did you have a chance to take a look at the scene I sent to support? any comments, suggestions?
pl
did you have a chance to take a look at the scene I sent to support? any comments, suggestions?
pl
Hey,
Still hasn’t reached me yet. Let me bump the support guys…
Cheers,
Would you please let us know the ticket number, subject or sender email address since I am not able to find your case into our system?
Thank you very much in advance.
Support Ticket Opened [#336246]
Got it! Will check it and get back to you ![]()
Hey,
At long last I have an update on this issue - the foam did not die exiting the grid as it looked to me at first, but instead just stopped there. The reason for this was that foam behaves differently above and below the liquid surface - when it flies through the air, it moves ballistically and is affected by gravity and the air drag, which you can control using the Falling Speed parameter. When the foam particles are under water, they directly use the velocity of the liquid at the position of each particle, and the Rising Speed moves them up towards the liquid surface. The problem in your scene was that the surface of the liquid inside the simulator was a bit lower than the surface of the surrounding ocean, so particles that were floating above the water surface and moved sideways, crossed the container walls and stopped immediately, because they found themselves under water, and their velocity was set to the velocity of the liquid outside the simulator, which is zero. There is nothing wrong with your setup and whether this happens or not is just a matter of chance.
So, for Monday’s nightly I made a change that allows foam particles outside the simulator and below the liquid surface, to continue moving in the direction they were headed before exiting the simulator. This would allow the foam to spread out of the container without getting cut off like that. The velocity of such foam particles outside the grid is now damped with time, so they won’t continue to move to the sides infinitely - this is controlled by the Rising Speed param now - I am still wondering if we should have a different parameter for that. However, it would be best to have the container as wide as the foam would spread, otherwise if the velocities near the borders turn towards the inside of the grid, they would not be able to affect the particles that are already outside and then a visible rift in the foam wiil appear there.
Something else that I noticed is that you have foam amount set to zero, which eliminates all natural foam birth and you depend only on the ‘foam on hit’ setting for creating foam from splashes. I think that you can get more rich and diverse foam if you allow it to be born both ways and find the right balance between the amount of naturally born foam and foam born from splashes.
Hope this helps, cheers!
Hi Svetlin,
I am wondering how if at all possible to make fire / smoke particles to live outside of the grid. If I have a adaptive grid, when smoke opacity goes to 0 then grid shrinks and it kills the particles too.
Hi Svetlin,
Thanks (a bit belayed:() for your resp[once. All you say makes perfect sense and I decided to try your tips.
I took a scene I don’t remember what I simulated it with, last year, (https://www.youtube.com/watch?v=uSinQowodbc&feature=youtu.be 1000 frames simulation, folder size 111 GB) deleted the boat and put another boat in the same simulator.
Following your recommendations, I changed the Foam amount form 0 to 1. Started the simulation last night and this morning I noticed it was not not as far as I thought I would be.
I noticed the executive foam particles count, changed the the foam amount from 1 to 0.5 and let it continue. few frames later the sim had stopped due to overflowing the 250 GB SSD the simulation was running on.
my question is: it it to be expected a scene like this:
to generate frame files of 2+ GB:
or should I be supposition of something wrong going wrong with my scene?
simulated with:
cheers
pl
Hello,
170 Million foam particles is quite high particle count and it will lead to big cache files. If you have all the channels exported for the particles (pos,velocity,age,id) this will increase the size as well.
Do you mean the drag particles? This is a good suggestion, I think it won’t be hard to implement it. Thanks!
It seems normal - 160 million foam particles is quite a lot, and furthermore it depends on what foam channels are exported - if you have velocity, this would export a huge amount of data per each frame and even though the particles are compressed, the caches could go over 2 gigs per frame.
Seems like you need to reduce the Foam birth amount, but also watch out for the Foam on Hit amount as well - this one is a separate control and it could add a lot of foam as well. Come to think of it, we should add a way to either observe in the viewport, or at least count how many particles were generated either way, so you won’t have to find out by trial and error.
Hope this helps, cheers!