About Hems, Simulation and exploding Knickers

I'm a big fan of semi-solid or - even better - solid 3D hems (see the image on the left). No fiddling around with bump, normal or displacement maps, but pure topography, which leads to great rendering results.

But unfortunately - when it comes to simulating pull-off / stripped 3D clothes, both types of hems have their disadvantages.

When a garment (3D or not) falls down, you can usually see not only the front of the hem, but also the back. In this case, a semi-solid hem is not very useful.

The magic self crawling panties

This is a bit more technical. Let's try it: Every 3D object consists of vertices (= "points" in 3D space that form the object). If these vertices are too close together (which is the case when solid hems are used, but also when the 3D clothes have many folds and wrinkles), the simulation algorythm produces nonsense.

An example: The image on the left shows what happens when the simulation algorithm cannot decide where to place adjacent vertices for the next simulation step. This results in a deviation from the normal fall line (red) - the panties appear to creep out of the viewport.

Unfortunately, this happens not only when using solid hems, but also with "painted" hems when the algorithm sets the vertices too close together during simulation. I have even observed this effect when using a straight plane ... or in the QS3 "C" of my Knickers manual.

Caution: exploding Knickers

The next escalation step is exploding garments if the vertices are too close to each other and a bad combination of parameters was chosen.

The picture on the left, which looks like something you'd wear on your head in the shower, is exploded knickers. Indeed, nice and interesting - but not suitable for pull-off / stripped 3D clothes.