Loading...
> >

Dforce Hair Not Settling

IceDragonArtIceDragonArt Posts: 12,740
in The Commons

So when I run a simulation on dforce hair, or hair that I have applied a dforce modifier and weight map to, most of the time, at the end of the simulation, the hair is still floating sometims a little, sometimes a lot.  This happens with both hair that comes with dforce and hair I modify myself.  I've tried lengthening the time and so it has more time to settle, I've upped the gravity to 2, still no love.

Comments

  • richardandtracyrichardandtracy Posts: 7,089
    Try increasing friction. Absorbs energy and slows movement. May work. Regards, Richard.
  • IceDragonArtIceDragonArt Posts: 12,740

    richardandtracy said:

    Try increasing friction. Absorbs energy and slows movement. May work. Regards, Richard.

     

    Thanks!  Where would I find that particular function?  Sorry, I've been away for quite some time and am struggling a bit with the massive relearning curve.

  • richardandtracyrichardandtracy Posts: 7,089

    Sorry, not at my machine, will try to show this evening.

    However, in words, without pictures: The co-efficient for friction is listed in the surface material parameters. Select one of the hair materials in the surfaces tab. At the top are the colour/texture/tiling parameters, and after you scroll down below that you come across the dForce parameters like weight/m^2, bend stiffness, shear stiffness etc. One of them is the co-efficient of friction. I suggest you take the figure that's there (probably 0.2 or 0.4) and double or treble the value to see if there's an effect when you simulate it again. If there is, it may be worth tweaking further up or down to get the result. Real values of the friction co-efficient would be in the range 0.1 (oiled metal on metal) to 0.8 (rubber on asphalt), but in DS you can have large and unrealistic numbers. Won't guarantee that there will be an effect, but there may be. 

    Regards,

    Richard

  • IceDragonArtIceDragonArt Posts: 12,740

    richardandtracy said:

    Sorry, not at my machine, will try to show this evening.

    However, in words, without pictures: The co-efficient for friction is listed in the surface material parameters. Select one of the hair materials in the surfaces tab. At the top are the colour/texture/tiling parameters, and after you scroll down below that you come across the dForce parameters like weight/m^2, bend stiffness, shear stiffness etc. One of them is the co-efficient of friction. I suggest you take the figure that's there (probably 0.2 or 0.4) and double or treble the value to see if there's an effect when you simulate it again. If there is, it may be worth tweaking further up or down to get the result. Real values of the friction co-efficient would be in the range 0.1 (oiled metal on metal) to 0.8 (rubber on asphalt), but in DS you can have large and unrealistic numbers. Won't guarantee that there will be an effect, but there may be. 

    Regards,

    Richard

    Thanks!  I should be able to find that without pictures.  Just wasn't sure where to look.  Appreciate it!

  • richardandtracyrichardandtracy Posts: 7,089
    I have a few images, but have only transferred the images to a stick - should upload in 12 hours or so. Also look for damping, dynamic damping, shear & bend damping. Regards, Richard.
  • IceDragonArtIceDragonArt Posts: 12,740

    richardandtracy said:

    I have a few images, but have only transferred the images to a stick - should upload in 12 hours or so. Also look for damping, dynamic damping, shear & bend damping. Regards, Richard.

    Thank you so much!  I will play around with this as soon as I get the chance.

  • richardandtracyrichardandtracy Posts: 7,089
    edited October 30

    The settings I think are pertinent are shown below using the example of the Linday 'Classic Long Wavy Hair'. Obviously, really only applicable to dForce fabric engine hair rather than the confusingly named dForce strand hair. If you're using dForce strand hair then I'm utterly lost with it.

    And after simulation (which didn't completely settle, but had very small movements between updates, I admit):

    [As an aside, I still can't get over how fast this all simulates. I'm used to doing Large Deflection Finite Element Analysis for structures - which is effectively what this is - and a 150,000 element/facet 3D surface model like this would take several hours and hundreds of iterations, but in my RTX3060 GPU, this takes under 3 minutes. Completely mind boggling.]

    Regards,

    Richard

    Hair1.jpg
    1036 x 789 - 286K
    Hair2.jpg
    469 x 455 - 81K
    Hair3.jpg
    1245 x 839 - 262K
    Post edited by richardandtracy on
  • IceDragonArtIceDragonArt Posts: 12,740

    Thank you very much!

Sign In or Register to comment.