Week 6

Mentor’s Feedback Summary

  • FX too chaotic, tone it down

FX Work List

  • Shot 2

    • Re-Sim FX

    • Export with AOV layers

  • Shot 5 *

    • Fix FX to match camera

    • Address clipping issues

    • Optimize renders

    • Export with AOV layers

FX Testing and Outcomes


  • Viewport_02

    • Viewport render before full render.


  • Shot 5*

    • This week for shot 5, I decided that I wasn’t happy with the current state of my file, and so I decided that I needed a full overhaul. I reworked and optimized the file so that I could make iterations quicker and to redesign the FX.

    • The biggest change that I made, was the switch back to a particle workflow instead of a FLIP based one. I made this decision for a few reasons:

      • To speed up my sim times

      • Because I wasn’t getting the benefits of using FLIP simulation (i.e. Viscosity/Surface Tension)

      • To thin out any strands created

    • Another big change is the choice to split the simulation in two, so that I can better art direct the shot.


  • Version 01

    • Initially I worked on matching the previous camera move, before we decided on the final move.


  • Viewport 01

    • I always made sure to do periodical viewport renders to make sure the pacing of my shot worked



  • Viewport_03

  • (Figures Below)

    • (Figure. 1:.)

      • The setup consists of scattering points in a tube volume that fits inside my pint

      • Then I use a bounding box to group the points into their respective ingredients

      • I separate them and merge them back together to create my starting points.

    • (Figure. 3:.)

      • The points are then fed into a pop network, where I create another bounding box group called ACTIVE

      • This makes it so that when the points enter the bbox they are affected by my forces

      • This is what makes the nice layering of the ingredients.

      • There is also a pop kill zone that deletes the points when they leave frame.

    • (Figure. 2:.)

      • After being simulated, the points are cached, and then go to be meshed.

      • For meshing, I use a series of VDB smooths and reshapes to get my desired level of smoothness.

      • The cookies, I just add a copy to points with a low poly sphere.

    • (Figure. 4:.)

      • The truffles are completely separate from the simulation, since I wanted to hand place each truffle to avoid and unwanted collisions

      • They are then transformed and then to animate them, I used a blend shapes node.

Figure. 1:.

Figure. 2:.

Figure. 3:.

Figure. 4:.



Previous
Previous

Week 7

Next
Next

Week 5