Wednesday, June 7, 2023

The Animated Athlete Returns!

Welcome to this sprint's blog post! I’m your host, Mackenzie and I’m here to give you all the latest on this project! 

Before we get into the nitty-gritty of what I did this week, let's take a brief look at our animation render.

And here is the Stepped Tangents version of the final product! 

So, what's different?

A lot, actually! Here is the initial submission I presented in my first presentation two weeks ago.

During this two week session, I implemented several changes, which I'll briefly run through here before we get into the meat of the blog post:

  • Slowed the animation down by 1.75x. 
  • Iterated on the animation based on teacher feedback.
  • Added rotation to the soccer balls.
  • Added movement to the hips.
  • Used IK/FK switching on the hand to make it stick to the ground.
  • Experimented with different tangent types in the graph editor.
  • Added a basic lighting setup, consisting of 3 directional lights at 0.8 - 1.3 intensity.
  • Did a simple Arnold render of my character rig. 
  • Cleared animation of any non-snapped keys.
  • Removed extraneous keyframes to maintain line-of-action.
  • Emphasized squash and stretch on the soccer ball. 
  • Implemented the infinity wall into the project.
  • Resized the infinity wall to make it less distracting. 
  • Added basic facial animation.
  • Adjusted Jira board to make up for the advancement of the lighting task to this sprint.
  • And more! 
So yeah, a lot went in behind the scenes, which is all very exciting, I know. 

But, I hope you at least enjoy the final product. I am very happy with how everything turned out and I'm excited to keep iterating. 

But before we get to that, I have one quick thing to get to... My blog! 

So, here is the post in question, which contains all the pitfalls, ups, downs, and loop-de-loops this project has taken me on over the past two weeks! 

Sprint 2 Retrospective

This sprint’s progress was better than I was expecting! I was lacking direction last week, as I was not fully satisfied with the animation, however, with help from Jonathan, I was able to identify key poses and create a better storyboard that followed line-of-action and animation principles. 

Overall, I was able to take a 20+ frame ‘storyboard’ and limit it down to 7 key poses. From there, the goal next week will be to find sufficient in-betweens for the key poses. 

Additional feedback I received during the presentation last week suggested that the animation was too fast. Since I was utilizing a GIF for my animation, the video render looked sped up and was a little too quick. 

Using Adobe Premiere, I was able to render a much more realistic target for my animation. I narrowed it down between 1.75x faster and 2.0x faster. 

Here is 1.75x faster.

And here is 2.0x faster. 

Thanks to feedback from Cheryl and Jonathan, I ultimately decided on 1.75x faster…

…However, I had a problem.

When I adjusted the frame spacing in Maya, it multiplied all the frames by 1.75, which resulted in a lot of funky numbers with decimals, which is never good for animation frames. Worse yet, I had 20+ frames of animation with weird numbers that refused to snap to keys. 

Thus, I decided to start over from the beginning. Yep, all the way from the beginning. I decided to take Jonathan’s advice to heart and really hone in on key poses. And I wanted to start piecing together Chris’s advice regarding center of gravity.

So I went back to the oldest version of my scene and started fresh. And here we are! 

I definitely worked a lot of the hips and line-of-action this week, as they were things that were not present in last week’s animation. In addition, I also tried to keep the head motion natural, as the neck got a lotta bit out of hand last week. 

Another thing I really worked hard on were the arms and hands. During last week’s submission, the hands floated and moved away from the ground and looked really unnatural and artificial. In this week’s rendition of the project, I figured out how to IK/FK switch with the hands, which made it easy for me to plant the hands on the ground. 

I will figure out how to constrain these objects and the ball to the ground later; I just want to get all my ducks in a row, first. Once I have a submission that pleases everyone, I will move forward and get into the more complicated elements of animation (i.e. camera sequencer, graph editor, constraints). I am a firm believer of fundamentals first, so I am going for that approach with this project.

Oh, also, along the way I added the lights and infinity wall, too. Chris and Nick suggested that I pull the lighting up in the schedule, just so I can iterate on it, which I did. Part of the reason for pushing lighting so far back in the schedule is because I struggle with it quite a bit, but I managed for this submission, at the very least.

Here is that lighting setup.

I wanted to put in area lights, but I was having trouble with them over-saturating my camera view, so I swapped them out for low-intensity Directional lights. I set them up in the formation we learned last semester, two angled up front and one behind the character. 

The render in Arnold is a little gloomier than I would want it to be, but I didn’t want my playblasts to be too white, so I decided to tone down the lights for now. It’s weird that Arnold renders out so much darker than the normal camera view. I will have to consult a teacher regarding this because I have no idea why or how it does this. 

The Arnold render in question is below. This is before I added the infinity wall, by the way. 


I also mentioned experimenting a little bit with different tangent types. I uploaded a bunch of videos of these different types and how they impact my animation.

Here's clamped:


Here's auto:


Here's linear: 


Here's flat:


Here's plateau:


And here's stepped, the one that is ideal for storyboarding and finding key poses! 


And this is what my graph editor looks like in stepped tangent mode:


I want to learn when and where to apply these various tangents in animation. I know some are good for bouncing ball motions, while others are better for more natural motion, which is one of my targets for next week-- figuring out how to make the motion the most natural.

Anyways-- that's a story for next week. I've got a PowerPoint to prepare for class tomorrow. 

Before I leave, have my Jira board! Woohoo! (It was too big to fit into the PowerPoint)


That’s all, folks! Thanks so much for tuning in! See you in two weeks! 

No comments:

Post a Comment