dForce Assistant stretches out the dress
irynakhymych
Posts: 27
Hi guys! I am using Joan G9 (https://www.daz3d.com/joan-9-hd) with this outfit (https://www.daz3d.com/joan-9-high-elf-add-on-bundle) and I wanted to make her sit on an animal in it. When I ran simulation obv it exploded, do I purchased this product (https://www.daz3d.com/dforce-assistant) which made the dress not explode but it streched the dress out...like a lot. here is a picture https://photos.app.goo.gl/XPymBn7sycpBb9ES8 ;
Can anyone tell me what I am doing wrong or how I can shrink the dress? Thanks in advance!
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
The link of screenshot seems not valid...
https://photos.google.com/share/AF1QipMZaF8jg0prLrHUHkLmVoTO9Kk-o-ws_PQUxjVh257xVbT-BFblmErZy84v9yC3fA?key=aFJuX3kzNzNLYjBLc1FqYmhsVk9NOC10MEhqakpR
Hello. I can't say that I've dforced Joan 9 in that dress on a fox, but I got it to work on a horse with dForce Assistant.
It can be done. I don't recall now, but it was not the easiest project. What pose are you using?
OMG your render is gorgoues! I am using this yawning pose (https://www.daz3d.com/ffs-faery-antics-poses-for-genesis-9) but moved the arms. For some reason Daz will not let me upload the image, I have tried different browswers and refreshing and I see from commons this happens a lot. Let me see if I can make a new link (https://photos.app.goo.gl/VPKAb87EZengdwCK9).
Just created a new link - should work now https://photos.app.goo.gl/V9pDpjQHU75BWMtH8
06 "Happy Yawn"? I don't have that pose set but I'll look around and see what I have that's similar.
I really appreciate it!
I'm at work now so I won't be able to tinker with Daz until tonight. Which options were you using in dForce Assistant?
Oh no worries, this is for a personal project so not urgent, plus you're helping me so totally up to you when you do it! I am using the Figure set up assistant and the chair set up assistant options. I just tried it again after adjusting the figure and still completely streched out skirt. If it helps, I am using this fox (https://www.renderhub.com/vyusur/fox-for-dog-nubis) from Renderhub, but I am not sure if the fox matters, a mesh is a mesh right? Pic of current simulation https://photos.app.goo.gl/jk8ohLrsFjRXEn1M7
It was definitely a challenge .. when I posted that image in my gallery last spring, I noted, "...it took some doing to dforce this dress, in this pose, on a horse." Which, if I said that, I meant it.
I wish I'd mentioned more specifically what the challenges were, but I bet when I open up the saved file and take a look I may remember more.
Oh yeah I bet! I've been playing around with it for 2 days now and just cannot figure it out. Granted I am not a pro at Daz but my other simulations are decent
Okay....I figured it out.....I completely forgot that I scalled the figure back to 55%, but when the simulation was initiated, it reverted back to 105% which was the original size. Then it shrank the figure as it was simulating it, the top part of the dress sharnk with it but NOT the skirt.....so I went to the initiation frame, shrank the figure and dress to 55% and it worked like magic. I feel so dumb. Pic https://photos.app.goo.gl/GeTkSjzvvri8oraZ9 ;
Aha! Good find! Glad you got it working.
Thanks so much for your willingness to help!
I find turning off "self collision" sometimes helps to avoid dForce explosion and stretching
dForce is kind of random though
In schools we teach that computers, like robots, are determininstic and repeatably precise, then we invent dForce :-)
That really is a lovely drape and image.
Ha! My husband, who is a programer, was asking me if it was deterministic and I was like It's temperemental
It's probably non-deteministic becasue it's a chaotic system where a small change has a large effect, and the small changes come from the fact that computation threads are scheduled by the opereating system not the user application, so the ordering can't be predicted, and that hardware floating point is not commutative i.e. A+B does not equal B+A which means a different ordering changes the result.
Hopefully it's that and not that it's doing random sampling.
Sensitivity to initial conditions is not the same as non-deterministic - if it were, one wouldn't be able to tell how seensitive it was to initial conditions since the outcome would vary. I believe, without being certain, that dForce simulations are deterministic - so given an identical starting state they should end up in the same place.
You are are correct: in programming, deterministic means you always get the same output from the same input. It has a different meaning to the common or philosophical one.
Note that hardware, and Windows or MacOS themselves, are NOT deterministic from the point of view of an application program, because the program is being run by the OS on the hardware, and cannot directly control everything that happens there. This makes it very difficult for a program to be truly determinstic.
One example of this is when you run a benchmark twice, the results vary by a small amount. (Here we would ignore the fact that the time is not determinstic, and claim that the pretty picture made by the benchmark is).
Another example is why chess programs sometimes return a different move given the same board position on the same machine and settings, because the order which their computation threads are "timeshared" on the CPU cores is not under its control, so it cranks through all the possible moves in a different order each time, and uses the best that it finds within the time limit allowed - sometimes it will therefore see the right move quicker. This timing-ordering issue is very problematic for floating point maths where the hardware evaluates A+B and B+A to a different answer sometimes.
If dForce is deterministic, which it could be, it would need to have been specifically engineered to be that way (with an associated reduction in performance). I suspect dForce is deterministic, at least up to the limits imposed by the hardware.
Glad you figured it out. You stumbled on the way to make oversized or looser fitting clothing by scaling the figure over time.
I get caught all the time by scale changes. I run a simulation, make a change in scale or pose, then when I run the simulation again I realize I was on frame 30 not 0 when I made the changes.