of the face, I was unsure as to whether I would need it or not. I decided against putting an IK/stretchy spine into the rig because it is a HyperReal rigging challenge and it was not specified in the requirements. I have of course set up my rig so that I could incorporate one at any time but have chosen not to at this point.

I'm actually dealing with 4 skeletons. 1 Bind Skeleton, 1 Control Skeleton which the bind skeleton's bone's rotations are connected to in the connection editor, 1 FK Skeleton, which the control skeleton's bones are orient constrained to, and 1 IK Skeleton, which the control skeleton's bones are also orient constrained to. With this setup, I can blend between FK and IK on any part of my skeleton simply by blending between two orient constraints on the same bones on the control skeleton. Since the control skeleton has all of the constraints it makes the bind skeleton free of any constraints or animation curves so that it can be reset to bind pose relatively easy.

If anyone has any feedback regarding the deformations, please let me know now so that I might address them before the challenge is over. If I do not post any additional submissions, please treat this as my final submission for the Torso/Head part of the challenge. Good luck to everyone and thank you for all of your help.

darkjedi1929
01-09-2007, 07:16 AM
hmm....good going n8. The deformations look good, but I dare say people would be able to judge em better if you posted pose specific camera angles instead of one fixd camera. Eg., I am not able to see how the back deforms, also unable to see how the neck deforms....

cheers m8 ;)

N8Girard
01-09-2007, 06:41 PM
...The deformations look good, but I dare say people would be able to judge em better if you posted pose specific camera angles instead of one fixd camera. Eg., I am not able to see how the back deforms, also unable to see how the neck deforms....

Thanks darkjedi1929. I did as you requested for the arms and shoulders. I posted the results on my arms thread located here. (http://forums.cgsociety.org/showthread.php?t=449270)

If you need me to redo videos for the torso again, I'll do so after I incorporate boob jiggle. I'm still running tests with that so I'll post the results later.

N8Girard
01-09-2007, 10:46 PM
Ok, I decided to attack the boobs. (That's a figure of speach!) What I mean is, I'm attempting to incorporate some dynamics into the rig on the breasts so that they jiggle a bit when moved. Of course I'll have a switch to control when the dynamics are on or off so that the animators can hand animate it if they like (though not to the degree that the soft body can). I'm using a nurbs soft body with goal weights as an influence object for the skin. Below is a diagram of the goal weights that I painted onto the nurbs object.

http://emanateimaging.com/images/Huntress_05.jpg

Here is the video file of a test using this setup: http://www.youtube.com/watch?v=5arWKd_xoZA

After I apply it to the model I'll capture another test to show how it influences the surface points on the mesh.

JamSession
01-10-2007, 12:57 AM
Your boobs look really good :)

it is very nice motion but you want to be careful that they don't go through the bottom of the boob hoisters. Instead of adding collision objects, I would try only weighting the visible section. No one would ever notice if the bottom of the boobs aren't deforming :)

Overall very nice, I don't need 5 camera angles to see good deformation.

N8Girard
01-10-2007, 10:47 PM
Here are the results of my boob rig test. http://www.youtube.com/watch?v=diHx6MGWCV8

http://emanateimaging.com/images/Huntress_06.jpg

Above is an illustration of the rig concept. I used a nurbs sphere that was skinned to the breast bones and made a soft body out of it. Then I adjusted the weighting on it so that it behaved well enough for an unrestrained boob (You'll understand my usage of the term unrestrained in a second.) Then I took an idea that Xenoid gave me from his ribbon spine set up and used a pointOnSurfaceNode to constrain locators to the surface of the soft body nurbs sphere. I then placed a Normals constraint onto each locator that was on the surface so that it maintains orientation with the soft body sphere.

After getting the locators to stay on the surface and oriented correctly, I simply added them to the skin as influence objects and weighted the breast to them. The locators are necessary because influence objects only work off of transform nodes, not based on the shape of the geometry itself. If you wanted it to work off of the shape, you have to use a wrap deformer, which in my past experiences often break or evaluate incorrectly, especially if the wrapped object is accidentally punctured. To avoid the wrap deform, I used the locators as influence objects. I wanted to note however that I did attempt to use jiggle deformers in an experiment but that I couldn't get them to evaluate as nicely or as correctly as I could with the soft bodies.

With all this being said, I don't think I'll quite use this rig. It evaluates nicely for a boob that is unrestrained, i.e. doesn't have a bra on, but you can plainly see that with the armour on, the boobs are being supported and if they are being supported by leather or metal, they would not be flexible. So only the TOP portions of the boobs can have any jiggle at all and hopefully not enough to make them fall out. So I'm going to modify my rig so that only the top portions will move at all and even go so far as to remove the controls that I have on the boobs.

N8Girard
01-10-2007, 11:17 PM
Your boobs look really good :)

it is very nice motion but you want to be careful that they don't go through the bottom of the boob hoisters. Instead of adding collision objects, I would try only weighting the visible section. No one would ever notice if the bottom of the boobs aren't deforming :)


You were right JamSession. I'm only going to use the locator on the top of the geometry to influence the boob skin on top. Everything that is being supported by the chest armor wouldn't move at all.

After showing my rig to some females, they also agreed with you that bra's and corsets generally restrict the lower portions of the boob from moving. But they also said..."If ya got `em, flaunt `em!" Hahahaha..... evidently the designers of the armor thought so also.

proton
01-11-2007, 01:57 AM
Looking really cool...

N8Girard
01-16-2007, 08:41 PM
Here is my final torso submission:

http://www.youtube.com/watch?v=XUfjzSs9PX4 (http://www.youtube.com/watch?v=XUfjzSs9PX4)

Demonstrated in it are the new breast controls that enable the user to turn on and off the soft body breast jiggle and also allows them to adjust the firmness of the breasts. Also shown are closeups of all the torso deformations, and the FK/IK Blend on the head controls.

As you know, I've already rigged the arms so some of the arm controls are visible during this video. Of the arm controls, only the clavical and shoulder deformations are illustrated in the torso section because they directly control surfaces on the torso. I have NOT built in a posable IK spine because this is a hyper realistic rigging challenge and also because I don't want the animators having to work with translation keyframes as well as rotationary keyframes. I feel confident that any animator could grab this rig and operate it with little or no help from me. All of the keyable attributes are on the controls, all of which have been oriented with zero-out groups so that the animator can just type 0,0,0 and it will go back to default pose. Some of the controls do have limits applied to them, simply to keep the deformations from wigging out.

I do still need to add the pose deformers on the chest armor piece, but I'm saving that for the armor section.

I hope the judges and fellow riggers enjoy the video. Good luck to you all. :thumbsup:

xenoid
01-16-2007, 11:40 PM
Very nice N8Girard. You've almost finished the whole challenge! :scream:

Just some comments:
- At around 2:50 of the video, when the arms are twisted, there seem to be some irregular deformations on the clavicle area. Looks like some flipping from the video. Maybe you wana check it out?

- The clavicle deformations are nice on the shoulder regions. But wondering if it might be possible to have some slight movement on the breasts area when the clavicles/arms are lifted up? That might add more to the realism of it.

That's all for now. Man, I'm finding it hard to keep up with you. :thumbsup:

N8Girard
01-17-2007, 02:55 PM
Thanks Xenoid! I took a look at the video and I agree with you. The left shoulder raises up a bit too much when the arms are straight up in the air. You have a good eye for details! I can fix that problem by adjusting the pose Deformer for that pose so it shouldn't be a big deal to fix.

The breasts could move out of the way automatically when the arms move up but I'm afraid that the shoulders range of motion can be interpreted differently for different rotations so I can't implement a set driven key. I might be able to write an expression for it though. I decided in the end to leave the breast controllers on the rig though so that animators could have a bit of freedom with them. They could manually move the breasts out of the way of course, but you are right, the automation would probably be better in the end. I'm curious to see if other animators and riggers think it would be beneficial or not so additional comments are welcome on the subject. The only thing I'm worried about is if the expression would be too restrictive on the animators if they are hand keying the breasts around. I might be able to put an additional attribute on the controller for them to turn on and off the automation though, what do you think?

xenoid
01-18-2007, 12:39 AM
Nah I'm just picky with details. :D

I think it's always good that automation in rigs should definitely come with an attribute to turn the automation on/off. That provides much more flexibility for the animators. You do not want an animator to come up to you saying "I don't want your automated shoulder! Can't I just animate myself???".

Regarding the set-driven key for multi-axis drivers, you might want to check out using the poseReader, since you are already using the great poseDeformer plugin! I am planning to use that to drive all my autoshoulders and stuff like that for ball joints especially. I'll let you know once I have any good results with it. :thumbsup:

futurcraft
02-01-2007, 08:35 AM
Hey Guys...!

Awesome work you have here Girard...! I really liked the deformations you have and the extent of detailed research you have done for the boob rig... I was just wondering why you did not think of using Maya's dynamic Hair system for the boob rig.. Was it a matter of preference or did you feel that the soft body influence object provided you more control than what the hair system would have given you....?

Xen... I have another query I would like to ask you too... I've been told by some here at Gnomon that its better to avoid using plugins for a student demo reel.. This I guess is a very debatable topic, So was wondering whats your opinion on this issue... PSD certainly is a great system for use for muscle rigs, though it has one major minus point which is that its not quite supported by all the major game engines yet as it sits as a separate node in your scene file unlike using shapes and other generic maya features to achieve the same purpose of muscle deformations... and are quite compatible with most of the major game engines...

Would appreciate inputs from you guys...

n BTW, thanks girard for appreciating my work on the dragon... Though I've been more focused on the other toon rig which you would have seen at my site.. n I'm finally done with that & am presently lookin for an animator to animate it...

I'm actually workin on a Mocap rig with hair, cloth and fur simulation for a female character... I'll put it up for you guys to take a look once I am done with it in a week.. & feel free to rip it apart with your critiques....:D

Regards,

Nikhil.

N8Girard
02-01-2007, 07:07 PM
Nikhil,

I can give you some valuable advice regarding what to put on your demo reel. You see, I actually got an interview with ILM for a Creature Technical Director position so I know what questions they will ask you. For the record, I got the interview with my student TD Demo reel that is on my website.

ILM was looking for TDs that can do three things very well: simulations (cloth, fur, hair), rigging, and scripting tools for animators. They use Python and Mel to do a lot of their scripting and programming. What probably got me the interview was that I had all three on there. I had a small but effective muscle simulation that I used for Jessica Mouse's boob rig, I had of course her rig, and I also had a tool that I made for compositing as well as an interface that I made for animators to use. The interface I actually made from a Gnomon tutorial and customized it to work with my character. They wanted to see more cloth simulation, fur simulation and hair simulation on it because at any point a TD could be working extensively with those things. They also wanted to see a creature rig, not just a normal biped which there are thousands of tutorials out there about. You see, a creature rig challenges your problem solving abilities and forces you to think about not only how the creature will move but also how simple you can make the controls for an animator.

So if you want to be a TD, put those three things on your reel. They don't care if you use plugins, so long as you use them correctly. They liked to see the muscle system. They like to see cloth. But, that said, if you develop your own plugin (not like you'd have all that much time to do so as a student, but if you do...) DEFINATELY put that on the reel and show it off. Also, be consistant with your rigs and naming conventions. They may ask to see one of your rigs in software form so be prepared to hand one or two over to them so that their current TDs and animators can check it out.

In case you are wondering why I'm not currently working for them, my availability conflicted with their hiring schedule and also they had artists coming off of a project that they pushed into the new project to avoid training time. You see they value their artists there enough to try to keep them employed for as long as possible. It's still a contract based industry, but by restructuring your departments and keeping contracts cycled properly, you can keep your work force employed for longer without having to bring in fresh talent all the time and training them on your systems. I really respect that and appreciate that they do that for their artists. I'm still applying for work their and one of these days I'll hit the timing right but for now I'm working on developing those items that I listed above and trying to incorporate them into my demo reel.

Good luck with your student reel and if you have any further questions, don't hesitate to ask.

N8

futurcraft
02-02-2007, 03:44 PM
Hi Girard....

Thanks a lot for your valuable advice on my demo reel and for sharing your interview experience with ILM. I appreciate your value of ethics with regards to the job scenario in the industry. Such qualities are very important in the long run to be successful & have a good reputation no doubt...

Thank you for clarifying that part of my doubt on the use of 3rd party plugins. As far as writing a plugin myself, i learnt the basics of creating an empty plugin from Alias's "Writing Custom deformers" DVD by Eric Miller. Though I had to work on the other aspects of my reel as a result of which I postponed incorporating the plugin-in into the final reel. But now that you have mentioned it, I guess I will get back to that once I am done with these character rigs.

Hair, fur & cloth is exactly what i'm working on with a mocap driven female skeleton.. I am also working on an Ogre next after completing this present character. I plan to use a fat/muscle system for the ogre. But I have a few doubts, which I would like to ask you.

1. I had worked on a toony rig earlier, the one you would have seen on my website. The body & facial rig was developed by me and so were the GUIs for each of them. But the blendshapes for the face were modelled by the modeller.

Would it be a necessity for the rigger to model all the blendshapes for his demo reel? This doubt has been in my mind for quite a while so I thought since you adviced me on this subject I figured I should ask you....

2. What if not all the rigs in my reel are animated....? The mocap driven one would surely be animated and so will the Ogre by another friend of mine. But the toony rig may not be animated. Though I can always demo the rig's capabilities from a rigging standpoint. I'm actually on the look-out for an animator though.. but it aint tht easy for me finding an animator who would have the time to work on this. Atleast the guys I've been workin with are now completely tied with their projects & other work.

I'm doing the best I can from the rigging side though my main hurdle is the animation part.. So do you think this could in affect the final value of my reel in any way? How do you think I can couteract this situation?

Sometimes we have to draw a balance on how much content we can put into the reel inspite of the time constraint we have in hand. But still we do strive for the best we can...

Appreciate your advice on these issues...

Thanks,

Nikhil.

xenoid
02-03-2007, 12:58 AM
Sorry for the late reply. Frankly speaking, I'm not very familiar with the review process of the rigging reels, so strictly personal opinion and just my 2 cents here based on what I've seen from the good reels so far:

Plugins for student reel. If what you are trying to showcase is in fact achieved by the plugin itself, then I would think it's not good to use the plugin. For example, it will be good if you can showcase how you have achieved a certain level of good deformation using nothing but your own techniques. Then, to further enhance the realism, you can show how you use PSDs to do that.

For muscle system, most of the studios will be using their own proprietry software for that. So, I think it's good if you can show how you use the existing muscle systems to create a muscle rig, for e.g. Comet's muscle system. The key thing is, you are familiar with the muscle anatomy of the human/creature and can create a muscle rig out of the systems available. It's definitely a bonus if you are able to develop a muscle plugin of your own and put it in your reel.

Modeling for blendshapes. At my side, the shapes are actually modeled by the modelers, but I guess this differs from studio to studio. I do know some studios require the riggers to model the shapes as well.

As for animating the rigs, my personal feel would be to show the features of your rigs. However, showing animation on them would be able to give people a better idea of what range of animations will your rig be able to do.

One last thing, custom your reel to the studio you intending to apply to. If they are doing features, showcase your toony rigs. If they are doing creature visual effects, realism realism realism. :)

Hope this helps. And all the best for your reel production!

P.S. Sorry N8Girard! This thread's supposed to be for your rigging challenge. Nikhil, if you have further qns, might be good to pm us abt it?

futurcraft
02-03-2007, 01:05 AM
Thanks Xenoid.. & Girard... for your valuable advice.. n wish you guys all the best for your rigging challenge too...!!

N8Girard
02-05-2007, 10:47 PM
Hey Nikhil,

Xenoid is right on several points. Definitely customize the reel for the company that you are applying for. That's a given. If you are going for VFX, don't even bother putting in a toony rig, but if applying for a cg feature film company, the toony rig will be invaluable!

You can get away with not having an animation on a rig if you animate it like what Xenoid is saying to do by using it to illustrate different key elements that you put into the rig. Thing is, you want to keep this to a minimum because it takes up a lot of time on your reel and can be very boring for someone to look at for too long. Even just a 10 sec. animation piece can illustrate how efficient your rig is without boring your viewer.

As far as finding an animator is concerned, I have an opinion that I'm not sure that you want to hear but I feel that you need to hear it as someone wanting to go into rigging. I feel that every rigger should start off as an animator. If you don't understand animation and the problems that animators go through, you can't understand how to address the problems that they inherantly have with animation. You won't be able to develop tools that will make their lives easier if you haven't walked in their footsteps for a while. Does that make sense? I feel like every rigger should be able to get a performance out of their own rigs, otherwise why would you expect someone else to be able to when you can't? Now, this opinion of mine is not very popular among other riggers who feel that their specialty is the rigging and not in animation, otherwise they'd be animators. But I consider the ability to be a rigger to be a much more difficult challenge than that of animators because you have to understand all of the animator's problems to be able to address them effectively at a stage before animation has even begun. Don't get me wrong, animators have a skillset and a job that is very difficult and neccessary to do! They breath LIFE into these rigs! But you should be able to do that as well, even if only on a limited scale, so that you can relate well with the animators that you will always be working with. Besides, if you can show an animator a performance that they are happy with, they'll gain respect for you and what you do. Otherwise you'll always be just the fix-it man down the hall. I strongly encourage you to develop relationships with animators and I'm glad that you are already engaged in this process. There will be a constant dialogue between you and the animators which is desirable by employers.

I also recommend that your animators REFERENCE your rig within the files that the animation resides in. That way, there is a SEPARATION of the RIG and ANIMATION so that the animators don't have to redo all of their animation every time that you fix something on the rig. In order to do this though in practice you have to follow certain rules.

1.) DON'T DELETE ANYTHING IN YOUR RIG FILE! This will jack up the animation file which is searching for that thing that you deleted.
2.) DON'T CHANGE THE HEIRARCHY IN THE RIG FILE! Same reason as above only the heirarchy acts as a map for the animation file to find things.
3.) ALWAYS STORE A BACKUP OF THE LAST RIG THAT WORKED! I have a folder with the reference rig in it as well as a backup folder within that reference folder where I copy the last referenced rig that worked with the animation. I store all of the previous rigs that I gave to animators just in case.
4.) USE THE SAME NAME AND LOCATION AS THE OLD REFERENCED RIG! So it can reference the right file.
5.) ITERATE EACH RIG FILE AFTER YOU MAKE A MODIFICATION. Every time you make a modification to a rig, save it as a new file so in case that file doesn't work with the reference rig inside of the animation file, you can revert to the previous iteration and troubleshoot a new solution to the rigging problem at hand.

If you follow these criteria, you won't have to transfer animation between rigs every time you make an update to it. The software will automatically transfer the animation to the new rig that it is referencing. This also allows you to continue working while the animators are working and neither of you waste time. You can continue adding rig options while the animators are working with the basic rig. (i.e. adding things like a facial animation controls to the rig while the animator is working with overall movement and performance) Some people don't like the use of reference rigs simply because there are things that can go wrong, but if you always save a backup of the last rig that worked, you'll be safe not matter what goes wrong.

I really like the concept of using cloth, fur and hair on a mocap driven character. That is a GREAT thing to show off on your demo reel that shows that you have a solid understanding of simulations. Just remember to cache the simulations after you get the performance that you like so that it doesn't have to recalculate it every pass that you render.

Lastly, keep your demo reel short. The shorter the better. Definately less than 5 min. Just show them enough to get their tastebuds salivating. Be sure to show them one of each of those points that I mentioned above: rigging, scripting and simulation. Filling it with animation is not the answer. Just one shot will be enough to show that your rig works. I like to show the animation first and then the rig that made that animation possible, with a focus on specific rig requirements that you met and scripting involved.

That's it for now. This has been a very good dialogue and will be very beneficial for anyone reading it, not just Nikhil. I wish someone would have shared this stuff with me while I was still a student because it was all I could think about at the time and stressed me out to no end.

N8Girard

futurcraft
02-06-2007, 06:42 AM
animators have a skillset and a job that is very difficult and neccessary to do! They breath LIFE into these rigs! But you should be able to do that as well, even if only on a limited scale, so that you can relate well with the animators that you will always be working with. Besides, if you can show an animator a performance that they are happy with, they'll gain respect for you and what you do. Otherwise you'll always be just the fix-it man down the hall. I strongly encourage you to develop relationships with animators and I'm glad that you are already engaged in this process.

I also recommend that your animators REFERENCE your rig within the files that the animation resides in.

I really like the concept of using cloth, fur and hair on a mocap driven character. That is a GREAT thing to show off on your demo reel that shows that you have a solid understanding of simulations. Just remember to cache the simulations after you get the performance that you like so that it doesn't have to recalculate it every pass that you render.

Lastly, keep your demo reel short. The shorter the better. Definately less than 5 min. Just show them enough to get their tastebuds salivating. Be sure to show them one of each of those points that I mentioned above: rigging, scripting and simulation. Filling it with animation is not the answer. Just one shot will be enough to show that your rig works.

This has been a very good dialogue and will be very beneficial for anyone reading it, not just Nikhil. I wish someone would have shared this stuff with me while I was still a student because it was all I could think about at the time and stressed me out to no end.

N8Girard

Hey Girard...!

To start with... I never expected to get such valuable advice when I entered this thread.. I completely understand the importance of a rigger knowing how to do the animation. The animators I've worked with do have some idea about rigging & I have some idea about the animation though not in depth. But now after you stressed the importance of the animation knowhow, I guess I will have to take a more serious study into the animation process too. Though I would be sticking to being a rigger which I feel is my strong hold. I perfectly understand you when you say "Fix-it man" & I do appreciate you for being frank and outspoken about these things which not many would want to put forth upfront. I guess when a rigger knows more about animation & is able to fix things himself before handing over the rig to the animator, it certainly will have much more value & he also earns the respect for his expertise.

The referencing process is a very good thing to do. Thanks for sharing that piece of advice on how to work on it. It certainly saves a lot of trouble and time right through the pipeline..

The mocap rig with the simulations was something I thought was the best way to showcase a combination of features in one single character & avoid repetition in the reel. Glad you liked the idea...

My main concern on the animation of my rigs has certainly reduced thanks to both yours & Xen's inputs & I will also be getting the DVD by Aaron Holly to work on my Ogre rig after the current mocap simulation one. Rigging, scripting and simulation is exactly what my reel will have & I'm happy to hear that I am on the right track....

I really appreciate the time you guys took for sharing such valuable advice on the whole rigging process & the content for rigging demo reels...

Thanks a lot Girard & Xen... N Wish you all the best for the Rigging challenges & more... in your future endeavours...!

Regards,

Nikhil.

nlz
02-14-2007, 12:29 AM
I didn't rig so far at all. Anyway it's niiiice :thumbsup:

Just look at the popularity: Boobrigging > All zomfg :D

http://img78.imageshack.us/img78/251/boobrigginggk7.jpg

futurcraft
02-14-2007, 07:27 AM
Oh yeah...!!
Girard....!! You Rock..!!

N8Girard
02-15-2007, 06:49 AM
nlz,

it might have something to do with the word boob in the title of the video clip and other ppl on youtube are looking at it who are just looking for some nudity. It could be just pervs. But if ppl are genuinely interested in the process I went through, I could write up a tutorial or something though I pretty much explained my process earlier on in the thread.

Good luck with your own rigs. I'm curious to see how everyone approaches the issues. I guess that it's time to attack the legs soon.

N8

MrHobby
02-19-2007, 05:00 AM
Hi, just wanted to reply to your awsome techniques.
If this is what you can do in a short time for a CGTalk rigging challenge, your dragon rig is going to be spectacular!.

mysrh
03-02-2007, 11:56 AM
Hey N8,what an excellent rig. If you are willing to put tutorial on the jiggling stuff, you will make me one happy guy ;)

RobW720
03-12-2007, 10:44 PM
hey, i have been readin on your thread here and its super informative. I have learned quite a bit just by checkin out what you have been doing and trying to do something like it for myself. However this whole thread/challenge has dried up. any word on the progress of yourself or this challenge?

and yes, jiggely tutorials would be purdy cool :-D

N8Girard
03-12-2007, 10:54 PM
Hey guys,

I'm still in this challenge but because of the long terms that it has been set for, I think that a lot of the riggers lost interest after the first month or so. It needed to be set for a new body area every week or two week period at the max instead of a month apiece. I'm at the stage where I need to rig her legs and armor but haven't gotten around to it just yet.

So everyone wants me to demonstrate the jiggle techniques that I'm using huh? I explained some of it earlier but they were pretty vague references so if you didn't follow then I don't blame you. I'll see what I can do about putting together some sort of tutorial. I've never really assembled one before but it shouldn't be too difficult to figure out. At least you could learn some techniques from it like pointOnSurface nodes and soft body simulations.

I'm a little busy redesigning my website right now but I'll soon be finished with that and I'll post a tutorial later on this week or beginning of next week. I'm glad you guys are still interested in the rigging challenge. I know it's a great opportunity. We just need to work on the structure some more.

L8r,
N8

xenoid
03-13-2007, 12:51 PM
Just to add on to N8Girard's words...

I still think 1 month is a good time to explore and experiment the techniques... :)

Now for me, I did not lose interest in it, but had been caught with helping a friend with some stuff, and it's taking almost all my time at home for almost 2 months now, and still counting. I had to put this challenge aside for him (Sorry Todd!)

Hope this doesn't stop the effort of more rigging challenges in the future! (Directed at Todd :D)

And N8Girard, hope to see you finish this challenge... Can't wait to see and play with your final rig. :thumbsup:

ds5000
03-15-2007, 06:48 PM
that was an awsome idea to solve that rig issue with the boob bounce... is there any chance we could get a tutorial on the steps you took to create it? i for sure would be interested!

nottoshabi
03-17-2007, 09:03 PM
Nice job on the wrist can you elaborate a little on how you keep the volume throw all of the motions?

Thanks...

N8Girard
03-22-2007, 12:12 AM
Hey everyone,

Sorry I haven't posted anything recently. This new website of mine has been sucking up all of my time. But the good news is, IT'S DONE!!! So I have a special treat for everyone interested in checking out my work. http://n8girard.com CHECK IT OUT!!! I've got lots of my 3D stuff on there streaming over real-time in flash! I even posted some artwork and animations that hasn't been posted anywhere before. Below is a screen capture of the website. Enjoy everyone!!!

http://n8girard.com/images/big/clients_n8v2.jpg

Djampa
04-07-2007, 02:59 AM
Great rigging ! Congratulations !

Nice website and artwork too ! Keep it up !

CGTalk Moderation
04-07-2007, 02:59 AM
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.