Help - Search - Members - Calendar
Full Version: Fluids Tutorial
Hash, Inc. Forums > Technical Direction and Development (Learning Animation:Master) > A:M Rendering, Compositing and Special Effects > Materials Laboratory > Water
swooster
The new Version 15 includes fluids.

For those who are impatient, you can just just load the following tutorial project:
Click to view attachment (for something less goopy, try changing the Goop material's surface tension to 200 instead of 300)

Tutorial:

Fluids are based on blobbies but their physics and some of their properties behave differently. Fluids are made of many droplets that interact with each other, in particular via pressure, surface tension and viscosity.

Create a new project and a new choreography.

From the library's models tab, drag the \Tutorials\smoke and \Primitives\Box models to the choreography. The smoke model will be our fluid emitter and the Box model will be used as a nozzle.

Create a new material, right-click its attribute and change its type to Particle System > Fluid.

Pop open the fluid emitter's properties.
Set the "Rate of Emission" and "Initial Velocity" to 700.
Turn on "Object Collisions".
Set the "Size > At birth" to 3cm.
The default droplet physics settings should work just fine.

Before continuing, make sure particle physics are turned on, your timeline is at 0 (so you don't inadvertently do a lot of calculations), and your view is in wireframe (wireframe renders fluids much faster than shaded mode).

Drag the fluid material onto the smoke model in the objects folder of the project workspace.

Another problem is that the emitter tends to spray the droplets instead of shooting out a stream. One reason is that the droplets are dispersed and their surface tension doesn't have much of a chance to cluster them into a solid stream. To help with this, rescale the emitter by 35% on the x axis and 50% on the z axis. The other reason is that the emitter randomly places droplets, and sometimes two droplets are put next to each other, causing them to push each other away with a lot of force. To help with this, we need a nozzle to collect any wayward droplets and give the fluid's pressure a chance to stabilize. The box model will be our nozzle; scale it by 30% on the x axis and y axis and rotate it by -45 degrees on the x axis. Translate it up so that it doesn't intersect the ground. Then rotate the fluid emitter by 45 degrees on the x axis and place it inside the nozzle at the very bottom.
Click to view attachment
The emitter and nozzle should look like the above image.

At this point, going forward in the timeline should yield a relatively coherent stream of droplets. If they formed a coherent stream then try viewing them in shaded mode. Be warned: Currently, the time it takes to generate the fluid surface heavily depends on how far spread out the droplets are. A coherent stream of fluid may be fast to render but if a single droplet gets very far away, it can slow things down a lot.

At this point, you can edit the surface attributes of the fluid material to make it look like water. Be careful to set your time to 0 before changing the surface attributes.

Finally, try doing a progressive render.
Click to view attachment


Another fluid example contributed by Stian <agep>
KenH
Wow. It renders faster than I thought it would. Lots of potential.
johnl3d
Tutorial as pdf



Click to view attachment


Mr. Jaqe
QUOTE(Paul Forwood @ Aug 22 2007, 04:48 AM) *
QUOTE
I think it is the pressure at which the particles are thrown on a surface...

Yes, that seems to be part of it but my experiments show that fluid will pass through a surface if the emitter is closer to that suface than the width of the blobbie even if the pressure is as low as .1. Also if I use the surface of an object as a fluid emitter and turn the pressure down to .01 some of the fluid particles will percolate through to the opposite side of the the normal.

That explains quite a lot of problems I've been having with the fluid emitter... Thnx Paul!
jzawacki
How do you make the fluid desolve? I was playing with it (works awesome) and I have bits that splashed out, and then just seem to "flop" around.

My Fountain (48MB, 10 sec, 10 hours to render)
agep
I'm playing around with fluids, but I have a few questions. Is there a way to stop the flow of fluids, while still simulating the fluids that is already been created? Also, it seems the fluids wont cast shadows (eventhough the "cast shadows" is set to "ON" in the material property)
jzawacki
QUOTE(agep @ Jan 8 2008, 06:34 AM) *
I'm playing around with fluids, but I have a few questions. Is there a way to stop the flow of fluids, while still simulating the fluids that is already been created? Also, it seems the fluids wont cast shadows (eventhough the "cast shadows" is set to "ON" in the material property)


Isn't my fluids casting a shadow in this thread: Fluid floating away?
michaelangeloart
The .prj file opens in the browser window as a text file, and won't download.

Any thoughts on this?

Curious,

Michael Angelo
John Bigboote
I believe when it opens as a text file in your browser that you would then 'Save As' and make sure it has the .prj extension.
johnl3d
right click and choose save as and it should come as a project file
agep
Does anybody else have problems with fluids? Im getting "Exception #035. Problem transforming hair/particle system." when I'm trying to render. I've tried many projects: The tutorial project, all my own projects, started new projects... I'm running 15c, tried to uninstall and reinstall, no luck

*edit* I've sent it to AM Reports, but I would like to know if I'm the only one with the problem
frosteternal
QUOTE(agep @ May 23 2008, 04:20 AM) *
Does anybody else have problems with fluids? Im getting "Exception #035. Problem transforming hair/particle system." when I'm trying to render. I've tried many projects: The tutorial project, all my own projects, started new projects... I'm running 15c, tried to uninstall and reinstall, no luck

*edit* I've sent it to AM Reports, but I would like to know if I'm the only one with the problem


Report has been made, is fixed in v15D
Rodney
QUOTE
I've sent it to AM Reports, but I would like to know if I'm the only one with the problem


Even when I install v15c A:M is reporting that it is running v15b so I'm not sure how accurate my testing is here.
I *think* I am running v15c...

Rendering the Fluidtest project found here in this topic works well in both quick and final render.

The referenced fix for the forthcoming(?) v15d can be found at A:M Reports here: http://www.hash.com/reports/view.php?id=5130
agep
Thanks! Rodney, have you tried to render several frames? I often got the error message on the second frame
ruscular
When I was working with fluid, I got it to render in action window but would not work in Choreograph window.

Here is my project that I did for a commercial.

http://www.gentlechifitness.com/wip.htm

click on baby being fed with a new spoon on the right side.
johnl3d
at announcement for 15c I posted a qt showing the multi colored fluids

http://www.hash.com/forums/index.php?showtopic=32173
Rodney
QUOTE
have you tried to render several frames? I often got the error message on the second frame


The realtime render was multiple frames but the final render was only one frame about 3/4 of the way through.
I'll have to try a sequence when I find a little more free time.
Also after downloading and installing again A:M is finally reporting v15c now... so that may make a difference.
agep
I'm really looking forward to the fix. I got a few projects I eager to test out
case
Just wondering but is the fluid controlled by newton physics?
Fuchur
As far as I can say no.
Newtonphysik has a "Fluid"-option, but it is only available for closed surfaces (so you can simulate a ship, dancing on waves).

*Fuchur*
John Bigboote
No...A:M fluids are their own entity... do not confuse: A:M Fluids...A:M SimCloth...Newton Dynamics
agep
Just testing fluids. This time I tried to combine it with movie clip which I tracked. The render settings are low, not much shadows, I might tweak them later. Anyway, here you go:

Click to view attachment
NancyGormezano
QUOTE(agep @ Jul 24 2008, 02:17 PM) *
I just did a new fluid test.


Oh Funnnneeeeee....and Yuckeeeee...

made me laugh.

John Bigboote
Gotta get that roof fixed, Stian.

What was the test? What was 'broken' that is now fixed?

OH- and what tracker software are you using?
agep
Fluids would not render in 15c, got an error message. But this is fixed in 15d. I used PFHoe 2.1 for the tracking
case
I was asking because it seems like newton physics having a channel in fluids would be a cool idea for more fluid like water effects...Just something I felt like i needed to know LOL!
This is a "lo-fi" version of our main content. To view the full version with more information, formatting and images, please click here.
Invision Power Board © 2001-2012 Invision Power Services, Inc.