Daz 3D is part of 
Connect
DAZ Productions, Inc.
7533 S Center View Ct #4664
West Jordan, UT 84084
Licensing Agreement | Terms of Service | Privacy Policy | EULA
© 2024 Daz Productions Inc. All Rights Reserved.
Comments
Out of curiosity, has anyone tried plastic bags with dForce? A product I'm working on could have various objects in a plastic bag - think vegetables inside a supermarket bag, one of the small clear ones. I can make a bag easily enough, but I have no idea what effect dForce would have on it. I'm sort of hoping that if I had a bag filled with vegetables, I could get a 'clinging' effect. Is that likely? And would I need to 'lump' the vegetables together as one object, or could they be separate?
I don't have enough experience with dForce to guess at the results, or even know if it's possible t get what I want.
TIA
mac
@maclean I haven't but I'll give it a try. I have some fruit and veggie props so I'll see what I can find in my runtime to use as a bag prop.
First attempt at a bag of fruit. I found a trash bag in Trash Stash. I modified the shader so it was a gray translucent color. I used fruit from MHE: Fruit. Part of the fruit is one prop containing multiple fruits, and then I added in individual fruit items. I applied a dForce dynamic modifier to the trash bag and positioned the fruit in the middle. I applied a weight map to the bag and reduced the influence at the knot so the bag would drape from there. Good news was the bag did simulate without crashing (I have found not all props will do that) and the bag deformed smoothly (sometimes how the mesh was constructed interferes with draping). I used dForce surface properties.
The first image shows the starting point and the second shows after the simulation. Overall it worked although I think it could be made better with some tweaking. I wouldn't say it has a clinging effect yet.
A second try. Same setup as before, but during the simulation I scaled the fruit up and moved it down which made the bag wrap around some of the outside fruit and also moved the fruit closer to the bottom of the bag. Since the polygons in the bag are large, it leads to some squarish corners, so I also tested adding 2 levels of subD which smoothed it out (1 level did as well).
After simulation:
After applying subD:
Here I swapped in a mesh shader and enlarged the fruit a second time. You can see there was polygon stretching given how the mesh looks around some of the lower apples.
Wow! That was fast! And it answers my question perfectly - it can be done. I'm going to make some different bags and try this.
Interesting about the mesh effect. That was something I'd been thinking of (a bag of onions or oranges).
Thanks so much for trying that out!
mac
Glad to be of help. It went faster than I expected too (that doesn't always happen, I can assure you). I'll write up some details on how I did it for everyone's benefit.
24. Effect of Surface Properties Using Simple Sheet Drop – Part 2. This continues the discussion from Part 1. Instead of resetting the 6 scene surface properties to default one at a time, one default surface property is set to the original scene’s properties. This is another way to see how surface properties affect the draping. After looking at parameters one by one, several are combined at once to show interactive effects.
a. Default values
b. Stretch Stiffness (0.01 vs. default of 0.8)
c. Shear Stiffness (0.01 vs. default of 0.2)
d. Bend Stiffness (0.01 vs. default of 0.5)
e. Buckling Stiffness (2% vs. default of 5%)
f. Buckling Ratio (98% vs. default of 70%)
g. Density (80 vs. default of 180)
h. Stretch, Shear , and Bend Stiffness at 0.01
i. Buckling Stiffness at 2% and Buckling Ratio at 98%
j. All 6 original scene parameters
k. Conclusions. It’s easiest to see differences if you open each image in a separate browser window. You can then flip through them to see what is happening.
Part 1 of 3
25. Effect of Surface Properties Using dForce Primitive Flag Test. The primitive flag test is a scene in the dForce Starter Essentials, which comes with Daz Studio 4.10. It uses color-coded flags to explain the effect of mesh resolution and a number of surface properties. After you run the simulation, you need to interpret the results yourself using the brief explanation provided in the dForce – Start Here forum thread. Since it’s not easy to see what’s going on, I took the liberty of running the simulation and then pulling out various combinations of flags so it’s easier to see what’s going on.
a. The first thing to look at is the effect of mesh resolution which is the number of polygons in the mesh. The more polygons the more opportunity for the plane to drape in a more complex manner. From the forum post, the flags are color coded to show different mesh resolutions:
Below are the simulation results at 3 different timeline positions. Frame 120 is just after the flags have become vertical, frame 150 is halfway to the final draping, and frame 180 is at the end of the timeline. I show it this way as it helps see how the various parameters affect draping.
Frame 120:
Frame 150:
Frame 180:
As you can see, the high polygon meshes will drape nearly vertical while the lower resolution meshes look stiffer and have a less compact shape. The former would better represent a silky fabric while the latter might best represent a stiff fabric. Of course, resolution only helps if you are creating a primitive or your own model where you have control over how many polygons are present. If you are using a commercial mesh, you will be stuck unless you have a decimator to reduce the number of polygons. Sub-division adds virtual polygons but they do not have an effect in the dForce simulation; however, they can soften sharp edges that occur during a simulation.
The flags above all have the same dForce surface properties:
Continues in next post
Part 2 of 3
b. The flag test scene expands upon the 4 resolutions in the middle (yellow, cyan, blue, purple) by testing combinations of Buckling Stiffness and Buckling Ratio. In this next series of photos
In the images below, BuS = Buckling Stiffness and BuR = Buckling Ratio
c. We’ll start with the higher resolution yellow flags (96 divisions).
Frame 120:
Frame 150:
Frame 180:
BuS: 0.02 1.0
BuR: 0.95 0.05 0.05 0.95
Buckling Stiffness appears to have the bigger effect (left flags vs right flags in frame 150 and 180) with lower values causing draping closer to the pole. Buckling Ratio had a lesser effect.
d. The Cyan flags have 64 divisions.
Frame 120:
Frame 150:
Frame 180:
BuS: 0.02 1.0
BuR: 0.95 0.05 0.05 0.95
Again, Buckling Stiffness appears to have the bigger effect (left flags vs right flags) and more of an effect than on the higher resolution yellow flags. The higher Buckling Ratio (outside flags) seem to drape better to my eye.
e. Blue flags have 48 divisions (similar to what I used for the banner flag in section 22).
Frame 120:
Frame 150:
Frame 180:
BuS: 0.02 1.0
BuR: 0.95 0.05 0.05 0.95
Again, Buckling Stiffness appears to have a bigger effect (left flags vs right flags) than Buckling Ratio.
f. The last series is the purple flags with 32 divisions.
Frame 120:
Frame 150:
Frame 180:
BuS: 0.02 1.0
BuR: 0.95 0.05 0.05 0.95
High Buckling Stiffness makes for a stiff draping (left vs right in frame 150 and 180). Buckling Ratio had little effect.
Continues in next post.
Part 3 of 3
g. There’s four extra blue flags that are supposed to show the effect of Density. It actually is multi-variable because low density is combined with low Buckling Strength and vice versa.
Frame 120:
Frame 150:
Frame 180:
BuS: 0.02 1.0
BuR: 0.95 0.05 0.05 0.95
Den: 100 650
The higher density/buckling stiffness flags have stiffer folds and less draping. They also rotated to the left from the right as they finished simulating.
h. If you load this scene yourself, you’ll note that there’s a large sphere next to each flag and the flags start in a horizontal position. The flag first drapes over the sphere, then when the flag has mostly wrapped over the sphere…
…the flag pole begins to tip to the vertical. Once the pole is vertical, the animation runs until the flags drape completely vertical as shown in step g, frame 180.
The use of a sphere does make for a good looking flag drape but it’s not intuitively obvious why you would simulate something in this fashion. I hid the spheres and ran the simulation and without the spheres, adjacent flags can catch each other and their mesh explodes. I then started with the poles vertical and the end drape is similar to that from using the spheres (frame 180). The biggest differences are in the intermediate stage, where the flags without the sphere have folds in the lower-right half but the fabric is flat in the upper-left half.
You never get a drape that looks like step g, frame 120. By the time the folds appear in the upper-left half, the lower-right has mostly collapsed towards the pole like step g, frame 150
Is it important if you miss that intermediate stage? Maybe, maybe not. It depends upon what look you are trying to achieve. If you don’t get the results you expect with dForce, you may need to get creative and do it as multi-step process (a good example of that is what Areyna did to simulate a dress on a horse rider, starting with the dress and horse upside down and flipping them over during the simulation).
Hey Rich, thanks for all these amazing illustrations! Is the flag scene somewhere in DS? Is it something we need to download to practice this test? I don't have any flags.
Yes, it's part of dForce Starter Essentials. I did nothing special to get it - I think it was in DIM as a download when the first version of DS that used dForce was released. If it's already installed, you can find it from Smart Content by searching with the word dForce. Or from the Content Library, look in Products / D / dForce Starter Essentials. If it's not installed, check DIM or go to your product library and search for dForce. I can't post a direct link as it's not a store product. It contains 3 sample scenes and 4 wardrobe items.
I've updated the pdf to version 5. It now has 91 pages. The PDF is a handy way to see the examples that have been posted to this thread. The first page has a menu (as does the first post in this discussion thread) to make it easy to jump to the section you are interested in. The images in the PDF are smaller than what you see online but you can zoom in as needed to see details. Comparison between images are easier online as you can open images in browser tabs and switch between them to see what's changed. Although both can stand alone, they also work well in combination.
Hopefully some of what I have been posting is proving useful to others. Next I'll be summarizing how to put fruit in a bag that I posted recently.
(Note: I had difficulty posting the pdf. I had to reduce image resolution to reduce file size. Let me know if that presents any issues; otherwise I'll have to split into two parts.)
Edit: new version is available - go to first post for link
Ok, thanks so much. I must have missed that when reading early on.
I just downloaded this new pdf and scrolled through all the pages and the images look fine to me when compared with images from the previous pdf docs. Valuable resource for sure.
Oh good, thanks for letting me know.
Great lot of information, Batman! Well done!
Quite a lot of information for a total noob like me, but I will have a lot to play around with, thanks to this piece of digital paper.
Could You mayhaps test dForce on fluids sometime in the future? Like.... using molten chocolate to cover a cake or something like that? Or lava flowing over the edge of a cliff... I guess there's quite a lot possibilities to be tested..
thanks!
Those are good ideas - I'll take a look
If you want something to fall until it connects, but not to deform at all, what would you adjust?
Thank you for the info. It's quite strange to me that the Wind Node is only available there under the Create drop down menu. I would think it would be under the D-Force tab menu as well!
Ok, here's a weird one. The 'start from memorized pose' option doesn't consider a figures Scale as a pose change. How do I turn a scale change into a pose change so that dForce recognizes it?
Animated drape, maybe? Using the timeline.
Ah! That is working... mostly. It catches on the skin at the neck and wrist so it bunches up instead of sliding off. But it's a better starting point than what I was trying.
The mysteries of Daz menus! There's so much buried in so many spots, can be hard to find or keep track of. I guess that what comes with a complex program.
@thistledownsname Here's what I have on the dropping object question. I'm not aware of any surface property combinations that do what you ask.
26. Dropping an Object onto a Surface. The question was asked “If you want something to fall until it connects, but not to deform at all, what would you adjust?” I didn’t have an answer so I did some experimentation and got it to (mostly) work.
a. Create two primitive cubes of 10 divisions each. Raise one so it starts in the air above the first and add a dForce dynamic modifier to it.
b. Go to the Simulation Settings pane and set Collision mode to Good. This preserves the cube’s shape; otherwise, it begins to collapse in on itself.
c. Run an animated simulation. You’ll find that as soon as the simulation starts, the top cube not only falls but also begins to distort. It will land on the lower cube and stop dropping but will continue to distort, although it mostly retains the shape of a cube.
d. Repeat the above steps but make the upper cube only 1 division. Now as it drops it will not distort; however, as it lands on the lower cube it will shrink slightly (hard to see in these images but easy to see in an animation). I could not find a surface property combination that eliminated the slight shrinkage but it doesn’t matter for our purpose.
e. Add an object to the scene and position it at the bottom of the upper cube. I found it works best if it’s mostly inside the box with a little extending below.
f. Next add a rigid follow node to the box. Go to the Tool Settings pane and from the drop down box choose Geometry Editor. In the viewport left-click on the bottom face of the upper cube. Then right-click in the viewport and select Geometry Assignment/Create Rigid Follow Node from Selected… and give it a name when the dialog box opens.
g. Now when you go to the Scene pane you’ll see that the upper cube has a child with an icon that looks like an I-beam. Drag the hammer on top of that node to make it a child of the rigid follow node:
(You don’t have to add the rigid follow node before the simulation. You can add it anytime, even after an object has undergone a simulation.)
h. Rerun the dForce simulation. As the box drops, the hammer drops with it. When it hits the lower box, it stops falling but does rotate (most likely because the polygon face is undergoing some twisting transformation that isn’t visible on the screen).
Frame 0: Frame 11: Frame 29:
If you hide the cube, you just see the hammer dropping:
Frame 0: Frame 7: Frame 11:
i. Although it works, I’m not sure what value this flat drop has. If making static images, you could just put the hammer on the cube. If making an animation, put the hammer in the air at frame 0, and at frame 11 put it on the cube. There is no physics other than gravity so you don’t get a bounce or other reaction. However, it may be useful with some forms of motion.
j. I made the lower cube into a ramp and sized the hammer to be the same size as the side of the box. When it drops, it hits the ramp, rotates, and continues to do so down the slope. The hammer follows along, and with the cube hidden, it look like its bouncing.
This gif show an animation of the drop and bounce from frame 0 to 42. (There was a glitch with the animation from frame 44-49 where the hammer flipped upside down, then flipped back at frame 50 and stayed that way until the end frame 60. I have no idea what in the simulation caused that short-time reversal.)
I like that bouncing hammer! But it's not what I was going for. This is the scene I was trying to set up. Some parts were dForce'd, others were just moved around a lot. Still curious to make it better using more dForce tricks.
L'Adair had a clever idea - using a circular transparency map on a square plane and draping it as cake icing. You can see the result here.
I've been working on creating a kite with primitives and getting it to move with dForce. Turns out to be harder than I expected but I'm making progress. Will post the result once I get it all figured out. dForce's behavior is not as straightforward as you would hope but then I'm most likely using it in ways it wasn't intended.
dForce Trees and Leaves is a new product in the store that shows a non-clothes application of dForce. The companion YouTube video has a section that shows how some parameters affect the fall of leaves which you may find helpful in other applications of dForce.
Thanks for the heads up about the video. I missed that on the product page.