How to Use dForce: Creating a Blanket, Draping Clothes on Furniture, and Much More [Commercial]

1141517192037

Comments

  • sapatsapat Posts: 1,735
    edited June 2018

    Wow, that's genius.  Gonna try it right now.

    Edited to ask: With the Torus, what do you use for the Minor Diameter setting? I left it at the default of 30" but it looks like a giant bagel that fills the entire viewport.

     

    minor diameter setting for torus.JPG
    440 x 468 - 29K
    giant torus bagel if left at 30 inches minor diameter.JPG
    704 x 647 - 30K
    Post edited by sapat on
  • RGcincyRGcincy Posts: 2,695

    I went back and edited a couple of my posts to put them in a format more similar to the previous sections and added them to the index on page 1. When I add them to the pdf, I'll likely flesh them out a bit more. By this point, some of the details I routinely included in early sections may not be as needed for those who have been following along.

  • sapatsapat Posts: 1,735
    RGcincy said:

    I went back and edited a couple of my posts to put them in a format more similar to the previous sections and added them to the index on page 1. When I add them to the pdf, I'll likely flesh them out a bit more. By this point, some of the details I routinely included in early sections may not be as needed for those who have been following along.

    I've been folllowing along, but have no means been able to do a lot of them due to real life, so I'd rely on those details. I'm a slow learner due to an illness and it takes longer for me to grasp things.  I'd love to see them fleshed out as you say, but would it be extra work to leave those details in there?   Sorry if this is something I shouldn't ask given all you've done for us.

  • MadaMada Posts: 1,317

    I love seeing all the clever ways you all come up with to use dForce :D very cool

  • RGcincyRGcincy Posts: 2,695
    sapat said:

    With the Torus, what do you use for the Minor Diameter setting? I left it at the default of 30" but it looks like a giant bagel that fills the entire viewport.

    I used 4 inch for major diameter and 0.25 inch for minor (or you can think of it as the minor being 1/16th the major if you are making it larger or smaller.

     

    sapat said:

    I've been following along, but have no means been able to do a lot of them due to real life, so I'd rely on those details. I'm a slow learner due to an illness and it takes longer for me to grasp things.  I'd love to see them fleshed out as you say, but would it be extra work to leave those details in there?   Sorry if this is something I shouldn't ask given all you've done for us.

    Actually it's helpful to know the details are useful. It takes a bit more time to write but I don't mind. 

     

    Mada said:

    I love seeing all the clever ways you all come up with to use dForce :D very cool

    Thanks! dForce turns out to be able to do more than you originally think it can.

  • jardinejardine Posts: 1,138

    i'm just starting to get into this.  but one thing i'm already *really* wishing for is some sort of central command module (or module hierachy) in the primary dforce panel.

    'include in simulation' on/off should be at the top there, on top of a hiearchal list with the old familar show-hide eyeball appearing alongside each item.  having to enable or disable each item's visibility to dforce in the individual item's parameter tabs is a royal pain.  there should also be some flagging involved, so that items which are not things that would be likely to have any dforce properties do not appear.

    then some interface for adjusting the layering of the items that are interacting, and their individual dforce simulation settings. 

    part of that a)  dforce at this point seems to be a bit slack about figuring out what is supposed to be on top of what.  it's kind of the like the basic studio principle you wind up figuring out eventually.  put your pants on before you put your shirt on, if you don't want to have bits of your shirt trapped weirdly in your pants.  do it the other way 'round, and you won't have any problems.  but kitbashing with dforce, for whatever reason, dforce items from different products with different default settings can give each other some really strange wedgies.  i've had a top and pair of pants dforce themselves together so devotedly that you couldn't even tell that they were separate objects. 

    part of that b)  some simple slider + numeric value strength input to determine how strongly the dforce values will be applied to the other objects/physics layers in the scene.  perhaps you want your top and long flowing sleeves, responding to windforce, to glide very lightly (0.33) along your wide belt, and interact with your pants lightly (0.48).  apart from being great for wrangling clothing, this would be a godsend for dealing with dforce hair and/or dforced hair.  

    :)

    i'm not complaining in any way.  dforce is pretty awesome.  and the PA's whose dforce clothing i've experimented with this past week seem to really know what they were doing.  trying to figure out how they've worked the magic is probably going to be about as tricky as trying to figure out why their surfaces always seem so wonderful. 

    but that's kind of the thing.  i don't really want to have to mess around with excellent default dforce simulation presets to get my kitbashery to work.  i want to be able to leave those alone, or maybe finetune them a little, and then--in a separate interface--try to figure out how to make my physics layers get along. 

    randomly, some script or series of scripts that could weld wayward vertices in pre-dforce clothing (or at least approximate that enough to keep dforce-insecure clothing from falling apart) without losing JCMs would be awesome. 

    :)

    j

  • MelanieLMelanieL Posts: 6,412

    That sphere-and-torus experiment worked out really well - this thread continues to amaze!

  • OdaaOdaa Posts: 1,540
    jardine said:

    i'm just starting to get into this.  but one thing i'm already *really* wishing for is some sort of central command module (or module hierachy) in the primary dforce panel.

    'include in simulation' on/off should be at the top there, on top of a hiearchal list with the old familar show-hide eyeball appearing alongside each item.  having to enable or disable each item's visibility to dforce in the individual item's parameter tabs is a royal pain.  there should also be some flagging involved, so that items which are not things that would be likely to have any dforce properties do not appear.

    then some interface for adjusting the layering of the items that are interacting, and their individual dforce simulation settings. 

    part of that a)  dforce at this point seems to be a bit slack about figuring out what is supposed to be on top of what.  it's kind of the like the basic studio principle you wind up figuring out eventually.  put your pants on before you put your shirt on, if you don't want to have bits of your shirt trapped weirdly in your pants.  do it the other way 'round, and you won't have any problems.  but kitbashing with dforce, for whatever reason, dforce items from different products with different default settings can give each other some really strange wedgies.  i've had a top and pair of pants dforce themselves together so devotedly that you couldn't even tell that they were separate objects. 

    part of that b)  some simple slider + numeric value strength input to determine how strongly the dforce values will be applied to the other objects/physics layers in the scene.  perhaps you want your top and long flowing sleeves, responding to windforce, to glide very lightly (0.33) along your wide belt, and interact with your pants lightly (0.48). 

    Select all objects set as dforce. Go to surfaces tab. select the surfaces you want to be less affected by the simulation. Scroll down the surfaces tab until you come to a button labeled "dforce simulation on/off) with a series of dials/sliders on a different colored background. "Reduce Simulation Strength" is like the second or third option in that list.

  • maikdeckermaikdecker Posts: 2,118
    RGcincy said:
    Odaa said:
    Or is it just going to freak out and try to resimulate from scratch on frame 26-30, and lose all the work you've already done?

    It freaks out.

    You could animate to the end, hide the cylinders and export the plane as an object, then reimport as a morph and simulate from there. A bit of work though.

    You can reduce the effect of the cylinder a bit though by reducing it's scale to 0.0001% so most of the plane isn't connected to it anymore... haven't tried it out excessively, but it worked on the few tries I did

  • sapatsapat Posts: 1,735

    Ok, it appears I am clueless.  I can't figure out how to move the sphere up, over and down.  I don't know how many 'clicks' I need to do on the sphere for each frame of the animation from beginning to end.  Do I click 1 frame and move the sphere up a bit, then advance a frame and move it up again?  Or do I click and move the sphere up several clicks, advace another frame and do that over and over til the sphere reaches the positon you have it?   I tried to follow along and have my ball positioned in the spots that your progressive images showed.  I thought I could do this but I've never created an animation, just ran them.  But I figured 'how hard can it be?'  I guess my slow brain needs those details!

    Then after clicking to move the sphere up and over on the Ytrans advancing frames til I got to 30, the sphere looked like it was in pretty much the same position as yours, so I ran it, and the plane took a nosedive down and disappeared below the viewport. I had the plane, sphere and torus positioned (although I don't know how far up on Y the torus is).  But they all had the correct parameter settings.

    Help Rich!

  • OdaaOdaa Posts: 1,540
    edited June 2018

     

    sapat said:

    Ok, it appears I am clueless.  I can't figure out how to move the sphere up, over and down.  I don't know how many 'clicks' I need to do on the sphere for each frame of the animation from beginning to end.  Do I click 1 frame and move the sphere up a bit, then advance a frame and move it up again?  Or do I click and move the sphere up several clicks, advace another frame and do that over and over til the sphere reaches the positon you have it?   I tried to follow along and have my ball positioned in the spots that your progressive images showed.  I thought I could do this but I've never created an animation, just ran them.  But I figured 'how hard can it be?'  I guess my slow brain needs those details!

    Then after clicking to move the sphere up and over on the Ytrans advancing frames til I got to 30, the sphere looked like it was in pretty much the same position as yours, so I ran it, and the plane took a nosedive down and disappeared below the viewport. I had the plane, sphere and torus positioned (although I don't know how far up on Y the torus is).  But they all had the correct parameter settings.

    Help Rich!

    Use the frame 4 references Rich shows as a guideline. At Frame 4, y-translate the sphere upward until it is in line with the hole in the torus. Add a keyframe (on the basic timeline, there's an icon at bottom far-right with a key and a plus sign inside its handle-use that). At frame 10, move it closer to the Torus, add keyframe. At frame 24 pull it through the torus, add keyframe and so on. The timeline will extrapolate the movements for the "in-between" frames on its own, as long as the keyframes are set correctly.

    Post edited by Odaa on
  • RGcincyRGcincy Posts: 2,695

    Odaa's explanation covers it. The  frame numbers in the image names are almost always where I make a change in position. So frame 0 has everything in the starting position, frame 4 moves the sphere up, etc.

    In terms of the plane taking a nosedive, that's likely from using either a plane and/or a sphere with too few segments. How much dForce holds onto depends upon the objects resolution. So be sure the plane has 50-100 divisions and the sphere has 16 segments and 16 sides or more. Be sure the sphere is embedded in the plane and not sitting above it. 

  • jardinejardine Posts: 1,138
    Odaa said:
    jardine said:

    i'm just starting to get into this.  but one thing i'm already *really* wishing for is some sort of central command module (or module hierachy) in the primary dforce panel.

    'include in simulation' on/off should be at the top there, on top of a hiearchal list with the old familar show-hide eyeball appearing alongside each item.  having to enable or disable each item's visibility to dforce in the individual item's parameter tabs is a royal pain.  there should also be some flagging involved, so that items which are not things that would be likely to have any dforce properties do not appear.

    then some interface for adjusting the layering of the items that are interacting, and their individual dforce simulation settings. 

    part of that a)  dforce at this point seems to be a bit slack about figuring out what is supposed to be on top of what.  it's kind of the like the basic studio principle you wind up figuring out eventually.  put your pants on before you put your shirt on, if you don't want to have bits of your shirt trapped weirdly in your pants.  do it the other way 'round, and you won't have any problems.  but kitbashing with dforce, for whatever reason, dforce items from different products with different default settings can give each other some really strange wedgies.  i've had a top and pair of pants dforce themselves together so devotedly that you couldn't even tell that they were separate objects. 

    part of that b)  some simple slider + numeric value strength input to determine how strongly the dforce values will be applied to the other objects/physics layers in the scene.  perhaps you want your top and long flowing sleeves, responding to windforce, to glide very lightly (0.33) along your wide belt, and interact with your pants lightly (0.48). 

    Select all objects set as dforce. Go to surfaces tab. select the surfaces you want to be less affected by the simulation. Scroll down the surfaces tab until you come to a button labeled "dforce simulation on/off) with a series of dials/sliders on a different colored background. "Reduce Simulation Strength" is like the second or third option in that list.

    thank you, Odaa...

    that's very helpful.  :)

    j

  • sapatsapat Posts: 1,735

    I hope I can figure that out.  i've never done an animation, so don't know what key frames are.  I see the little button to add one, but don't know any more than that.  Guess I'm in over my head.  I'll try it, but I'll have to find a tutorial that explains the timeline and key frames.  What you're saying then it kind of works like 'Tween' in Photoshop/Image Ready?  You put your object at start postion, then say in the middle and then where you want it to end up. Then hit the play button.

     

    RGcincy said:

    Odaa's explanation covers it. The  frame numbers in the image names are almost always where I make a change in position. So frame 0 has everything in the starting position, frame 4 moves the sphere up, etc.

    In terms of the plane taking a nosedive, that's likely from using either a plane and/or a sphere with too few segments. How much dForce holds onto depends upon the objects resolution. So be sure the plane has 50-100 divisions and the sphere has 16 segments and 16 sides or more. Be sure the sphere is embedded in the plane and not sitting above it. 

    I entered all the parameters you had in your tutorial, but it dropped like a rock.  I'll try again.

    Thanks for the kind help Odaa and Rich.

  • N-RArtsN-RArts Posts: 1,191
    RGcincy said:

    It looks like part of the bandeau may be intersecting the skin at the start (or during animation) and so is getting locked into place. To test if that's the case, increase the Z or X scale a bit to be sure all is clearing the skin. If that's not a problem, you can try a lower Dynamic Strength (like 0.95 or 0.97) on the middle band material.

    I have tried adjusting the Z and X scales, but I'm still getting the same results. Dynamic Strength didn't make much of a difference either. The only thing that has made a difference is turning off Self Collide on the default and lower bands. If I alter too many settings, both the simulation and Daz become unresponsive (then I have to restart the computer frown ). 

     

     

    better.png
    413 x 492 - 130K
  • OdaaOdaa Posts: 1,540
    sapat said:

    I hope I can figure that out.  i've never done an animation, so don't know what key frames are.  I see the little button to add one, but don't know any more than that.  Guess I'm in over my head.  I'll try it, but I'll have to find a tutorial that explains the timeline and key frames.  What you're saying then it kind of works like 'Tween' in Photoshop/Image Ready?  You put your object at start postion, then say in the middle and then where you want it to end up. Then hit the play button.

     

    RGcincy said:

     

    I don't use Photoshop, but yes, that's how the timeline works. Here's an abstract explanation of how key frames work in general: https://en.wikipedia.org/wiki/Key_frame

  • RGcincyRGcincy Posts: 2,695
    edited June 2018
    N_R Arts said:

    I have tried adjusting the Z and X scales, but I'm still getting the same results. Dynamic Strength didn't make much of a difference either. The only thing that has made a difference is turning off Self Collide on the default and lower bands. If I alter too many settings, both the simulation and Daz become unresponsive (then I have to restart the computer frown ). 

    Two things you can try:

    (1) Add a smoothing modifier (Edit/Object/Geometry/Add Smoothing Modifier). You can set number of iterations in the parameters plane.

    (2) Add a push modifier (Create/New Push Modifier Weight Node). After adding the map, set it to a positive value (may need to be higher than 1) and use the weight node map tool to paint over the areas sticking into the skin to raise them above.

    (I cover how to use smoothing and push modifiers several places in this thread. If you can't find them, let me know and I'll point them out.)

    Post edited by RGcincy on
  • RGcincyRGcincy Posts: 2,695
    sapat said:

    I hope I can figure that out.  i've never done an animation, so don't know what key frames are.  I see the little button to add one, but don't know any more than that.  Guess I'm in over my head.  I'll try it, but I'll have to find a tutorial that explains the timeline and key frames.  What you're saying then it kind of works like 'Tween' in Photoshop/Image Ready?  You put your object at start position, then say in the middle and then where you want it to end up. Then hit the play button.

    This post has links to two older pages that describe the DS timeline controls. Once you understand the basics, it's pretty straightforward for simple animations.

    • Open the timeline pane
    • Click the key icon
    • Move the slider along the ruler to frame 15
    • Go to the scene pane and select an object
    • Move the object by dragging it to a new position or by using the parameter dials
    • Go to back to the timeline pane and move the slider to frame 30
    • Reposition the same object
    • Hit the play button and you'll see the object move.
  • sapatsapat Posts: 1,735

    I selected the sphere, moved it as shown in your images adding key frames at the places where you have them. Then in render settings I set it to Image Series.  When I played the animation, all that moved was the sphere.  The plane stayed flat on the ground. I don't understand. How do you get the plane to come along with the sphere?

     

    plane not moving with sphere.JPG
    747 x 660 - 70K
  • RGcincyRGcincy Posts: 2,695

    It looks like the right setup. After creating the animation, did you then run the dForce simulation using the Animated (Use Timeline Play Range) on the Duration tab of the Simulation Settings pane? That will create the draping movement.

  • sapatsapat Posts: 1,735

    Yes I did.

     

    set up for animated with timeline.JPG
    414 x 347 - 36K
  • RGcincyRGcincy Posts: 2,695

    I'm out of ideas. If you PM me your .duf scene file, I'll take a look and see what's going on.

  • sapatsapat Posts: 1,735

    Yes I did.

     

    RGcincy said:

    I'm out of ideas. If you PM me your .duf scene file, I'll take a look and see what's going on.

    Ok, I'm really sorry. I don't know either.

  • 3dOutlaw3dOutlaw Posts: 2,452

    Is there a way to tell in a scene, which items have a dForce modifier?  Now with many including it, when I hit simulate, I am expecting to see one thing, but at times other things occur as well.

  • RGcincyRGcincy Posts: 2,695
    3dOutlaw said:

    Is there a way to tell in a scene, which items have a dForce modifier?  Now with many including it, when I hit simulate, I am expecting to see one thing, but at times other things occur as well.

    I'm not aware of an easy way, but each object in the scene with a dForce modifier will have an extra property group in the Parameters pane titled "simulation".

  • RGcincyRGcincy Posts: 2,695
    edited June 2018

    I thought it might be helpful to see the animation for the shawl and the cloth through the torus. Here are two gif's:

     

    dForce shawl.gif
    480 x 384 - 3M
    dForce cloth thru torus.gif
    480 x 384 - 2M
    Post edited by RGcincy on
  • RGcincyRGcincy Posts: 2,695

    I updated the pdf for Part 2. It now covers through section 48. Part 1 was not changed. The link can be found at the top of the first post.

    Both sections together are 172 pages. This is getting to be book size!

  • sapatsapat Posts: 1,735
    RGcincy said:

    I updated the pdf for Part 2. It now covers through section 48. Part 1 was not changed. The link can be found at the top of the first post.

    Both sections together are 172 pages. This is getting to be book size!

    Thanks for the UD Rich. It sure is a book!

  • RGcincyRGcincy Posts: 2,695
    edited June 2018

     

    48. Cylinder Rigidity. You can use a long cylinder as a stand-in for a rope, chain, or other object; however, I have found they tend to collapse in on themselves.  This was a study of how resolution and buckling stiffness and ratio can overcome this.

    a. Create a long thin cylinder 3 feet long and 3 inches in diameter. Set axis to X-positive and use 10 segments and 8 sides.

    b. Create 3 more cylinders, this time using 12, 16 and 32 sides.

    c. Apply a dynamic dForce modifier to each cylinder.

    d. Create a plane to act as a collision object.

    e. Line up the cylinders at an angle so one end hits the ground first.

    f. Run a simulation with the default surface properties. You’ll find the eight-sided cylinder retains its shape, the 12-sided shows slight flattening in the middle, the 16-sided shows a lot of collapse, and the 32-sided collapses completely.

    g. Use the following settings for the Simulation surface properties. These were determined by trial and error to minimize collapse and not have the object fly away.

    Sides

    Buckling Stiffness

    Buckling Ratio

    8

    5% (default)

    70% (default)

    12

    35%

    70%

    16

    62%

    0.25%

    32

    10%

    0.1%

    With these new surface settings, the 12-sided retains its shape and the 16-sided has only slight flattening. Nothing helped the 32-sided cylinder.

    h. Create two more 12-sided cylinders but with 20 segments instead of 10. For one of these cylinders, set Y-scale to 200%. Position them parallel to the 12-sided cylinder created in step b. Use default simulation surface properties for all three.

    i. Run the simulation. You’ll find some flattening for all three cylinders.

    j. Use the following settings for the Simulation surface properties. These were determined by trial and error to minimize collapse.

    Segments - Yscale

    Buckling Stiffness

    Buckling Ratio

    10 – 100%

    35%

    70% (default)

    20 – 100%

    50%

    70%

    20 – 200%

    35%

    70%

    All 3 retain their shape. When there was more sides in the same length, we needed to increase the buckling stiffness. But if we increased the length so the segments were of the same size (10 segments and 20 segments with 200% Y-scale) they can use the same value.

    k. Create a small cube and change its scaling to be a beam. Position it so the cylinders fall down across it. As you can see, the 8-, 12-, and 16-sided cylinders mostly retain their shape while the 32-sided cylinder collapses across the beam and into itself. Fine-tuning of the buckling stiffness and ratio would likely help the 16-sided cylinder which has some flattening.

    Doing the same with the three 12-sided cylinders.

    In conclusion, if you want a rope or chain to not flatten in on itself, it’s best to use fewer sides.

    dForce cylinders 4 resolutions default surfaces.jpg
    524 x 368 - 104K
    dForce cylinders 4 resolutions.jpg
    500 x 360 - 97K
    dForce cylinders 12 sided default surfaces.jpg
    500 x 298 - 76K
    dForce cylinders 12 sided.jpg
    461 x 285 - 68K
    dForce cylinders 4 resolutions across box.jpg
    515 x 414 - 104K
    dForce cylinders 12 sided across box.jpg
    504 x 374 - 77K
    Post edited by RGcincy on
  • DanaTADanaTA Posts: 12,353

    Is the Part 2 PDF finalized, or will you be adding more to it?  Or...will you instead be starting on a Part 3?

    Dana

Sign In or Register to comment.