Basic questions

Discussions about the use of Realflow 2012
Builtdown
Posts: 14
Joined: Wed Oct 12, 2011 11:58 am

Basic questions

Postby Builtdown » Mon Oct 24, 2011 9:23 am

Hi,

I´m collecting here some basic questions that come to mind.


1. When I use normal Particle Emitter with high viscosity values, violent small explosions happen all over the place. Is there any way to avoid these?
(When I do the same with Grid Emitter, it´s no problem however...)

2. I want to make an animation of small play-doh-drops dropping on each other. I use Grid Emitter + Standard Mesh and the dropping works well. However I would like that the drops wouldn´t "merge" to each other but to stay as more "stiff" drops on each other. I have tried to adjust the viscosity/compressibility etc but still the mesh keeps merging to one large ball. Any tips of the settings needed for this kind of job?


Thanks!

User avatar
Otuama
Posts: 1155
Joined: Fri Apr 02, 2010 12:56 am
Location: Swansea
Contact:

Basic questions

Postby Otuama » Mon Oct 24, 2011 9:49 am

Ok, when using high viscosity values you need to change the substeps. The reason for the exploding is because your substeps are too low.

I'm sure PixelPro won't mind me using his VIDEO as an example.

If you look through the comments you find that he had a viscosity value of 200 & max substeps at 2000-3000. If you search for 'A3' you'll see someone with the same problem as you.

----------

For the play-doh simulation you're going to need a VERY high viscosity setting. Also, the reason they're merging into 1 large ball is probably because you have them attached to the same mesh.

----------

NOTE!! Increasing the substeps will increase simulation time and could also crash your scene so save often and maybe make backups of saves.

-Sean
Otuama - Vimeo
'Strength does not come from physical capacity. It comes from an indomitable will.' - Gandhi
'Realism is in the eye of the beholder' - Me ;)

Builtdown
Posts: 14
Joined: Wed Oct 12, 2011 11:58 am

Basic questions

Postby Builtdown » Mon Oct 24, 2011 11:51 am

Thanks for the answers!

I tried increasing the substeps (in Preferences?) but didn´t see much difference with exploding particles. But that´s not a problem because I think the Grid Emitter works better with high viscous stuff anyways.

For Play-doh, using different meshes is a good idea!


Next questions :


3. When I add an object (like a sphere or a vase), is it possible to increase the quality of the objects mesh? The vase for example is very "blocky".
Of course I can make and export objects from 3D-apps but for quick testing in RF the RF-objects are handy.

User avatar
Otuama
Posts: 1155
Joined: Fri Apr 02, 2010 12:56 am
Location: Swansea
Contact:

Basic questions

Postby Otuama » Mon Oct 24, 2011 12:31 pm

To change the substeps for a particular scene you need to click the arrow attached to the simulate button and select options.

I'll do a few examples of substep differences and put them on Vimeo. They won't be up for a while though.

Yeah, by default the Rf objects aren't a good resolution. To create Rf objects with a higher res, you'll need scripting...

Open the batch script editor (F10). Try these.

Sphere:

scene.addSphere(2) # Default setting
scene.addSphere(3) # Medium
scene.addSphere(4) # High
scene.addSphere(5) # Very High

Vase:

scene.addVase(10,1,2.0) # Default setting
scene.addVase(20,10,2.0) # Medium
scene.addVase(40,10,2.0) # High
scene.addVase(40,20,2.0) # Very High
Otuama - Vimeo
'Strength does not come from physical capacity. It comes from an indomitable will.' - Gandhi
'Realism is in the eye of the beholder' - Me ;)

Anaxarchos
Posts: 843
Joined: Wed Mar 19, 2008 4:21 pm
Contact:

Basic questions

Postby Anaxarchos » Mon Oct 24, 2011 1:15 pm

I'd not use grid fluids for a setup like this as at all. Standard Particles are better for this scenario as you will have more control and you can have far more more detail in your sim.

Set substeps to a high settings, as said before do not use the settings in Preferences but click on the small arrow next to Simulate. Maybe 50 (Fixed) or 50 -100 (Adaptive should work.

When playing with high viscosity, do away with Surface Tension and set this to 0.0.

Anyway, did you try simulating these Play- Doh drops with nCloth? I think this could be easier.
www.fk-fx.com

User avatar
Otuama
Posts: 1155
Joined: Fri Apr 02, 2010 12:56 am
Location: Swansea
Contact:

Basic questions

Postby Otuama » Mon Oct 24, 2011 1:52 pm

Yeah, but a substep setting of 50/100 is going to be way, way too low for play-doh.

The nCloth idea is a good one. It should be much faster as you can set the rigidity of the cloth.
Otuama - Vimeo
'Strength does not come from physical capacity. It comes from an indomitable will.' - Gandhi
'Realism is in the eye of the beholder' - Me ;)

Builtdown
Posts: 14
Joined: Wed Oct 12, 2011 11:58 am

Basic questions

Postby Builtdown » Mon Oct 24, 2011 2:25 pm

Thanks for the tips!


I did a test with Particle Emitter, ground plane (box object) and a Standard Mesh to produce something similar to Anaxarchos websites caramel animation.
Viscosity is 200, surface tension 0.0
I also adjusted the options next to Simulate button : Adaptive, MIN Substeps 50, MAX Substeps 100.

There´s still some nasty explosions going on...

By the way, Anaxarchos, how did you manage to keep the "creases" so beautifully separate? And how is the mesh not flickering at all?


Question 4. :

Is there somekind of method to "expand" a liquid? Let´s say I make a shaving foam to fill a cup to the middle. Then it would slowly expand so the cup would eventually be filled. Is this possible?

Anaxarchos
Posts: 843
Joined: Wed Mar 19, 2008 4:21 pm
Contact:

Basic questions

Postby Anaxarchos » Mon Oct 24, 2011 4:43 pm

Try simulating at 100FPS, this works good for me when doing high viscosity stuff. What did you set for viscosity?

The foam effect you were talking about could be achieved with some scripting.
www.fk-fx.com

User avatar
Otuama
Posts: 1155
Joined: Fri Apr 02, 2010 12:56 am
Location: Swansea
Contact:

Basic questions

Postby Otuama » Tue Oct 25, 2011 2:59 am

I've put a video on VIMEO showing different results with substeps.

For something like play-doh, you're going to make things much harder for yourself using Realflow. I'd try the nCloth route.

The first batch of tests were with a viscosity of 300.

When I tried a much higher setting, still not high enough for play-doh, it took so long and the substeps still weren't high enough.

I then tried a higher setting but it was taking too long so stopped it.
Otuama - Vimeo
'Strength does not come from physical capacity. It comes from an indomitable will.' - Gandhi
'Realism is in the eye of the beholder' - Me ;)

Builtdown
Posts: 14
Joined: Wed Oct 12, 2011 11:58 am

Basic questions

Postby Builtdown » Tue Oct 25, 2011 7:03 am

Wow, thanks for the video!


So it seems that RF is pretty useless for super high viscous stuff...


Return to “Realflow 2012”

Who is online

Users browsing this forum: No registered users and 1 guest