3D Models
3D Print Models
3D Scans
Animation & MoCap
Textures
Materials
Skin Textures
2D Game Assets
Sound Effects
Brushes & Tools
Reference Photos
Stock Images
HDR Images
3ds Max
Maya
Blender
Unreal Engine
Unity
ZBrush
Cinema 4D
LightWave
Marmoset
SketchUp
FBX
DAE
DXF
GLBDAZ: About dForce, Drapeables and Dropables
119Thread Activity
Pushee-RiWed, Sep 03
Pushee-RiWed, Sep 03
Pushee-RiTue, Sep 02
Pushee-RiTue, Sep 02
Pushee-RiTue, Sep 02
Beautiful MisfitsSun, Aug 31
Drapeables and dropables (*) are a cool feature that can enhance your scenes. But why are there so few of them?
Unfortunately, most producers are conditioned to one-click solutions... or convince themselves that users are subject to this conditioning.
Far from it!
With a little information, you, dear reader, are quite willing to make a few more clicks... especially if this extra work allows you to elevate your renderings far beyond the DAZ mainstream and give your scenes momentum and dynamism.
I am not writing the following tutorial/infotorial in one go, but will respond to myself with new blocks of information. So please forgive me if I publish the various parts bit by bit (after all, I have to continue working on my PAOs
The following topics are planned:
* POSITIONING IN 3D SPACE the be-all and end-all of any work with dropables
* THINKING LIKE A FILM DIRECTOR after all, a picture can lie better than 1000 words
* ABOUT GLUING AND STAPLING how to keep parts of dropables in place
* ABOUT HIDE AND (SEEK) SHOW AGAIN this is a WIP
* ONLY FOR THOSE WITH TOUGH NERVES: SIMULATING OBJECTS ONE AFTER THE OTHER (Argghhhh!!!!)
For this tutorial, I am using my DropLace. However, the information provided can be applied to other Dropables.
(*) Drapeables and dropables are objects that are not tied to a model ( " parented ", such as clothing), but are used without reference to the model (with the exception of collisions).
Unfortunately, most producers are conditioned to one-click solutions... or convince themselves that users are subject to this conditioning.
Far from it!
With a little information, you, dear reader, are quite willing to make a few more clicks... especially if this extra work allows you to elevate your renderings far beyond the DAZ mainstream and give your scenes momentum and dynamism.
I am not writing the following tutorial/infotorial in one go, but will respond to myself with new blocks of information. So please forgive me if I publish the various parts bit by bit (after all, I have to continue working on my PAOs
The following topics are planned:
* POSITIONING IN 3D SPACE the be-all and end-all of any work with dropables
* THINKING LIKE A FILM DIRECTOR after all, a picture can lie better than 1000 words
* ABOUT GLUING AND STAPLING how to keep parts of dropables in place
* ABOUT HIDE AND (SEEK) SHOW AGAIN this is a WIP
* ONLY FOR THOSE WITH TOUGH NERVES: SIMULATING OBJECTS ONE AFTER THE OTHER (Argghhhh!!!!)
For this tutorial, I am using my DropLace. However, the information provided can be applied to other Dropables.
(*) Drapeables and dropables are objects that are not tied to a model ( " parented ", such as clothing), but are used without reference to the model (with the exception of collisions).
! REPORT
POSITIONING IN 3D SPACE
First things first: after a simulation, the model's pose should not be changed. With Dropables, it's even stricter: the pose MUST not be changed!
For the workflow, this means:
POSING then POSITIONING then SIMULATION then GROUPING
Tip: Don't try (as I unfortunately keep doing, despite writing tutorials) to position a dropable in perspective view! It's frustrating and takes longer than necessary.
In the following example, I have posed the model and want to hang the dropable (green) on the right hand (1)
Therefore I switch to the real (!) top perspective (i.e., not the rotated normal perspective) and position the dropable where it should (approximately) be (2 + 3).
Then I switch to a real (!) side view and position and rotate the dropable there as well (4 - 6).
Il make a few more corrections with (7 + 8). Ready to simulate
VERY IMPORTANT: Check the position of the dropable from all angles! There must be no overlap with the model!
First things first: after a simulation, the model's pose should not be changed. With Dropables, it's even stricter: the pose MUST not be changed!
For the workflow, this means:
POSING then POSITIONING then SIMULATION then GROUPING
Tip: Don't try (as I unfortunately keep doing, despite writing tutorials) to position a dropable in perspective view! It's frustrating and takes longer than necessary.
In the following example, I have posed the model and want to hang the dropable (green) on the right hand (1)
Therefore I switch to the real (!) top perspective (i.e., not the rotated normal perspective) and position the dropable where it should (approximately) be (2 + 3).
Then I switch to a real (!) side view and position and rotate the dropable there as well (4 - 6).
Il make a few more corrections with (7 + 8). Ready to simulate
VERY IMPORTANT: Check the position of the dropable from all angles! There must be no overlap with the model!
REPLY
! REPORT
THINKING LIKE A FILM DIRECTOR
My standard example: Do you really believe that people in movies beat each other up when the going gets tough?
OK ... there is one exception (especially when it comes to Oscars): Will Smith! But otherwise,
it's all just ACTING AND FAKE!
On Dropables (which can generate a lot of errors), this means that it doesn't matter what you see in the viewport when you rotate and flip the view in search of errors...
... what matters is only what the camera sees from your desired camera perspective. Because that's what creates your image - regardless of whether you can see in the background that the model isn't wearing pants, that the dropable has produced errors and glitches, or that Don Taco is picking his nose behind the model.
None of that matters for your image. So forget about all the mistakes... and enjoy the camera view.
And if minor errors do occur in the desired view of the Dropables, simply move them carefully until the errors are no longer visible
My standard example: Do you really believe that people in movies beat each other up when the going gets tough?
OK ... there is one exception (especially when it comes to Oscars): Will Smith! But otherwise,
it's all just ACTING AND FAKE!
On Dropables (which can generate a lot of errors), this means that it doesn't matter what you see in the viewport when you rotate and flip the view in search of errors...
... what matters is only what the camera sees from your desired camera perspective. Because that's what creates your image - regardless of whether you can see in the background that the model isn't wearing pants, that the dropable has produced errors and glitches, or that Don Taco is picking his nose behind the model.
None of that matters for your image. So forget about all the mistakes... and enjoy the camera view.
And if minor errors do occur in the desired view of the Dropables, simply move them carefully until the errors are no longer visible
REPLY
! REPORT
Masterstroke
Karma: 3,506
Sat, Aug 30Allthough I agree with you on a reasonable level, my gut feeling, my lizard brain is a f-ing purist.
In any of my rendered scenes, I made sure, that clothes or items look as perfect as possible from any angle, even out of the range of the camera.
I'd render a portait of my character in a winter outfit and made sure, that anatomical elements don't have unwanted mesh deformations during posing.
Why would I care if joints look right on a fully clothed character?
It shouldn't matter, as you point out, but it bothers me to no end.
In any of my rendered scenes, I made sure, that clothes or items look as perfect as possible from any angle, even out of the range of the camera.
I'd render a portait of my character in a winter outfit and made sure, that anatomical elements don't have unwanted mesh deformations during posing.
Why would I care if joints look right on a fully clothed character?
It shouldn't matter, as you point out, but it bothers me to no end.
Pushee-Ri
Karma: 33,953
Sat, Aug 30Yep - that's what distinguishes an artist (you) from a designer (the Pushee):
I work with a visual focus, purposeful and concentrate on the effect (!) of the image ... while artists work expressively and... well... artistically. There's nothing wrong with either of them, right?
I work with a visual focus, purposeful and concentrate on the effect (!) of the image ... while artists work expressively and... well... artistically. There's nothing wrong with either of them, right?
INTERMEZZO: GONE WITH THE WIND
In addition to the normal dForce method (gravitation), DAZ also offers a super blower: the dForce Wind Node. It is likely due to the somewhat fiddly handling of the node (hence " gone with the wind " ) that one rarely sees renderings in which it has been used.
Without my usual blah blah blah, here is a screenshot with explanations. (OK, just a tiny bit of blah blah blah: you really have to tinker and experiment with the Wind Node: it took me 17 attempts to get the necklace to look more or less the way I wanted it to).
In addition to the normal dForce method (gravitation), DAZ also offers a super blower: the dForce Wind Node. It is likely due to the somewhat fiddly handling of the node (hence " gone with the wind " ) that one rarely sees renderings in which it has been used.
Without my usual blah blah blah, here is a screenshot with explanations. (OK, just a tiny bit of blah blah blah: you really have to tinker and experiment with the Wind Node: it took me 17 attempts to get the necklace to look more or less the way I wanted it to).
REPLY
! REPORT
Masterstroke
Karma: 3,506
Sun, Aug 31So wind force works for you?
I've never been lucky with it. I never could get the parameters right, and in nine out of ten simulation runs, I got clothes explosions.
I've never been lucky with it. I never could get the parameters right, and in nine out of ten simulation runs, I got clothes explosions.
Pushee-Ri
Karma: 33,953
Sun, Aug 31" So wind force works for you? "
Yep - with nerves of steel and 17 attempts ... hmm, actually, probably not
I usually play around with the sliders (as I do with scene lighting) until something looks (reasonably) right. Ultimately, however, I usually end up resorting to my old trick, which I will SHARE HERE AS A TIP:
Don't wait for the simulation to end, just stop it when it looks good
Yep - with nerves of steel and 17 attempts ... hmm, actually, probably not
I usually play around with the sliders (as I do with scene lighting) until something looks (reasonably) right. Ultimately, however, I usually end up resorting to my old trick, which I will SHARE HERE AS A TIP:
Don't wait for the simulation to end, just stop it when it looks good
ABOUT GLUING AND STAPLING
dForce objects (and therefore also Dropables) collide with everything and everyone in the scene. This also applies to simple overlaps. This circumstance can be used to keep parts of a Dropable (more or less) in place during the simulation.
In the image below, you can see a blue object. This is a simple cylinder from the DAZ construction kit, which I scaled to form a kind of board and pierced through both the Dropable (a simulated, simple mini skirt) and the model. At this point, the simulation will later attempt to drop the mini skirt, but will be prevented from doing so by the blue object.
Please note: this method often produces unsightly results at the point where the dropable is attached to the model. These range from simple distortions to jagged materials. For precisely this scenario, I gave you my tip above on how to think like a film director and plan for these mistakes. The only thing that matters is what the camera sees!
One more tip: don't make the collision object too narrow! The narrower the object, the more glitches and errors you will get. If the objects are too narrow, the Dropable may even tear and simply slide further down.
dForce objects (and therefore also Dropables) collide with everything and everyone in the scene. This also applies to simple overlaps. This circumstance can be used to keep parts of a Dropable (more or less) in place during the simulation.
In the image below, you can see a blue object. This is a simple cylinder from the DAZ construction kit, which I scaled to form a kind of board and pierced through both the Dropable (a simulated, simple mini skirt) and the model. At this point, the simulation will later attempt to drop the mini skirt, but will be prevented from doing so by the blue object.
Please note: this method often produces unsightly results at the point where the dropable is attached to the model. These range from simple distortions to jagged materials. For precisely this scenario, I gave you my tip above on how to think like a film director and plan for these mistakes. The only thing that matters is what the camera sees!
One more tip: don't make the collision object too narrow! The narrower the object, the more glitches and errors you will get. If the objects are too narrow, the Dropable may even tear and simply slide further down.
REPLY
! REPORT
ABOUT HIDE AND (SEEK) SHOW AGAIN
Hiding objects can speed up a simulation - that's well known. But did you also know that hidden objects are not recognized by the simulation algorithm?
This is a useful feature that can be exploited. Imagine, for example, a Drapable bedspread with one arm and one leg of the model lying below the bedspread and one arm and one leg lying above it.
Of course, this can also be done with a lot of fiddling around, by " somehow " positioning the model around the bedspread ...
or you can simply hide the body parts that are to lie on the Drapable later on before the simulation.
The image below shows you a simple example, but despite its simplicity, it would be difficult or impossible to achieve without hiding parts of the body.
Note: hiding body parts is a bit tedious. I am currently working on a few scripts that will do this for you in a fraction of a second. The scripts will probably be available as a freebie in Beautiful Misfits' little shop of Goodies.
Hiding objects can speed up a simulation - that's well known. But did you also know that hidden objects are not recognized by the simulation algorithm?
This is a useful feature that can be exploited. Imagine, for example, a Drapable bedspread with one arm and one leg of the model lying below the bedspread and one arm and one leg lying above it.
Of course, this can also be done with a lot of fiddling around, by " somehow " positioning the model around the bedspread ...
or you can simply hide the body parts that are to lie on the Drapable later on before the simulation.
The image below shows you a simple example, but despite its simplicity, it would be difficult or impossible to achieve without hiding parts of the body.
Note: hiding body parts is a bit tedious. I am currently working on a few scripts that will do this for you in a fraction of a second. The scripts will probably be available as a freebie in Beautiful Misfits' little shop of Goodies.
REPLY
! REPORT
DoroThee237
Karma: 12,998
Sun, Aug 31So there I am in awe, achieving this result in a few days leaves me speechless
Pushee-Ri
Karma: 33,953
Sun, Aug 31Thank you very much for this wonderful compliment
... but actually, it took years of working with dForce and dealing with all kinds of sources of error, etc. (though I still find some
... but actually, it took years of working with dForce and dealing with all kinds of sources of error, etc. (though I still find some
OUTTAKE
Oops! ...I think I scaled too much on the X-axis ...
Oops! ...I think I scaled too much on the X-axis ...
REPLY
! REPORT
Beautiful Misfits
Karma: 15,389
Sun, Aug 31and over the other ear as well and it can be used as bling bling over the mouth
Pushee-Ri
Karma: 33,953
Sun, Aug 31Arrggghhhh... don't give me ideas like that. I'm just glad I've finished the promos
Beautiful Misfits
Karma: 15,389
Sun, Aug 31*lol* I'm innocent, it's because I hit my head with the fork and the bump causes crazy ideas
Pushee-Ri
Karma: 33,953
Sun, Aug 31Crazy ideas? I love it!!!
Was it a large fork or a normal one that I use in the garden?
I definitely have to check that out
Was it a large fork or a normal one that I use in the garden?
I definitely have to check that out
Beautiful Misfits
Karma: 15,389
Sun, Aug 31It was a normal one, one can use in the garden BUT no plastic shit, real iron and ergonomic formed.
And you have to use a spade, hit a root from a tree, where the fork is somehow connected, and with that push from the spade, the fork will "jump/fall and you have to have your head in the perfect direction.
Well, I made this all without thinking *lol* and for sure not planned (panic included because of the blood thinner).
But I think I've proven it by now: Unkraut vergeht nicht (weed won't die, meaning: Ill weeds grow apace, Bad weeds grow tall)
And you have to use a spade, hit a root from a tree, where the fork is somehow connected, and with that push from the spade, the fork will "jump/fall and you have to have your head in the perfect direction.
Well, I made this all without thinking *lol* and for sure not planned (panic included because of the blood thinner).
But I think I've proven it by now: Unkraut vergeht nicht (weed won't die, meaning: Ill weeds grow apace, Bad weeds grow tall)
Thanks for this useful mini tutorial! I wasn't aware of the hidden object that is completely ignored by deforce. This opens up a whole new field of possibilities!
REPLY
! REPORT
Pushee-Ri
Karma: 33,953
Sun, Aug 31Thank you very much for your kind feedback
As soon as the scripts are ready and published, I will get back to you. Then the hiding process will be even faster.
As soon as the scripts are ready and published, I will get back to you. Then the hiding process will be even faster.
Beautiful Misfits
Karma: 15,389
Sun, Aug 31Don't break your brain, the tools are already out there.
https://www.daz3d.com/daz-studio-4-scene-tools-set-1 (on sale and cheap)
https://www.daz3d.com/easy-geometry-hider (not on sale and pretty expensive)
https://www.daz3d.com/daz-studio-4-scene-tools-set-1 (on sale and cheap)
https://www.daz3d.com/easy-geometry-hider (not on sale and pretty expensive)
QUICKTIP: GROUPING
Dropables and Drapeables are (unlike 3D clothing) not " anchored " (parented) to the model. This means that when you move the model, the Dropable will not move - neither before nor after the simulation.
To move both objects, use this simple trick:
* Shift-select the model and the (simulated) object in the " Scene " pane.
* Choose (from the main menu) Create -> New Group ...
* If you like, you can enter another name than " Group " and then hit " Accept "
You can now move or rotate your new group - the simulated object will follow.
If you want to assign a different material preset to the dropable, expand the group in the " Scene " pane, select the object, and proceed as usual.
Dropables and Drapeables are (unlike 3D clothing) not " anchored " (parented) to the model. This means that when you move the model, the Dropable will not move - neither before nor after the simulation.
To move both objects, use this simple trick:
* Shift-select the model and the (simulated) object in the " Scene " pane.
* Choose (from the main menu) Create -> New Group ...
* If you like, you can enter another name than " Group " and then hit " Accept "
You can now move or rotate your new group - the simulated object will follow.
If you want to assign a different material preset to the dropable, expand the group in the " Scene " pane, select the object, and proceed as usual.
REPLY
! REPORT
QUICKTIP: START BONES FROME MEMORIZED POSE
When you create a brand new scene, the " Start Bones... " option is enabled. There is nothing wrong with this, except that in some cases the simulation can be accelerated. For example, all of my scenes that you see here in the thread or in the gallery were generated with this option turned OFF
Here is a brief comparison:
Start Bones ON
Plus: better draping of clothing (and sometimes dropables) during simulation.
Minus: simulation time is very (!!) long - approx. 8 to 10 times longer (and more) than when the option is turned off.
Start Bones OFF
Plus: simulates faster
Minus: the draping is (sometimes) not as precise as with the option turned on
So you have to decide for yourself which option you want to use to run the simulation. As a general rule, I always recommend turning this option off for smaller Dropables.
When you create a brand new scene, the " Start Bones... " option is enabled. There is nothing wrong with this, except that in some cases the simulation can be accelerated. For example, all of my scenes that you see here in the thread or in the gallery were generated with this option turned OFF
Here is a brief comparison:
Start Bones ON
Plus: better draping of clothing (and sometimes dropables) during simulation.
Minus: simulation time is very (!!) long - approx. 8 to 10 times longer (and more) than when the option is turned off.
Start Bones OFF
Plus: simulates faster
Minus: the draping is (sometimes) not as precise as with the option turned on
So you have to decide for yourself which option you want to use to run the simulation. As a general rule, I always recommend turning this option off for smaller Dropables.
REPLY
! REPORT
Pushee-Ri
Karma: 33,953
Tue, Sep 02Oops - I just forgot: you can find this option under:
" Simulation Settings " pane -> Simulation -> Initialization
" Simulation Settings " pane -> Simulation -> Initialization
QUICKTIP: DROPABLES - DIFFERENT POSITION, DIFFERENT RESULT
When working with Dropables, you may notice self-similarity in the results (even with slightly altered poses of the model). This is due to the underlying Cloth Simulation in general and DAZ dForce in particular: the algorithm cannot take into account all the subtleties you are used to in real life.
To achieve a variety of results, you should therefore use DAZ's transformation options. In addition to rotating a Drobable, you can also scale it - either evenly or only along one axis. Some Drobable products also offer additional morphs that allow you to position or distort the object BEFORE the simulation.
By combining rotation, scaling, and (if available) morphs before the simulation, you can avoid or at least reduce the self-similarity of the results.
When working with Dropables, you may notice self-similarity in the results (even with slightly altered poses of the model). This is due to the underlying Cloth Simulation in general and DAZ dForce in particular: the algorithm cannot take into account all the subtleties you are used to in real life.
To achieve a variety of results, you should therefore use DAZ's transformation options. In addition to rotating a Drobable, you can also scale it - either evenly or only along one axis. Some Drobable products also offer additional morphs that allow you to position or distort the object BEFORE the simulation.
By combining rotation, scaling, and (if available) morphs before the simulation, you can avoid or at least reduce the self-similarity of the results.
REPLY
! REPORT
QUICKTIP: ABOUT DISTANCES AND DIFFERENCES
DAZ simulates in a limited number of steps. I have never counted them, so I will just say 10 steps. If the dropable is too far away from the model, the simulation first wastes valuable steps to overcome the distance before the collision simulation can even start.
Therefore, it is very important to position the Dropable as close as possible to the model and rotate it if necessary (think of a necklace, for example: if your model is in an upright pose, you can rotate the necklace towards the chest). This leaves the algorithm with more steps, which can then be used to improve the draping of the object.
DAZ simulates in a limited number of steps. I have never counted them, so I will just say 10 steps. If the dropable is too far away from the model, the simulation first wastes valuable steps to overcome the distance before the collision simulation can even start.
Therefore, it is very important to position the Dropable as close as possible to the model and rotate it if necessary (think of a necklace, for example: if your model is in an upright pose, you can rotate the necklace towards the chest). This leaves the algorithm with more steps, which can then be used to improve the draping of the object.
REPLY
! REPORT
