Early Production:
Our group began work back in late February with ideas for plots and stories we could explore using the theme of “nature” that the instructors gave us. It was an overall fun, productive, and intuitive process in the classroom.
Dispite the wide arange of ideas we explored, a lot of the sotries we tried to look at mostly followed the same idea. A couple of animal characters are struggling to do something and then killed unceremoniously as a joke. Saying it like that makes it sound less funny but it was supposed to be a show of how dedicated but uncaring nature can be.
One idea that stuck with us was a short where a couple of swordfish were fighting over a pufferfish and get eaten by a shark. Part of it was because we kept returning to ideas about aquatic life, and the other part was because fish not having any legs would be a simple solution to the -no legs- rule we had to work around. In the end this one was the one we stuck to and continued to develop and change as time went on.
At some point, I think it was when we were making the Discord group, the name Swordfish and Caviar came to be and… it never really left. Even though we all agreed that there never was any caviar or anything to do with it in the short so it didn’t wholly make sense, but we never got around to changing it. That said, we recently changed the short itself to “Plenty of Fish”.
Characters:
So once we decided on what the sotry was (though it changed a lit during production but still mostly consistent) we began working on the characters. We had the pufferfish, a swordfish, another swordfish, and a shark, an earlier itteration of the script involved another fish riding the shark like a car and picking up the pufferfish.
By the the time it was March we already decided on the four characters, and dropped the other fish. The pufferfish we decided was supposed to be femenine, though balancing between cute and sexy was where we were exploring. For the two swordfish the idea of them being the stereotypical woman chasing guys was proposed so we made on the muscle-head, big chinned, “Chad” swordfish, and the other one was the most french looking stereotype ever of a swordfish, hence sometimes called “Frenchie”. The shark was a bit of a struggle, as in one version of the script he was vehicle, and in another a romanitc rival (???) to the swordfishes that outdoes them both. After some deliberation we decided to make the shark its own entity.
References:
While we were still deciding what role the characters would fulfill, Pinterest reared its head up and I made a board with various reference pictures of pufferfish, sharks, and swordfish, both real photos and cartoons designs. It also included Chad memes (as many as I could find on Pinterest anyways), sports cars with shark designs, and ancor tattoos (later on for Chad). While not as large as other Pinterest boards, it had enough references for designing the characters. I would later on need to add a few more reference pictures to it as I made the characters and certain details I gave them later on. Namely the ancor tattoo and colouration.
Concept Art:
I began doodling the characters prety early on as I wanted to get a good feel of what we were going to do with them. I strated with frenchie as I wanted to get used to the style before moving onto the other characters.
After agreeing on the characters it was decided that we would try and split up the workload into even bits for each one of us to pitch in and help as well as to make sure no-one person would do more or less work than everyone else. This included the characters, meaning that we each chose a character and finalized the designs to present to the group. So I chose the “chadfish” seeing as he needed to be developed a lot more.
So after making boards, and finalizing the character’s roles, we began sketching. I did some sketches for Chad, mostly exploring the head and shape of the fins. I wanted to lean into the macho meathead dynamic so I made his overall bodyshape triangular, but nothing really threatening so the edges were rounded off.
I liked the giant chin look so I looked around for some way to make it work with the swordfish nose, so that he could still use the sword dispite the massive chin. I also experimented with the eyebrows as I didn’t want something too aggressive looking but something still simple enough to make in Maya.
As seen in the picture above and below, he has Pacman like eyes. This is because while designing the characters someone proposed that we lean into the 1920’s style of cartoons as inspiration since the idea of guys showing up and trying to one up each other was more prevalent and accepted then as opposed to now. I agreed with that idea and made sure to keep that visual in mind while designing Chad.
At this point I should probably come clean and confess that I never actually made a final design of the Chad Swordfish. It was the middle of April and I hadn’t yet done it. I have no prior experience with modeling so I wanted to start as soon as I could to refine the model as much as I could eary on so it wouldn’t be a last minute thing. I took what I had in the sketches and made an amalgamation of the traits that Chad’s designs had.
Storyboard
Initial Ideas:
We had several different ideas at the begining of the short for the storyboard. As mentioned earlier, one idea included the shark being used as the ride for another fish. While the two swordfish were fighting he would come out of nowhere and eat them both before picking up the pufferfish and taking off. In another idea the pufferfish would encounter wach swordfish and they would show her something “interesting” about themselves. Chad would flex and pose with his muscles, and Frenchie would be sculpting a statue with rubbish. Since both are trying to get her attention they duel, while the increasingly anxious pufferfish is about to blow up. Then the shark comes and eats them. The pufferfish tries to approach the shark with romantic intentions, but the shark is still hungry so they aren’t reciprocated emotions.
Refined Ideas:
We later cleaned up the storyboard a bit so that the pufferfish sees each swordfish individually. First Chad swims up to her and while flexing accidentally knocks her away into Frenchie. We also later axed off the whole part about the romantic interest in the shark, instead it simply turns and eats the pufferfish while she is still in shock.
In mid April we presented what we had of our project and animatic to the teachers…
Animation
Animatic:
In all honestly we didn’t work on the animatic that much but we did put one together with some of the old storyboards. Special thanks to Eimer who put it together for us. She also included music and sound effects to the animatic so that we could all get a good sense of what it would sound like later on when we got to the previz and animation istelf. The animatic has the most bare parts of the story we have for the project, although it’s an older version of the script so not a lot is really in the final animation.
The final Animatic is
here
Previz:
While we were working on the previz, animatic, and character designs Eimer took charge and made a schedule for our team to follow and get the assignment done in time.
The previz we got working to rather quickly, before the animatic was really done. But a lot of the content was more up to date with the script. The models were just very simple deformed primitives to represent the characters, but later on we made the primitieves look more like the charactes by way of adding eyes, arms, mouths, and other spahes to make them more distincive and their motions more easily understood. Initially we had the part including Frenchie showing the pufferfish his sculpture, and the part in the end where the shark gets them both, before turning around and eating the pufferfish. But after we showed it to the techers they recomended that we cut down all the parts of the story that were “unnecessary” or took too long. Namely that we get rid of the statue, as it never has any relevance to the short. So we cut that part and replaced it with Frenchie making a painting of the pufferfish instead, also the scene with the shark was greatly cut down so that instead of several panning shots of the shark breach, turning, and looking down, it just swims straight up and eats both swordfish. We did include a bit in the end where the shark swims back around and gets her. It was decided that this part would be an optional ending depending on how long the short would be we would keep it or cut it.
The final previz short is
here
Modeling:
This was the part that took the most time but also the part I worked the most on in the project.
We were told by the instructors that we had the option of modeling our character either in Maya or Blender, and at the time I was still trying to get used to Blender and its setup. Given how it was still new to me and I was already working on another assignment with Blender, I decided to make the character model for this assignement in Maya as I was more used to how it worked there. I started by combining primitives together to make the head and chin, and later deformed a few others for the body and fins.
After a while I added the nose/sword, head fin, brows, and eyes. The mouth was tough for me, not knowing how to make one and waiting for the instructors for help, I didn’t do it for a long time. But when I finally did do it, I used a primitive and deforemed it into the shape of a smile.
I 
I did make a few more changes to the model as I went on, namely that the arms/fins were two separate pieces, which… in retrospect should’ve been the one thing I didn’t change. I also added a bit more to the head so that the mouth could sit less aqwardly.
After that I began working on the rigging for the Chad model. The rigging part of the modelling process was honestly the hardest part for me to get, and even though it was the most simplistic skeleton I could construct, I still found errors and problems with it up until the penultimate week of production. Chad being a swordfish, even if anthopomorphised, only needed to move his head, arms/fins, and tail, everything else was either stationary or would move with the rest of the body. So I made a cross shaped rig at the start but as will become evident later, it would change a lot.
After showing my work to one of the instructors, Alec, I was advised to make the model using fewer parts combined together and to extrude and deform fewer primitives instead. So I did that, and I do have to say that that was a more better looking build overall in comparison.
I did also get helpful advice for how to make a mouth for a character model, so I finally added that in.
After that I learned how to add an aim control and rig to the eyes so that they could pivot in place and look in whatever direction I need them to.
It took a bit of getting used to and more fiddling with the controls to get it right.
I later on found out how to make facial expressions thansk to instructors once again, and I used blendshapes and the shape editor to make a smiley face and frowny face for Chad. I had to learn and make rigs to control those too, which at the time was a bit hard to get but now I think its kinda fun to do.
I did also include a blendshape to make the eyes blink, which took longer than I antisipated to make on account of shennanigins with the multicut tool and edges not quite working how I thought they would. I eventually did get it working though.
After getting all that done with the rigging and modeling I did the UV mapping and texturing fo it and applied it to the model.
I had a few issues with UV mapping as some of the polygons were separated or joined up in some weird ways so it took a while to get it right. Additionally while texturing I had to go back and fix the UVs again which meant retexturing it again, which happened acouple of times. But once done I was able to get the textures on it and sorted out.
I still have problems with getting used to Substance Painter 3D so that was part of the problem with the texture phase of the project. The other problem that took the most time for me to sort out, and ironically had the simplest and quickest solution was adding the textures to the Maya file. I couldn’t figure for a couple of days why the textures wouldn’t render correctly in Maya, but then as it turned out I had pressed the 6 hotkey to render it in.
However there were still a lot of problems with the rigging at that stage of development.
I asked for the instructors on how to fix the problem and as per their investigating and analysis I had to give the rig a complete overhaul.
The things I changed were the skin binding, the size and orientation controllers, adding joints to the eyes and brows, and undoing all the parenting of the joints and instead adding aim constraints and point constraints (later), and grouping all the controllers and joints into each other for more organisation and simplicity. I am really thankgul to Henry for helping me sorth that out as it was tremendously helpful and easy to follow.
After all that the next phase was the animation phase, I was more comfortable here so it wasn’t very problematic or difficult, save for a few instances where I had to sort out some skinning issues with the models.
Late Production
Animation:
As with the character designing process we split up the animation of the scenes so that each one of us animated a pair of scenes and no one was excluded from the work, or was overworked. I decided to do scene 3 and 5, which were the scenes were Chad is flexing his muslces to try and impress the pufferfsh, and the scnene where he and Frenchie are confronting each other and staring one another down. I was the first one to finish the rigging and modeling so I needed to wait for the others to polish and fix their models, so in the meantime I used simple primitives to mark where the other characters were and how they’d move. Thankfully the other guys pulled through and got the models ready for use in the animation, so I was able to include the model for Frenchie and the pufferfish. After that everything else fell into place. The lighting as a little tricky but I was able to make it look acceptable thanks to the arnold-render view. Initially too bright, my team-mates suggested I tone down the intensity of the lights so that they would wouldn’t overdo the shots. There was also a model for the 3D Background in the group that I used for the shot too, thanks to Jonathan for puting together for us.
We did have a slight issue of communication and inexperience when dealing with the backgrounds a we originally wanted the background to be 2D but were struggling to get that working properly. We decided to use two different methods of rendering the scenes so that we could include the 2D backgrounds we made.
Rendering:
In the rendering phase of the assignment we had a problem figuring out how to include the 2D Backgrounds. We had two approaches on how to do it, one was to render the scenes as tiffs and then added the backgrounds in post the other was to layer the 2D backgrounds into the maya files and render the layers seprately. I opted for rendering the scenes as Tiffs and adding the backgrounds in Adobe After Effects.
I was still a little apprehensive of the rendering process and the file so I tested the render settings before fully rendering both scenes.
It actually took me a while to find out how to splice all the frames correctly as I made the frame rate to slow at one point and made a 9 second video 4 minutes long accidentally. But I did fix that, and managed to finish both of the scenes with enough time to focus on the rest of the post production.
Finalized Scenes:
The final animated scenes I worked on are
here
and here
Final Animation:
The final animated short is available here
Animated short
Thanks to Eimer to editing all the scenes together and keeping the whole short film within the time limit.
And thanks to Mark and Jon for their contributions to the short.
Reflection:
A lot of things happened during the production and development of this group project. But I’ll try to get straight to the point.
In our group setting a schedule and dividing the workload between each one of us worked extremely well for us all. The schedule helped to keep us in line and getting certain parts of production done quick enough to work on the next part of the process. I think that splitting up the work so that each one of us was responsible for one character, including concept art, design, modelling, and animation, was the most beneficial thing that we did in the group as it meant we were focussed precicely on our individual characters and didn’t have to worry about having to design and make the other characters. With exception to the storyboarding and animatic, the rest of the process allowed us to do so as for when we were animating we would share our models with one another.
In my own opinion the one thing that didn’t work so well was communication with each other. I noticed as the project went on there was a lot of miscommunication and lack of information at points where they were essential. This was most apparent to me with the animatic and storyboards as we hadn’t really all agreed to one official storyboard as the one we’d follow through and through. As a result our animatic I think was very different from our final animated short. I was a bit nervous as I was never clear on what final script was, and I can’t speak for my team mates but I was rather uncertain about that.
If I were to restart the whole module again, I’m not sure how many things I’d do differently but I know a couple of things I’d do differently and a couple I wouldn’t do. Firstly, I would try and make a finalized and completed Character Reference sheet for my character, even if not very detailed or expansive I would have preferred having one to reference directly as opposed to a couple of sketches. On that note, I probably should’ve focussed more on my character design as opposed to sketching the other ones when time was of the essence. Had I not done so many sketches of Frenchie I could’ve made a good ref sheet for Chad. I also had the issue of uncertainty when it came to storyboards and decision making. Had I been more outspoken and directly asked about it and tried to get everyone into agreement about what the final story was that we had to follow I think that would’ve solved the issue.
Finally the other thing, or the other two things I had trouble with were the rigging for the character and the textures. I’m still not the best when it comes to using Substance Blender 3D so I think that I need more time with that to get used to that part of the process. Whiloe it wasn’t a really big deal in the end, I think getting used to Substance and the different textures in it would go a long way. As for the rigging, I also need to work more on that as well. The rigging really got me as the deadline was fast approaching, and the fixtures kept piling up on each other. Studying and practicing with Maya is the way to go. If I can learn to handle the rigging an making the skeletons and joints for models that would save me a lot of hassle in the future.
Overall I think this year I’ve had in the class has been a good one. I’ve learned from last semester some things I should and shouldn’t be doing. While it took some time to stop doing some things, I have had time to work on the things I need to be doing more of this year, and am all the more productive for it. For example I learned last year that I should try to finsih with the concept art stage more quickly as I take a long time to finish that. Also to work as soon as possible as problems will inevitably occur and its better to get help sorting them out sooner than later. So I’ve been able to do that this year. While I have been able to talk with the team members more, I haven’t been really out going this year so I haven’t had a good change to know my classmates mush this year. That I don’t think I would have traded this university experience for anything. Everything came out with the best outcome it could’ve have been in all honesty.
And that’s good enough for me.