Greetings and salutations readers!
It has been a little while since I have filled out a post detailing my time on Bathroom Stall Galactica's production. Between Spring break, another production pass, and life's interference, its been hard to fill this post in about our time on BSG. So let's summarize what has happened, and what is going on presently.
My work on the character was the last update that I shared here, and even after that post, my work on the character did not end. On three separate instances, I had to return to the character's base model to edit his geometry, whether that was adjusting his shirt, fixing the amount of inside facing geometries, or redoing his mouth bag to alleviate some of the weird gaps between his teeth and cheek. Work on the character didn't yield any major external change, yet it caused a lot of re-rigging and delays. Ultimately, we ended up in a point where the character's deformations were live-able but not great, and we will hope that our animators can work around it in a consistent way.
After my teammate finished a version of his rig with a facial control, I spent several hours working on bringing his rig into Unreal Engine. For our main character, we were using Advanced Skeleton's facial tools to animate his face. There was some brief attempts where we had a facial rig with joints trying to go into Unreal Engine, but we quickly realized that Unreal Engine wanted a blendshape facial rig instead of a joint based one. Thankfully, the switch was easy. The real challenge was importing the facial rig into Unreal, and combating an error we kept encountering. Routinely, every single time that I attempted to import the rig, I got an error message saying that "rig could not be imported due to duplicate named joints "geo" were found". This particular error was super frustrating, since after painstakingly checking every single joint in the rig, not a single one was named "geo". The eventual solution, was that the rig referenced in both the published model, but had the imported stand in geometry as well, so the rig was somehow driving multiple copies of itself. It was a little bit of a misleading error, but after deleting that the rig came in with facial controls using the Advanced Skeleton FBX publish options. Once the character was in, I imported him into all the different Unreal sets, so that we could attach his hair groom to the character.
The next project that I worked on was the control panels for the bathroom/abduction bay. The work on this particular asset was largely focused on making different types of buttons/controls for the panel, so that it would be visually overwhelming and add more stress to our main character. Here is a quick in progress shot of the board's design with the template buttons nearby.
My plan is to place all the buttons, and then cut and delete any faces that are clipping through the displays. This particular model has the potential to be really time consuming and polygon heavy, so keeping stuff inner penetrating is of the utmost importance to me.
Additionally, I worked on some of the smaller props in the film as well, and created real rigs for them.
One of those assets was the clipboard that our alien checks when Jeremiah is abducted. I wanted this rig to have multiple sheets of paper that the alien could flip up and through, so rigging a piece of paper convincingly was a challenge. The clipboard rig required a lot of joints, but by evenly spacing them out, every subdivision line was just alternating between weights of 1 or .5 for every joint, to give it a nice clean bend and curl. I also created the abduction forms as decals and filled them with names of people who've helped the film in some way.
To make the document feel more authentic, I asked Avery to hand write all the different fates possible, and then I asked my father to use his nice hand-writing for the "Last Abduction" celebratory message. I chose this because I didn't want to hunt online for a handwriting font, that would be pretty obviously typed out.
From this point on, I worked in the different Unreal Engine projects to import all the new changes to the models, textures, rigs, and cameras, and began rendering out all the shots in the film. This pass was pretty rough, since we updated our rig with new geometry and hair sims, but the animation was still from the previous iteration where the main character was blocked geometry. This led to a lot of stretching, a lot of misplaced hats, and a lot of really awkward movements. It was also the first time that we re-rendered the entire film from the preview screening, so it put a lot of stress on the team getting renders out and organized properly. The renders themselves were slowed down by some anti-aliasing settings that took longer than I previously anticipated, and for allowing engine warm up frames for the hair and fog particle simulations. We managed to get all the renders in, except for the frames generated from the operating room. For some reason that is still unknown to us, when we hit the render button on the operating room, our main character would turn completely invisible, despite being visible in the viewport and being added to the set the exact same ways as all the other projects.
The production pass came, and my team and I got pretty eviscerated by the comments. Many assets were still missing or in their blocked form, the animation was non-existent on the main character, and there were the chunks from the previous iteration still present. The feedback was largely on our time management, but I had a couple peers personally call out my leadership of the film as well.
In that moment, I realized just how challenging it was to handle the leadership of this film, trying to be an artist who made stuff while also dictating how my teammates worked. I had my hand in so many pots, that it was really exhausting just talking about other's peoples work with them, let alone doing my own work. What I chose to do, was re-delegate and try to promote my other director into a more hands on oversight role of the film, so that I would be less involved with other people's work. I split our thesis film in half, where one half of us was only visual development and assets, while the other half was animation. By doing this, I took myself away from shot production and animation work and enabled my teammate to really oversee the quality of his brigade's work, freeing myself up to solely think about modeling assets and working with my brigade to make cool stuff.
Spring break came and went, and some personal work for other projects did as well. Part of me was reluctant to return to the grind that was BSG, and I delayed my effort a bit over the past week out of that reluctance. However, I dedicated a day to working on just the minigun, and got to enjoy the simple act of modeling without any rules or expectations on me.
Our thesis film is running out of time. However, stepping away from it for as long as I have has made me really appreciative of what we have made, and how lucky I am that I get to work closely with my best friends in the DAC. I would not be nearly as motivated or invested to contribute to this film if it weren't for the amazing people I get to spend my days with making jokes and gags in this film. I look forward to sharing more progress on this film as we wrap up this semester.
Comentários