hi,

Good looking result. I'm a bit confused by the numbers. Offset in MrSkinner is in meters? Poser Collission offset and depth are in cm.

Oh I see, you wrote in the last paragraph: ... really be twice the collission depth.

You mean: more than twice the collision offset. Didnt you?

Does the belt has a thickness? Collission depth should better be less than half that thickness (upper limit). Without a limit, anything goes.
My hint was to start collission depth at the offset, and double it once or twice. So I don't really understand the extremely small 0,001 = 1/100 of a mm.

Because... as you take 16 steps per frame and 1 frame = 1/30 of a second, 1 step = 1/480th of a second and during that time any vertex should not move more than that 1/100 of a mm. In other words, the max speed of a vertex, pushed inwards while the belt is shrinking in, is limited to 4.8mm/sec. So when the belt shrinking starts say 4.8 cm from the body, it takes say 10 sec = 300 frames + some extra for settling. And I'm sure you did not do that.

Quote - Poser Collission offset and depth are in cm.

Ah. I did not know that. Offset in Mr. Skinner is in Poser Native Units, I think, so my 0.001 Mr. Skinner belt will be 0.262cm, and the 0.0025 one will be 0.655cm. That implies that with the 0.0025 belt, the collison offset should be 0.3 or less.

Yeah, I got mixed up with offset and depth. :) 

Quote - Does the belt has a thickness?

No, it's single sided. The idea was to use this belt only for the simulation, then replace it with something better afterwards.

Even so, there's a limit to how much I can increase collision depth. I'm running tests now with Collision offset = 0.2 on belt and figure. I started Collision depth at 0.2 and doubled it for each test. Going beyond two doublings, where collision depth = 0.8, seems to make results worse.

It's getting late however, so I'll continue this tomorrow, I hope. 

I used to use the cloth room fairly successfully in earlier versions of Poser, but in Poser 11, I go through all the necessary steps for a successful simulation. It goes through the simulation alright, but then returns back to it's initial form as soon as the simulation is finished. I am using a hi-res square from Primitives and draping it over a V4 and also plain old Andi. Same results. Have I suddenly become just plain ol' stupid, or can anyone explain what's happening? I have followed step by step YouTube Poser tutorials, but all for naught. It returns back to plain old flat primitive hi-res square when simulation has finished all 30 frames.

Thanks in advance for any help. I am at the ends of whatever wits I have left.

You need to have saved the scene before running the simulation.

Is your cloth plane "Locked"?

main menu / object / lock object

The cloth room is just too weird for me, Menu -> Figure -> Conform To... this works for me just fine.

The Conform mechanism is just too unreliable for me.

And too unrealistic for my liking, particularly pants and skirts.

So, this is where we've come over the years: many still prefer the "ease-of-use" of conforming cloth.

Fair enough.

I'm lazy, and prefer letting the computer do all the hard yards of actually letting cloth drape on a figure, as it does in real life. So, what I've been doing is using Poser in its name-sake role: to pose figures, and (roughly) pose the cloth. Then, I bring that figure and cloth into Blender, and let it finish (finesse) the job. Blender drapes like nobody's business: it has tons of parameter settings for ... you name it. And just like with the cloth room, once you get your head around it, it's slick as owl-poo to use.

@RobynsVeil: I am with you there. Conforming just does not do the trick for end result unless it is for skin-tight underwear. I am even more lazy and use Marvelous Designer as simulation engine. This allows me to use the original definition with flat cloth state for strain zero and original fabric properties definition.

Yes, I know about the MD startup price tag and I sincerely hope someone in the Blender community will pick up the bridging possibilities:

• Fabric propeties and fold lines come in the metadata files (.xml)
• Fabric strain per facet is in embedded in the geometry as exported.

This should be enough to set all the garment related part of the simulation parameters in Blender.