Stephanie came to us already very talented and skilled and went through our Python 101 course. We are happy to have been able to contribute to her training and help make her even more efficient. Join us in our interview with her and if you want to read about another talented character TD check out our past interview with Gio Coutinho – Rooster Teeth Productions
Welcome Stephanie Wagner Rigging TD MPC and Rigging Dojo Almuna.
Let’s get started with a bit of background about you, and what led you to computer graphics and rigging as your specialty?
I have always had a technical and artistic curiosity. I created my first website when I was around nine years old and played around with Ubuntu Linux in my teens. One day I was watching a video game “making of” and was fascinated by a cgi arm in motion. After watching a ton of videos and reading up on the subject I applied and was lucky enough to get an internship at a motion-capture company called metricminds where I learned a lot about the animation pipeline. I also enrolled into the Games Academy – art & animation class where I was involved in rigging and animating characters for student projects.
At first I found it challenging to learn rigging but it quickly became the field I felt most passionate about. I really enjoyed the structural, problem solving element of rigging and loved the feeling when it all came together.
Can you share your learning curve and experience over the years as a TD, from starting out to working on new tech and across different studios.
During the first weeks working at MPC I was determined to learn as much as I could about working with the pipeline and understanding procedural rigging. I also needed to learn a lot more about anatomy since I was responsible for a muscle digital double rig. I tried to ask as many questions as I could to get myself up to speed and found this was the best way to improve. Working closely together with a lead really helped with learning the inhouse tools, improving my anatomy knowledge and my workflow in general. I enjoyed deformation rigging the most and specialized more in this field. Over time I became confident placing anatomy, using deformers and setting up simulations which has become my strength and passion. I really benefit from sharing knowledge and brainstorming with colleagues which I find to be the most rewarding aspect of being a TD.
In general Every studio has a different approach and a different tool set available to their artists. Overall it does not take a lot of time to start working using inhouse tools and conform to a new rigging pipeline. Being exposed to various tools helped me grow as a TD and broadened my skillset.
We talk a lot about model topology and the default pose to rig, do you have a preference?
A relaxed and symmetrical pose is preferable meaning slightly bend legs, arms and fingers. However if I get a model in a different pose I like to adjust it to my needs and rebind the skinCluster in some cases. I also like to adjust the topology if I have to or add some more details to certain areas such as armpits.
If you adjust the pose do you then have to send it back to modeling? What is the back and forth process look like then with the other department?
Usually I receive very good models to begin with and the changes are very minor. The model in it’s default pose is usually matched to reference from the client and can’t simply be changed. Adjusting the pose in rigging – to create a perfect tPose for example – is not a change that needs to be picked up by modelling since I can save different poses in the rig itself.
If anatomy or topology needs to be adjusted in the beginning of a project I can always propose reasonable changes and get the model updated by modelling. Towards the end of a show it usually is much quicker to adjust the model in the rig directly and add the changes as a front of chain blendshape.
Deformations are always a balancing act between character design, topology and animation needs. How do you approach skinning and final deformation work?
It’s essential to get your pivots right. I usually lay out my joints quiet roughly and do a test bind to find the right placement of the joints. I also like to place actual skeleton geometry in my rigs to find the best joint position.
If I am skinning a character for an animation rig I take a bit of an old school approach. Classic linear bind, 1 max influence set to interactive skinWeighting. I block in all the weights from the core to the limbs. Afterwards I set my max influences to 2 to 3 and use the paintWeightsTool to smooth the skinning. For polishing I set the max influences to 20 and use the weightHammer. I try not to use too many helperJoints to keep the animation rig as fast as possible.
For a simple deformationRig I like to add some helperJoints to the rig. On top of that I sculpt correctives for better volume preservation and to create more interesting deformation. It does not take a lot of time to slide the topology around, move some edges out to create wrinkles and sculpt a bit of a flexing biceps shape for example.
If I had to do a setup in vanilla Maya I would create a shape in the default pose, do a blendShape and control the envelope with drivenKeys. To make it more interesting I would apply a cMuscleNode and add some relax, smooth and slide attributes.
Most importantly for rigging deformation is having a bit of animation going on. A simple technical workout (elbow rotate in/out) while skinning or sculpting gives you a perfect idea of how the character is deforming in motion and if your weights/shapes are holding up between poses.
You have done a large number of really great creature and animal muscle systems, can you talk about the process and what non-software specific tips readers could use to learn or prepare for a job doing that.
Besides working with great tools and technology It mostly comes down to a good foundation, experience and an eye for deformation.
A foundation can be build up by studying anatomy, drawing muscles on top of a character – this will tell you a lot about muscles in motion. Looking at characters in (slow-)motion to learn about shapes of muscles, what they look like flexing. You slowly develop an eye for deformation and why some CG feels good or wrong. Sculpting can also be very beneficial to develop a sense for volume and proportion. Practicing will make you more confident and only benefit your work.
If you want to dive into deformation rigging I would recommend to take a certain body part, e.g. the arms and really try to understand the human body in motion. Just to get some questions out there: where do I need to connect the biceps muscle to create a nice flex? Why is there a dimp in my shoulder when I lift up my arm? How do the ulnar and the radius function? How does the deformation of the lowerArm change when I lift my finger up?
Great references:
- Planet Earth 2
- Animal Anatomy For Artists by Elliot Goldfinger
- Human Anatomy For Artists by Elliot Goldfinger
- Anatomy For Sculptors by Uldis Zarins & Sandis Kondrats
- Bammes
Can you talk about your approach tacking on a character from scratch vs. taking over a character from another artist.
If I get the chance to do a character from start to finish I firstly take the time to roughly plan my schedule for each step required (jointLayout, skinCluster, muscleRig, correctives,…), look at reference and mostly work with range of motion animation until animation picked up the character. I usually try to get the character deformation to 90% final as fast as possible to have enough time to polish and work on feedback from the supervisor or further animation requests. Doing the last 10% usually involves working on details and dynamics which really bring the character to life but take up a little bit more time than the other 90% in my experience.
It is rare to take on major fixed from other characters. My leads usually try to keep the same person on their task. If a person leaves or is busy on a different show fixes can always vary from changing constraints, adding more controls or doing a simple model update. In vary rare occasions I have picked up a first pass deformation rig and reworked the muscleLayout up until the final deformation. From my experience TDs usually leave their work in a good place and it is easy to update/improve their work. Even though naming conventions might be different from artist to artist most of us have a similar work flow.
Code vs. Visual rigging systems- You took our Python 101 course at one point, do you still do much scripting as part of your daily job?
I mostly use scripting as a tool to make my day more efficient. I have a few lines of code for this and that, things I find myself doing a lot such as wraps, creating base meshes, clusters, etc. everything that saves me a couple of clicks.
I personally like to do scripted deformationRigs since it makes it easier to handle model updates or use the script to build a different character in the future. Scripting procedural props, facial- and controlRigs goes without saying in the companies I have worked for though.
Do you use any publicly available scripts or tools when rigging?
The first tool I download when I start a fresh Maya is “comet” – of course! If you don’t have it, get it. It is a life and time saver. Besides that I always found brave rabbit to be a great source for tools. I personally don’t use ngSkinTools but a lot of riggers really love it. Definitely worth a try.
How much stock Maya vs. studio tools are you using?
I am mostly using inhouse tools for rigging my characters (from orienting joints to final touches). That said I find myself using a lot of the maya modelling tools, transfer attributes, etc on a day to day basis.
Anything you wish would get fixed or changed in the software to improve production work and make it easier?
It would be great to see Autodesk reinventing/supporting a new muscle system. It can be very challenging to tackle cMuscle and there is only little help around on the internet and barely any documentation. The cMuscle binding is not very intuitive, cMuscle overflows the outliner nodes and the sculpting of a muscleShape is very limited.
Do you see any major advancement in rigging for characters in the near future or are the techniques mostly standardized now?
Knowledge is constantly being passed on in companies and there are a lot of ambitious TDs pushing the standards of rigging on a daily basis. SIGGRAPH is a great example for this.
I do think that the core process of rigging is mostly standardized now, but there is a lot of room for improvements. I think we are all aiming to create fast, intuitive control rigs and outstanding deformation. Understanding motion is challenging and I am very excited to see what will be achieved in the future.
What have you found , training wise to have been the most helpful to you growing as a character TD? Any last tips or advice you can give to someone that wants to improve their skills as a TD?
When trying to solve a problem I always go back to pen and paper and write down the issue along with solutions that come up on my mind. I find this very beneficial to staying focused and finding the easiest and fastest approach. This brings me to the best training you can do: ask questions. Ask yourself, the internet and colleagues. The Maya documentation is actually the best resource when trying to improve your understanding of Maya. Understanding what is going on in your rig is very rewarding when it comes down to debugging or improving on existing tech. It is very simple but in my opinion the most essential advice I can give.
I personally enjoy watching reels/tutorials on vimeo to get inspired by other TDs or just to see a different approach to an issue. I also like to watch documentaries about animals in my spare time to learn more about their movement/deformation.
Another helpful method is playblasting your rig in motion and taking a critical look at it – even better if you have real life footage playing along with it. If you have quicktime you can scrub forth and back through the timeline to see if the deformation feels anatomical correct. I often find myself playblasting and updating a certain area several times before I am satisfied with the result.
Lastly and more on a personal level. Don’t be shy – feel free to approach people online or offline to get advice. Be hands-on! If you don’t like the topology or default pose – change it. Be confident in your work, try not to be arrogant (it will make you easier to work with). Be passionate about the work you do and enjoy it.
At Rigging Dojo we get a high percent of female students and some of our most successful students have been women. If at all, has gender had an effect on your career and do you work with a balanced crew or is it mostly a male crew?
First of all I had the pleasure of working and meeting some amazing and very talented woman in VFX studios. No matter if it is production, a technical or artistic position – a lot of woman are not only good at their job but also very organized, friendly and easy to work with.
Nowadays there is already a lot of woman in VFX. However I would love to see more woman in technical and leading positions overall.
Even though the rigging department is mostly a male crew there is always more or less a handful of female riggers in the team. From my experience a lot of the girls showed a lot of interest in deformation rigging, but others were really ambitious about scripting and control rigging.
I think the position of a Rigging TD appears to be a very technical job until you take a closer look at it. It is a very versatile field in which – mostly if you are in a larger company – you can focus on a subject that really interests you. Some of these are more artistic than others, but scripting and coding can also be seen as a tool to create something beautiful. In the end the technical part of the job is very creative in itself.
Is there something that you see missing in current TD skills or reels?
First of all, I am very guilty of the following myself and I wish had known a few of these points when I put my first reel online 😉
Most rigging reels start with a tPose bipedal character, showing an IK/FK switch, moving the hip around and rushing through the foot poses often followed by three minutes of showing other characters with the same functionality. It seems like a good idea to show that you understood the basics of rigging but does not show anything out of the ordinary. Rather take the time to show that you have build a solid rig and/or found a solution for a specific issue.
I would much more prefer seeing a reel showcasing two to three of someones best pieces, all animated and presented well. A range of motion is quiet easy to animate and viewport 2.0 is a good tool to present a rig. There definitely needs to be enough time taken in the reel to see the deformation of the work and highlight one or two specialties of the rig or an autorigging tool.
Last question – what book are you reading right now or last finished?
I fell in love with Stephen King when I discovered his novel “11/22/63” and finished “Misery” a couple of months ago. After reading several of his books over the past few years I started “American Gods” by Neil Gaiman to give a different author a shot 😉
How can people best find you online?
It’s easiest to give me a shout on linkedin or vimeo, but for a general impression of my work it’s worth taking a look at my website: http://stephiewagner.weebly.com/
Thank you for your time
Bonus:
Production Discussion
What it was like working on “The Jungle book”
*check out the behind the scenes video to see some of the amazing work from the MPC team including Stephanie.
Working on Disney’s “The Jungle Book” was a great opportunity quiet early on in my career. I was lucky enough to work on the show from preproduction to final shot work. The big title of the movie attracted a lot of talent and created excitement within the team and across different departments. Every artist is really ambitious to deliver great and innovative work.
The rigging team on the show was amazingly talented and fun to work with. Instead of having one lead we had three leads looking after facial (Nico Sanghrajka), puppet (Andy Phillips) and deformation (James Hood) rigging as well as the character supervisor Ben Jones who had been working in the rigging department for many years himself. They all worked together very well, gave great feedback and kept the team focused.
One of the most challenging creatures I worked on was the Pangolin – such an adorable little feller. It was very important that the scales were set up correctly in the rig and not having to be adjusted in techAnim/creatureFX for every shot. In order to do this I am using an inhouse collision deformer on single-sided geometry on each scale and letting the lower plane collide with the ones above it. To create dimension I am pushing out the tip of the scale when the object is compressing. The double-sided geometry is wrapped to each corresponding geometry.
Breaking it down in a few steps makes it sound like a simple and straightforward set up but I remember struggling a lot with the skin showing through the scales in certain poses, too much volume being created in certain areas or scales breaking when the limbs were moving far forward.
A few days before the deadline I realized that I had to take a different approach. Instead of having the body drive the scales I needed to let the scales drive the body. I created a very smooth skinning on the entire body and wrapped the covered areas to the scales which works really well.
An other character I enjoyed working on and helped me become a better artist was the female white wolf Raksha, who raised Mowgli in the jungle. Working on a hero character is very exciting and it is key to not feel anxious or overwhelmed by the challenge. A hero will have a lot of screen time, close up shots and a wide range of motion. There will be a lot of focus and critique from the supervisors. As a Rigging TD you need to work very closely together with other departments to create something appealing for the big screen.
Doing the initial muscleRig was very straight forward, but the finishing touches took a lot more time and effort. Raksha is beautifully groomed so I created shapes to relax and flex her muscles along with wrinkles along the skin as well as skinSliding to bring the fur alive and make her look more believable. I was focused on having her skin feel very smooth and make the bones look very pronounced.
Looking back at my work now I regret not pushing the muscle dynamics and shapes even further to really draw the viewers audience attention onto her. Nowadays I know that it is better to exaggerate and then turn it down more when the first shot renders are done.
Can you talk about the difference between working on a massive VFX film like JungleBook and some of the other types of projects you have worked on?
On smaller projects it is very common to work close together with other departments, especially during the validation phase of the character proportions. Once I recieve a model I lay out a first jointLayout and skeleton geometry to check the overall anatomy and limb length. Afterwards I create a controlRig which I pass on to animation to do some testings and reference line ups.
While animation is working with the first pass of the rig I usually refine the skinning and adjust joint pivots. The skeleton geometry tells me a lot about the current proportions of the character and if it needs to be adjusted by modelling. As I said earlier I like to update the model myself as well and try to create a blendShape of the changes I am proposing or I do a drawing of muscles on top of the model to visualize the adjustments I have in mind. I think it’s important to communicate with modelling and animation a lot during this phase and spot every little issue that might come up in the future.
By the time shots pick up there are more and more requests from animation. The animation supervisors usually know what they are looking for. The demands of animation vary from character to character but most of the time their request are minor. Sometimes existing technology needs to be upgraded (a pathRig or extra controls for example) or it can be as simple as showing a different approach to an animator using the current rig. In general I always try to keep in touch with animation and help if there are any issues or requests, such as caching a rig.
If later on in the project modelling has to do model updates they usually try to keep the same topology and proportions. Mostly only volume or details gets slightly modified and can easily be updated in the rig. If the proportion are about to change it has to be discussed with animation since updating the joint pivot will affect every shot they have been working on.
In general – and a great part of the job – Riggers deal with many departments on a day to day basis almost towards the end of a movie. It might be a request from lookdev regarding the motion blur of a wheel, or a geometry cache that needs to be done quickly for grooming, a skinFix that can be done quickly in rigging instead of techAnim, and so on. Being hands on is very important but it is also fun to be a big part of the VFX pipeline.