Click here to go back to studio 2


Click here for Studio 1 - Eppies - 11/15/08 - Week 9


Click here for Studio 1 - Eppies - 11/8/08 - Week 8


Studio 1 - Eppies - 11/2/08 - Week 7


This week I've been working on finalizing the dynamics setup. It was taking too long to solve (almost 8mins per frame), but now I've got it down to around 1.2mins per frame. I also fixed the normal rotation like this:

timeblend1 freezes the initial pose at frame 1. Then timeblend2 takes that frozen single frame and stretches it till the end of the timeline. Then the attribute reorient sop manipulates the normals.

mooseface1.htm

I also got the fur to bend and frizz based on normal direction. So if the fur wire directly faces the camera (allowing us to see the skin), the surface color will change. This allows me to take this attribute into the fur cvex shaders to manipulate frizz, bend, density, etc. It's all based on a pretty simple vex setup. Here it is calculating cd:

I also started to finalize the dynamics setup. As you can see here, each wire is individually affected, but they are too loose, and spin all over. I've spent hours tweaking kangular and damping to no avail. I don't think it's correctly accounting for the hair length. Looks like i might have to create another paintable attrib...

There were so many problems I had to solve as far as the dynamics go. The simulated hairs were freaking out... spinning and stretching all over the place. I managed to control it by extending roots by 1 point. I also made another attribute that controls how many points are added to roots based on RGB. So we can exclude very short hairs, like the feet, from the dynamic simulation. This had to be done as the wire solver didnt seem to properly account for length...


No more pretty renders this week, as there seems to be a problem with the renderfarm's houdini licenses.