So before I get into the making of Double Denim, there is something I need to get out of the way. This has been a tough few months. It’s been a very frustrating project with several hiccups in the team and in the creation process that were devastating enough to bring and animation student to the brink of hysteria. (Skin weights I’m directly looking at you.) That being said however, I couldn’t have been happier with the group I worked with. Both team members were not only some of the chillest and patient people I’ve ever worked with, but they were some extremely talented, hard working individuals who love what they do. It was a fantastic and fun group to be in and I’m happy I got the chance to work with them
So, this project.
We tossed around several project ideas initially, all very good. We immediately decided that the theme, Bravery, didn’t just have to have a dramatic or grand plot. It could be something as small as doing something you wouldn’t normally do. This is what we decided. A guy decides he’s going to wear double denim like it’s 1995.
So the first thing was decide what this dude was gonna look like. I was already doodling by the time I got home and had centered on one or two points of inspiration. The first being Jim Davis’s infamous John Arbuckle; someone known for his poor fashion choices and cornball personality.
To be perfelty honest, I had an idea of where to start so long as I started drawing using Jon as a reference. I tossed around a few ideas until I managed to come up with a slightly greasier looking guy.
This was a nice start, but the group felt out guy was probably a little less confident than these sketches implied if the whole point of out skit was about the courage to try a fashion choice that was normally frowned upon. So they went with my final sketch of this tall, gangly looking dude who clearly was just a little bit uncomfortable in his own skin.
We named him Lenry.
While I worked on the character, Glenn began tossing around ideas about the jacket itself. I don’t remember completely, but from what I recall Glenn had gotten it in his head that it would be hilarious if we had a denim jacket with a massive, terrifying, over joyed face that would fly at Lenry in the dream sequence. (This was easily the funniest and creepiest part of our skit.)
So another thing we began working on aside from storyboards, was the room design that we wanted the story to take place in. I set to work on this part of the project. I was really looking forward to modelling a room at that point as I felt like I definitely needed the practice. At this point, Yazz had made a joke that because double denim had been popular in the 90s, Lenry should have an odd fixation about that era. This made the project about 50% more fun than it already was and allowed us to experience a good amount of nostalgia while researching what we should have in the room.
To start, I began to look at bedrooms from the 90s and early 2000s. Back then the most popular styles were either grunge or pastel, so it was difficult to choose. As we worked however, we did allow ourselves to stray from the original theme. We began to mix it up a little more and let ourselves use the idea of nostalgia rather than total obsession on Lenry’s part.
Checked cover on bed in red nineties bedroom
While researching and beginning the modelling process, we also had the presentation for our storyboards as well. At this point in the project, we had officially been reduced to a group of three rather than the initial group of five we’d started with. This expanded our work load considerably, but it wasn’t as daunting as I thought it would be. We very easily picked up the slack and threw together an extremely concise storyline that the audience could follow. Here are a few of the frames
This of course was changed after a bit to make the dream sequence less nightmarish and more inspiring. The idea of the story is that Lenry should be eager to take on his new style, rather than terrified into it. So we wound up changing the fear in the scene to something akin to confusion.
At about this point we’d had some really great progress with the room. The blocking and planning done, the only thing that was stopping me was the frustrating blanket. I was using nCloth feature with a gravity field applied to it and the cloth was just not falling properly at all. Here were some earlier versions
As you can probably tell, the blanket is way too thick, and doesn’t really fall very naturally, so I’d wound up redoing it so many times I lost count. While I wasn’t fiddling with the blanket, I also went and made several props for the room as well.
Around this time I had also begun to model the character. I started by creating a very, very, basic model in maya and then transferred it over to Zbrush so I could begin sculpting out the details. This was a frustrating but fun process that took about a max of ten hours. When I finished, I sent him back over to maya and began the retopology process so we could animate with him. This is where things got very complicated for me, as I”m not very experienced with retopology, but after several days spent working over the model, I had managed to simplify the mesh enough to begin the rigging and blend shape process.
(Except not yet actually. I didn’t do the best job with the head topology, so Alec took the model for a day or so to fix it. Thank you Alec.)
The rigging process was, to understate, tedious and torturous. Building the skeleton is the easiest part of the rig usually, but in this case I was trying something new. I made some IK handles, which I had never done before, and while it worked pretty well for a first attempt, it took out a large chunk of my time. By the time I’d gotten around to actually binding the skin to the skeleton, it had been two days.
If there’s one thing that I cannot stand, it’s skin weighting, and that was the next process. Now there are some things that I don’t mind about it, such as the end result and the feeling of progress. This time I went out of my way to try a new kind of weighting called Interactive Bind. This type, as opposed to the paint weights tool, allows me to weight chunks of the mesh with two controls rather than several.
Painting weights usually has me sitting there selecting vertices for ages, sometimes one by one for the sake of detail. It can be difficult for me to tell where I want my weights to start and end and can lead to extremely frustration if I flood the wrong part and get stuck undoing the now deformed mesh for ages.
With interactive bind, I’m allowed to weight the skin with the help of a capsule control that wraps around the desired joint.
(Image courtesy of Jonathan Hughes)
When you move the green controls up to down, it blocks out the weight influence along with limbs until you get the perfect joint you’re looking for. If you find you’re still having issues with the weights (one joint drawing influence from the opposite) you can just as easily resort to painting skin weights to fix it. This process got me through the hardest parts of the skin weighting, and was far less stressful than the painting weights method that I had been using previously.
The face was created using normal blend shapes, which taught me that in the future, it would probably be better for my emotional health to just create a facial rig. Blend shapes are created by sculpting the vertices of the model, but never under any circumstances, deleting or changing any of the original mesh. The number of times I accidentally changed something the resulted in the entire model breaking was embarrassingly high. (Like the pitch of my voice each time I had to ask “Why is this happening?”) By the end of it though, I had some workable expressions that finally cooperated with the mesh.
So with the character finally rigged, I had gone back to touching up the room while Glenn got to work on the textures. He gave Lenry several outfits that worked very well with the colour scheme we’d had going, and still maintained the dorky persona he had
He had also been modelling and rigging our jacket character, and he’d come back with an absolutely fantastic result and you can see his process and work on his blog HERE.
(credit: Glenn O’Neill. Nice work Glenn!!)
With the room, I had finally gotten the blanket to fall the way I wanted it to, and I got started with texturing. This I found to be the most fun part of the project for me. I had originally been very nervous to do UV Mapping at the beginning of the semester, as it’s not something I was entirely confident I was doing correctly. However this time I was ready to go and excited to see the result. Even more exciting, I got to do bump mapping as well, which I was thrilled to figure out.
The textures I wound up using were free stock photos, some of which I wound up editing to use as bump maps.
This was the first version I had before Yazz added the rest of the props for the room. It looks a lot nicer once it’s been fully textured and filled up with all the 90s nostalgia.Yazz modelled a lot of the vintage toys and items for the room as well. The following are three awesome samples of her work for the bedroom scene. her process and more of her work can be found on her blog HERE
Yazz also made the dream sequence as well, which was a major part of the project. It’s the secondary environment and the setting in which Lenry is graced with the inspiration for double denim. It was a scene that needed to be both dream like and realistic as well, and they captured it perfectly. The red carpet set up is supposed to be a nod to the red carpet fashions of the 1990s, some of which were glaringly denim. Specifically this tragedy:
The Red Carpet sequence looks quite nice. The ropes are detailed, the poles are nicely textured, and it was easy and enjoyable to work in.
Very well lit too, courtesy of Glenn. He also did this really cool flick at the end of the carpet for Lenry to fall onto and bounce off of.
So at this point everything was finally textured and ready to begin animating. We all had scenes we were eager to do, and mapped out a good amount of work for everyone to start.
The rig proved to have troubles however, which would resurface again and again throughout the progress, and there were a few texturing problems as well. We’d managed to get the actual jacket on using a wrap deformer, when it came to animating, the jacket was anything but cooperative. I wound up having to tweak the skin weights multiple times to get something we could use. On top of them, large chunks of the model would vanish when the jacket was applied due to what I think were conflicting colour and texture settings.
(Lenry, coming out of his room to shame mankind)
Attempting to resize misbehaving mischievous model
I don’t know what happened or why he’s doing this to us
GET BACK HERE ASSHOLE
(Lenry, consistently making me regret choosing animation for a career)
Problems aside however, we at last came to and end of animating and what would begin the render process for us all. The things we’d managed to put together were wonderfully animated, and again, I’m really lucky to have landed with this group for the project.
Something I’ve started to try and keep in mind after the last few projects are what I would have done differently. On this, I would have loved to spend more time of the character. I already spent so many hours on him, but looking at the finished result, I can say that there are problems that I should have paid more attention to. Given that this is my first legitimate, complex character modelling and rigging, I am willing to be a little lenient with my efforts. However looking at this, I am painfully aware of Lenry’s faults. (restricted leg, finger and arm movement, limited facial expression, sloppy looking mesh)
For the next project I would be more than willing to take on another character, and I know what I have to do to make myself better and achieve a greater result I could do professionally one day. I am eager to learn from my mistakes, and look forward to doing so in the upcoming semester.
As of today, our extended deadline, we’re ready to render everything out and will have the finished film hopefully before the 29th. I’m happy that we finished all our work to hand in and we’re looking forward to sharing the finished product with everyone.
EDIT: The finished film can be viewed here!
Jacket, lighting carpet and textures, Lenry textures: Glenn O’Neill
Room Props, Dream Scene Carpet: Yazz Herron
Interactive Bind Reference: Jonathan Hughes