Unit 5 : Arachnid Locomotion

In unit 5, we explore the locomotion of an arachnid creature. I have never animated a creature like this before so it was a new experience and learning curve for me. When I first saw the animation of the spider in the tutorials, I was initially overwhelmed with the movement and how many legs there were to animate. However, when the cycle was broken down, it became clear that it was a repetitive pattern of the legs working in opposition. It was just a matter of animating one leg, and copying with frames onto the next leg with an offset. After watching all the tutorials and technical videos provided, it became apparent that this animation cycle was simple to produce, especially compared to the other animations I have previously created on this course. According to Barth (2020), spiders move by spreading their weight over their eight legs, that all need to work together. They do this by moving their legs in groups of four, with the ones diagonally opposite moving more or less at the same time. This is very useful information to help break down this cycle and to create the rhythm needed for this locomotion.I enjoyed studying and understanding the fundamentals of this animation, particularly the unique movements of a spider's walk and the overlapping action of each leg.

To approach this animation, I downloaded "Raa the Spider" rig and created and set the project. I was concerned about how well this rig would be considering how old it is (at least 10 years old) and as stated in the tutorials was created by an inexperienced rigger. However, for this walk cycle, it's a fairly simple rig and worked just fine.

The only technical issues were the controls of the legs that were created on polygons rather than nurbs curves which is not the convential soway to have controls or industry standard, therefore it was approached differently. This meant that the controls were visible in the playblast as they were polygons, so they were added to a layer and turned on or off when needed, it did add another step to the work flow but did not make a massive difference on the key frames. Typically, I animate with stepped tangents and then add and refine the keys from there. But after watching the tutorial, I decided it was more time efficient to work with curved tangents from the first pass. The overall animation only had 3 different poses on a 16 frame cycle. The most time consuming part was copying and pasting the keys onto the other legs. The 1st front legs moved in opposition which was simple enough to understand. Then the 3rd set of legs mirrored the front. The 2nd set of legs moved in opposition of the first set and the back legs mirrored them. To help break this down, I also looked to the legs that moved the same direction. So for example, the front right leg and the 2nd left leg had the same movements. 

By starting with the left leg in contact pose, then the next keyed frame moving back, and then next the leg rises and finally back to contact pose (same as frame 1). So the actual movement was simple enough, and then by adjusting the tangents, there was a consistent flow of movement when the leg is touching the floor. I did, however, notice the translation Y for this rig was not true to 0; therefore, when the leg is making contact, it's actually below 0. This was a challenge as I had to add an additional keyframe at frame 5 to avoid the leg dipping down. I also found this was the case for every leg, so adjustments had to be made per leg. Copying and pasting the keyframes always has its challenges, but for the front right leg, it was fine, and I offset the leg by 8 frames. I should also mention the pre and post infinity cycles were applied to keep the cycle moving continuously for any duration of time without having to manually create the frames. The next step was to copy and paste on the next set of legs that mirrored the first set. The translations copied fine, but there were issues in that the 3rd set of legs (although this applied to all the remaining sets of legs) were technically originally positioned differently from the first set. Therefore, the keyframes weren't necessarily accurately transferable. I made adjustments to this set of legs as they were stretched out or dipped into the ground. Another thing to note is that the rotations of this set of legs were different from the front as well; the front legs were more rotated like tiptoes, and the rest were like bent feet. So this did affect the copying and pasting of the keys. However, as there were not many keys on each leg, it didn't take long to make these adjustments on each set of legs. The back legs also took more time to adjust because, again, they were positioned differently from the front legs. 

One new technical tool I learnt was "breaking" the tangents in order to adjust the curves on one side only. I really found this useful and applied it in multiple areas so the transition of the leg rising was slower, but didn't affect the previous unkeyed frames. Once I felt confident in the leg locomotion, I moved onto the body. This rig is really limited in it's controls so there wasn't much movement that could be added to the spider, but the again spiders don't have a wide range of movements either. So I applied the secondary action of the body and pincers and offset them by 2. I also added movement on the "eyes" and learnt a new technical way to invert the frames without having to manually do it yourself as the eyes moved in opposition by applying "*=-1" to all the frames. Again, something that is really useful for future animations. 

Once I refined the keyframes and tangents the animation was completed to the best of my ability. What I've learned from this animation is the principles of an arachnid locomotion or 8-legged creature. I feel confident in this new knowledge, however, I believe this is the more simple animation I've created so far. This made me want to push my creative side more on this animation as I wanted some individualism, but I found it hard to do so with the type of rig and resources out there. This is the final animation, shown at different camera angles. I am pleased with the cycle and feel I've grasped a sense of the intricacies and motion of this spider. The most significant thing I've learned is the way spider walk and the repetitive pattern of motion. I think this is most useful in future animations of creatures, monsters or anything with 8 legs. 

---

The Final Animation

I furthered this animation into a storyline. I wanted to create a quick and effective story with this type of rig and the animation I've already created. I thought it would be interesting to create a gigantic arachnid crawling across a city, but from the perspective of a civilian. I downloaded a low-ploy model of an apartment building, (to keep on theme with the rig) and placed the model in different areas to create a set. I did have to make an adjustment to the model to create a window, as the model's windows were filled faces. This didn't take long to add in edge loops and deleted the faces to create window. Then I animated the camera with small wiggles and movements to resemble a person. I think the overall idea is executed well, however, due to the rig being simple, the geometry of the apartments were visible through the legs and upon troubleshooting a possible reason may be due to low-poly or holes in the geometry of this rig. To which I wasn't going to spend time fixing as in the industry, this wouldn't be my department and also I would have to dive into the geometry of this spider model. Therefore, I tried to disguise this issue as best I could. There is definitely some improvement that could be made on the legs. There is some sliding of the feet and the motion of the feet hitting the ground could be faster and more solid to create a sense of weight.

---

Reflections and Refinements

Final render of a Spider locomotion

As I reviewed my work, I decided I was not pleased with the final shot of the spider walking across the town. It didn't feel smooth and believable, and I had issues with rendering out the set. It is also hard to see the spider fully walking, which rather highlights the weaknesses of the rig and animation. For example, as mentioned in my previous post, the translation values on the legs were not rigged accurately, making it hard to maintain a constant flow of speed on each leg. Therefore, the legs "slid" and "dipped," but when looking at the graph editor, in theory, they should be still and in place. After spending a considerable amount of time editing these frames and curves to the best of my ability, the issues were still visible. Part of this challenge might be due to the fact that the original animation is on the spot, and then using the main control to move it across and finding the right speed was hard. I decided to change the story or presentation of this animation to my advantage. I considered that since this rig is low poly anyway, it wouldn't need to have any close-up shots to show detail, or in this case, the lack thereof. I used the content browser to add in a tree and put in a plane for the sky. As spiders are naturally in trees and climb them, it seemed fitting to do so. It's not my most creative idea, but I wanted to enhance the quality of this animation by not highlighting its flaws. I did include side-by-side shots of the spider anyway so you can see the construction of the cycle. To advance this cycle, I might consider using a different rig and having the spider climb onto a different surface.

This exercise has also made me think about how useful having this skill is for animating an arachnid creature and becoming a creature animator. Looking at a blog post by Yellowbrick (2024), I can understand the importance of having this essential knowledge in order to create not just real-life creatures, but fantasy and fictional ones too, pushing creativity and possibilities. It's also key to acquire these technical and theoretical abilities to become an advanced animator with a wide range of skills.