Enna Alouette [Project3] - Statistics (Patreon)

This is the statistics of the Enna Alouette Project. It is one of the hardest project that I ever take. Rigging takes an entire week to make sure that the model rig could provide details towards the animation. I took numerous different approach on how I rig the model.

The first one is the clothes rigging. Previously on Isekai Pikamee Project [Project2], I rig the model the same way as I rig the main body. In this project, I managed to use an alternative method, which is Data Transfer Modifier. The way it works is by transferring the weight data of the body into the clothing. The result of this change is truly stunning, as I hardly ever encounter clipping issues between the body and the clothing.

Another approach that I managed to learn is dealing with hair physics. Previously, in [Project2], I use cloth physics to provide hair physics to the model. However, one of the biggest downside is that the model would not be able to maintain its geometry and always get flopped around by the wind or gravity. In this project, I instead use the softbody physics. The way this works is that softbody physics provides object with elasticity which responds to movement, gravity and wind. In addition to that, objects that gets move around would still retain its geometry whilst responding to the environment.

The third technique that I recently learn is eye tracking. In the previous project [Project2], the pupil becomes the child of the eyeball (parenting), and then the eyeball tracks the bones to move accordingly. However, one of the biggest flaw of using this method is that it is not suitable for cartoony/anime eyes. This cause clipping which makes the eyeball pops out from its socket. Instead, I make the pupils to track the bones instead of the eyeball. As a result,one of my problems with rigging have been successfully solved.

In addition to that, one of the problems that I encounter is that I haven't apply subdivision surface to the model. This caused by how a few vertices gets influenced by bones that are far away from the vertex. As such, during rigging, the model could not respond to the skeletons in a good manner and the weight painting process gets tedious. After the application of the subdivision surface modifier, the rig becomes alot more responsive and was able to maintain the geometry without breaking apart the model.

I learn a lot from this project, and I hope I could accomplish better times later on in the future.