Dynamic cloth under stress



  • I made a new dress to test dynamics boundaries. I think Poser did a very, very good job.
    0_1530981439944_R-DressN.png



  • Some more details:
    0_1530981530896_R-TrousersD.png

    0_1530981543870_R-TrousersE.png



  • Well done indeed!



  • @F_Verbaas said in Dynamic cloth under stress:

    Well done indeed!

    Thanks!
    Yes, Posers cloth room isn't that bad. If the mesh is appropriate (means: made with dynamics in mind and not just roughly converted from something) and defined dynamic groups matchs how the cloth should behave.

    I used 100 Frames, movement of the figure stops at frame 55. The rest is setteling.
    Most frames needed less then 10 seconds to compute, some only 1-3 seconds, others up to 30 (computed with an older I5 laptop). My Desktop machine should do it a lot faster.



  • Oooh that looks NICE! <3



  • @adp have you done much constraint-wise? Like top of shoulder straps and waist-band, or is it all dynamic? Is the button just a hard decoration, or have you modeled actual button holes and threads? Did you subdivide the mesh, or is it just sufficiently detailed already?

    This really is well done!



  • You know, the cloth room would actually be a good substitute for rigging a suit of armor. Just make almost everything rigid. After all, only the joints need to bend.


  • Poser Ambassadors

    Agree, Poser's cloth room is underestimated. I did a stress test with my freebie pants .They are hybrid, but even there was just a minor postwork necessary .



  • @eclark1849 said in Dynamic cloth under stress:

    You know, the cloth room would actually be a good substitute for rigging a suit of armor. Just make almost everything rigid. After all, only the joints need to bend.

    Actually, after thinking about it, Armor, doesn't really need the Cloth room or rigging at all. Again, allowing for joint movement, a decent suit of armor could simply be smart propped to a figure. On the other hand though, chain mail and links of chain would definitely be a good candidate for the Cloth Room.



  • I think it's kind of funny that when I started with Poser 8, I didn't know anything about conformers and how they worked, so I went to the cloth room almost immediately. I don't find it daunting at all. I've had many successes and a few absolute failures in there, don't get me wrong. But overall, I love the cloth room and what I can do in there.



  • Now that I think about it, I started using the Cloth Room with Poser 6. I had skipped P5, so P6 was the first version I ever used the Cloth Room in. And if I recall correctly it was the one that came with a Cloth Room tutorial. I may even still have it on my Mac.



  • @anomalaus said in Dynamic cloth under stress:

    @adp have you done much constraint-wise? Like top of shoulder straps and waist-band, or is it all dynamic?

    All dynamic. Except a single row of vertices around the hip from the trousers defined as riggid. But this is only needed for extrem poses, to prevent that the back is pulled down (like real cloth tries to do).

    Is the button just a hard decoration, or have you modeled actual button holes and threads?

    There are 3 Buttons. Defined as riggid decorated. And there is a "real" buttonborder. the Buttons pierces the two borders and hold the thing together. If the buttons are removed, the dress falls appart at the front.

    To move the buttons, I used a magnet with vertexmap active. I simply added the buttons material group to the vertexmap and can now move the buttons where I need them (even move them out of the way).

    Did you subdivide the mesh, or is it just sufficiently detailed already?

    The mesh is made to bend and fold. So it is relatively detailed and has a workflow to support folding in the crtical sections.
    A fold can't be smaller as a polygone, because the dyn-engine can't split a polygon. If they are to big the engine can't release from the stress, what results often in a poke-through. More stress from frame to frame, until it gets broken.

    For rendering, the mesh is subdivided one step further (to make fold-edges rounder).

    [Sorry for my bad english.]

    This really is well done!

    Thanks!



  • @Ladonna Just for the records: I did no postwork. Not with the morphbrush, not on the rendered image. Anything shown is what Poser has produced all alone :).



  • @anomalaus said in Dynamic cloth under stress:

    @adp have you done much constraint-wise? Like top of shoulder straps and waist-band, or is it all dynamic? Is the button just a hard decoration, or have you modeled actual button holes and threads?

    A bit more about this. Each clothpiece has a border (hem?) around all openings. This border simulates thickness.
    The innermost vertices of that border are defined as soft-decorated dynamic group. the outer vertices of that border are another dynamic group. The parameters for this group makes it a bit harder, stiffer as the rest.The dyn-engine obeys that and a real good looking hem/border is the result.



  • Here are some details to show mesh density.
    0_1531050777724_Bildschirmfoto vom 2018-07-08 13-47-47.png
    As you can see, the density of the dress is not mutch higher than (relatively low-res) Roxie's. And you can see the buttonborder connected with the 3 buttons (defined as rigid-decorated) holding anything together.

    0_1531050948784_Bildschirmfoto vom 2018-07-08 13-50-55.png
    The trousers are a bit higher in density because they are tighter and has to support smaller folds.



  • @eclark1849 said in Dynamic cloth under stress:

    Now that I think about it, I started using the Cloth Room with Poser 6. I had skipped P5, so P6 was the first version I ever used the Cloth Room in. And if I recall correctly it was the one that came with a Cloth Room tutorial. I may even still have it on my Mac.

    I can't remember since when I use the cloth room. But my first dynamic freebee at ShareCG is now 11 years old :)



  • Looks like I was wrong. The Tutorial was in Poser 7.



  • Another screenshot from the front.
    In the final version, I will add another one or two polygonrows to the buttonborder. To make sure the buttons have more "grip". At the moment, if the fabric has to less weight the buttons aren't able to hold it together.
    0_1531055233048_Bildschirmfoto vom 2018-07-08 15-05-22.png
    I'll upload the whole thing to ShareCG if I'm satisfied. May last a few days, because I'm busy with a Python script to export/import a figure as unimesh OBJ (as long as vertexcount stays unchanged) :)



  • The breast area has a higher density to support big ... I think you now why. Drawback here: the dyn-engine assumes that a higher poly-density means higher fabric-weight. Anytthing is good as long as this area is "stretched". With standard breastsize the fabric tends to sag in this area.



  • @adp
    I do not think it is a weight issue. It looks more like there is more shape in the mesh in the area than the figure can fill.
    If you assume shape is made by stretch, (in real life fabric this usually means ther are no darts or figure-seams) best start from a fit for a flat-chested figure and let the breasts 'grow in' by morph.