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
Those new features are great and finally I can animate chars and props without errors... But still the Fluid source doesn't work when parented to chars or primitives.
I made some test parenting the source to Genesis' head and making it move the torso around. Nothing appeared, simply empty.
I tried again using a static cube and parenting the source to it, nothing either.
In both tests I put the emitter really away from the figure to be sure the source have enough space to work.
Yes it does! Parenting can now be unlimited depth in the hierarchy.
I just tried it! Parented to Genesis's head, no output, PLY file size 1kb. Same with the Cube... What I'm doing wrong, then?
Here's my setup. Where's the error?
Difficult to say without more detail - I have an animation (not yet posted) of an animated G2F with Sources on both hands and it works fine.
Is your Source inside the Cube that it is parented to? If so, the Cube is treated as a solid and so the fluid can't go anywhere. And you are sure that the Source is inside the Fluid Domain volume?
Yes, the source is really distant from the cube, about hyerarchy just check the image.
*EDIT*
Doing new tests, it seems to work properly with freshly made Domains. I'll try renewing the domain in the saved scene...
Imago,. instead of "parenting" the fluid source to the object,. try adding add a "Track" modifier to the fluid source,. then select an object for it to track,. that object could be a figure, or a Target helper object which could be parented to an object , or just animated
Thanks for the tip, 3DAGE, one day it will be useful!
In the meantime I found the problem: I mistakenly ticked "Carrara Physics Engine" in the options. It seems to block the parented source to work properly. Once unselected that option, old saved setups works perfectly!
Now a normal question to Alberto: The manual says something about the damping force, but it says to not use the Carrara's default one... Where I can find the right one?
I looked everywhere in options and "Insert" and I can see only the standard one!
Imago :)
I'm seeing that as, ... Don't use the default Carrara force's "strength",. the default values are too strong for the fluidos sim.
it's an issue with carrara forces which has never been changed,. the default value are set too high for most simulations using carrara's own particles or physics.
Try with really low values. or adjust the Fluidos gravity force from -32 to -15 (or whatever) to get a slower moving fluid.
I'll normally set the default values of any forces I add to a scene to 0 "zero" whenever i'm adding forces to any scene,. whether i'm using physics, particles, or fluidos.
So I have to use the Carrara's default damping with low values? Ok, got it! Thanks man!
Anyone else getting a lot of "an error has ocurred" when you run the sim. ? (with the new version of fluidos (1.1) )
I have a simple sim set up with a vertex model (fluid) and the sim runs fine when that's in the domain,. but adding another object to that domain, then running the sim,.. causes the error message.
I also set up a New scene,. loaded in a couple of objects from my browser,. set them up in a domain,. tried to run the sim,. Error.
the objects i'm using have already been used in other sims with no error.
Any ideas !
Thanks for the update.
CarraraCafe news post is now updated with the latest version info:
http://carraracafe.com/news/the-fluid-sim-plugin-for-carrara-fluidos-1-01-is-released/
May I see the scenes and setup?
Thanks alberto :)
here's some stuff, .. hope it helps
https://www.dropbox.com/s/5etfktickblg850/fall_test.car?dl=0
This is a very simple vertex model and a fluid source,. i get an error when trying to run the sim (Ctrl Shift F),. selecting a folder,. confirm,. Error.
https://www.dropbox.com/s/6vdxxejojwdqqvl/fluid_glass.car?dl=0
this is a simple container,. with a fluid source above it. ...as soon as i confirm the folder,. error.
if i remove the container from the fluid domain,. I can run the sim.
https://www.dropbox.com/s/mhasc1uxknwdrhd/wineglass_blob.car?dl=0
Wine glass with some liquid (vertex models) the liquid is above the glass, and should drop into it.
as soon as i select thr output forder,. error,. ..If i remove the glass from the fluid domain,. it will simulate the liquid dropping,. but it's strange ,. see clip
https://www.dropbox.com/s/8rrwfe2zq0pfrt7/wineblob.mp4?dl=0
I'm also seeing a build up of the progress bars at the top of the screen, ..See pics
SPEC:
PC : OS : Windows 7 : Nvidia GTX 650ti 2Gb
pics
sorry havnt had a chance to play with new version
Andy, I didn't find so far the problem, I couldn't replicate the error. Nevertheless, about the strange behavior of fluid in the case of wineglass, it is an artifact of the simulator engine; it only appears in the scene when anisotropic reconstruction selected. Sometime ago, I found a similar artifact in diffuse particles, the cause was a simple bug in the engine, but I could fixed it.
Did you try running without OpenCL?, if not, try it, please, and let me know.
Just a thought: Check the mesh, search for some errors in its structure. I often get errors due to wrong oriented normals in other softwares and plugins, maybe it's the same here!
Maybe there is some point where vertices aren't joined or there is some hollow.
Running with open CL enabled / disabled doesn't seem to change anything,. ...i'll still get an error.
I've done some more tests today and strangely got a simple thing (vertex model Tray/box and a cube primitive as a fluid) to almost work, but only if i set the grid size to 0.3 ,.. any higher or lower and it failed. but at that size i'm not seeing any physical collision betweeen the fluid sim and the object,.the fluid just passes through.
so the grid size seemed to have some bearing on whether the sim will run or error,.
I then tried changing the object base thickness, which broke the whole thing, and it wouldn't simulate at any grid size.
All the objects are manifold (no holes) and made in carrara, (no imports) and most have been used in previous simulations with no issues.
on that note, if i load up any saved scene and try to run the sim,. Error.
I've found the problem,... it's me,... i'm an idiot.
I had grabbed the files from the download zip and dropped them into the extensions folder /fluidos .. without noticing that the DLL was also in there with the plugin files.
moving the dll to the main extensions folder to replace the old one,. seems to have resolved the issues.
it's now working.
sorry for the confusion, thanks for the help :)
Errare humanum est, perseverare autem diabolicum!
But sometimes being stubborn helps a lot!
Glad you solved!
Don't worry. It happens sometimes to us all
.
I'm glad you found the problem!
Anyone else is getting the "bad_module_info" error using the plugin?
Could you give more details?, when?, how?, please.
It seems to be basically Windows 10 new update (1709) and it's pretty random, but in Carrara I noticed that it happens always in determined conditions, like when there are more than 2 objects (beside the emitter) or when you stop the simulation with ESC button. I also think it could happen when the emitter (fluid source) moves around the scene when parented.
I also noticed some "standard" error when the fluid source mistakenly goes inside a prop from the outside.
I'll send more details when I find some other condition!
For the rest, the plugin is truly wonderful!
Hello, Alberto!
Came to back to let you know the errors stopped happening. Like they came, they left, dunno why!
That is good !
(Although for bugs hunting is bad
)
I'm doing lots of experiments with the plugin, I'll let you know immediately if I get some strange behaviour.
Last one I got was about two days ago, but I think it was my mistake, I made a too big domain and when I stopped the simulation, it gave me "An error has occurred" and the viewport stopped working with a new message "render failed". Then Carrara got stuck.
I'll bother you a little more, Alberto, for a little "request":
I managed to have the right size of everything, but I still get the problem of the fragmented, beady-looking streams with "high" velocities (20). The trick I used doesn't work too well, make the emitter longer just create horizontal pieces that looks even less natural...
But if I use 60FPS the stream looks perfect, even with higher velocity values... But my animations are 30FPS and the final result is too slow and doesn't follow the real movements and needs.
I made two different simulations, same scene, identical values but one at 30FPS and the second at 60. (Check the images)
There's a way to have the same result at 30 FPS? Could you add a slider or an option that makes the simulation "add" the missing parts at lower FPS?
Imago, maybe it was due to a high memory consumption. Sometimes, it was happened to me. Check the Carrara memory usage in the Windows task manager.
If fact, this slider exists. The actual rate is given by the Completion FluidS modifier; you can stretch and shrink as you wish. Only take into account the timing of the other animated objects.
Already tried, doesn't help. Shortening the completion also shortens the fluid animation and half of the animation is stlil. Plus it doesn't match anymore with the other animated objects. For example, if I make someone pouring a glass of water and mistakenly spill some on the table, the moment when this happens will not match with the fluids if I modify the completion slider.