Almost ready:
Looks good! Do note though, the morphing will be tested/worked-on at a later date.
We’ll have to go back and forth a bit on the results you want to achieve, but this should make for a good quick setup video for you guys.
That’s okay, I expected that (I’m sure at least a few months). I’m doing a couple of dynamic clothing pieces to include, feel free to use whichever you think works well. I’ll get it sent to you tomorrow. Thanks.
So here’s some clothing to test/demo with:
They all have their pivot set the same as the character’s and should pop right on, if not please let me know and I’ll work with you to fix that. There’s a custom retargeting ref pose for the mannequin’s skeleton and if you want any animations retargeted I can help you/do that for you if needed. There should not be any dependencies on other plugins, I’m pretty sure everything required is saved into the project files - please note, your plugin is not installed in this project.
The things we’d like to see explained are:
- How to set up the simulation to collide against the character’s physics asset
- How to set vertex constraints to fix points of a garment so they’ll stick to the surface of the character e.g. at the hem of the skirt or the top of the dress
- Whatever other features you have that you think are useful, e.g. if you have some form of weighting to control stiffness (tbh I don’t know how much of this you’ve implemented vs. what you’ve got planned)
Here’s a link to the project files, zipped:
Do you have some sort of max distance setting that a particle is allowed to simulate away from its original position? I’m being asked about that, thanks.
No, there are no hard/built-in bounds to the simulation. Not counting floating-point errors and spatial hash over-saturation after a certain point.
Perfect, thank you! I’ll try to get to this next week.
Thanks, looking forward to it.
I’m very interested in how it will work and look 
Please post a short video or gif please?
Have not forgotten, a few things came up on our game project and on VICODynamics (see recent commits for a preview of some awesome new things, I’ll go through them, officially, here later); so this got pushed a bit.
I’ll do my best to get something on this by Sunday (F1 race 
)
That’s what I was looking for (almost) buying when fluid support is added.
Did a first pass on the test project, set everything up with the currently available features and tweaked things. A few interesting things came up:
- Attach to Rigids feature does not record which bone the rigid belongs to in the component if the component is a Skel Mesh Comp. Once I add this, you should get much better, per-bone, attachment using that feature.
- The current Mesh to Cloth algorithm tries its best to maintain the original Mesh shape, which may not work as well for complex clothing.
- Once #1 is in, we’ll be able to better see if collision is an issue. The way NvCloth does collision is very different to the way I do it in VD. This is because NvCloth is purpose built for skeletal character clothing that works with animations, this means they can make a lot of assumptions and special tricks for better use case results. Whereas, I have taken a more generic approach because I’d like VD to be able to be used in a wide array of scenarios. As a result we get into “issues,” if you will, like this.
- Notice the trench coat has very little movement, this is because most of the vertices are very close to the rigid bodies and get auto-attached. Just a matter of tweaking a few things to improve this. Like making the Physics Asset fit the mesh tighter and playing with the Particle radius and Attach to Rigids Radius scale property.
Here is a little video from the first test: https://youtu.be/1G9HAhIXxc4
Let me know if you have questions, happy to answer them!
I’ll be following up on this soon.
Glad to see you got some results, whew. For your rope/chain simulation, do you place particles at every vertex in those cases also? If not how are they placed?
No, for those, I place a particle per segment essentially.
Cool, although I’m not a programmer at all it seems a good thing that you’re not strictly limited to placing particles at vertices so maybe the idea of placing particles at joint endpoints is not too crazy?
Definitely something to test out! Like I mentioned before, the only big issue is how to best determine rotation of the bones based on two particles which do not have a rotation.
Btw, I should be doing a few more tests on the scene you sent very soon.
Quick update:
Support for 4.16 will be up on the Marketplace next week.
And the upcoming Feature Update has become a very exciting one with a bunch of new features that I will reveal very soon! Most of which have been much requested.
[MENTION=529477]Flesh Forge[/MENTION]: I have not forgotten to do further tests! I have not had the time to commit to them just yet, apologies for the delay there.
That’s all right, I don’t expect to set your priorities for you. Please keep in mind we’re also looking for documentation/demonstration of the setup and authoring procedures though.
hey, any news about the additional features?
It certainly has been a while since I posted an update or a feature preview here, this is on purpose. I’m saving up for a big unveiling of everything very soon, though later than I hoped due to work on our game taking priority. Many new features, including those already announced, will be released and shown off here. I truly wish I had a solid date for you, but I cannot commit to one just yet. As soon as I can, I will post everything here for everyone! Thank you for everyones patience, I don’t think you’ll be disappointed.
And if you’d like access to the latest and cutting edge, just email me your invoice (ujen@.com) and I’ll get you setup with access to the GitHub repo.