Splash particle birth issue

[TSR_trix]

Im having a problem to generate splash particles when system unit scale is set to meters.
Even if I had threshold 10, I have to bump up the birth rate in to its upper limit (1000000,0) and still I got only 20 000 particles.
My scene is a river, but them main event is to drive a car through it so I would like to have a good control for the splashes when the tires touches the water.
Now, cause the threshold has to be low, it affects too much for the whole simulation.

Btw Ivo, is the Forward transfer still broken in 2.1? In my scene it looks better with FT than with classic

[Ivaylo Katev]

the particle birth is bound with the scene units, because if we use the cell size as base, the change of the grid resolution will change the particle production dramatically.
if you have system units meters, and the entire grid is few meters, this can explain the problem, because in this case the cell volume is very small. keep in mind that the volumes decrease 1000 times when the linear size is just ten times smaller.
if you have good solution how to make the particle birth more persistent at different scenes and scales, i will be really thankful. it's already considered to be related to the grid volume, but this will produce problems with the adaptive grids.

[TSR_trix]

Thanks Ivo.
Now I remember to asking this same question once before...
But its just hard to take consider of all the main factors when starting a scene. So, if my simulation is inside few meters, I should use 1unit=1cm and real world scale objects. But if the container is over 10 meters or so, then 1unit=1meter could be better.

[Ivaylo Katev]

btw, the particle sources in the last version are with changed control, they were like the natural birth in 2.0, i.e. volume based (particles per unit volume), however now they are just particles per second. this is possible because when we construct the source, we have the volume and can calculate the internal birth rate in order to produce predictable amount per second.
unfortunately this solution is not applicable in case of natural birth , because we do not know the future birth conditions.

[TSR_trix]

I continue here with other question, but this time about foam birth.

I have a liquid pouring scene with animated transform source object.

I get quite nice results by using strong surface mode. It makes the liquid to be smooth and good looking.
The problem is foam threshold. If I remember right, it acts differently when strong surface mode is OFF? Without it, I can control threshold much better, but with it I have to use quite low value to get even few particles.
And because my source has transform animation, it starts to burst undesirable foams form the very near of the source.

I can avoid this by running the simulation without SSM and have a good control with threshold, but the look is not the same.

have you had any thoughts to make a gizmo for excluding areas from the particle birth? This would solve many similar particle birth problems at least I have had.

PS. Any news about rendering gizmo for the meshed liquid?

[Ivaylo Katev]

yes, strong surface mode makes the particle birth problematic. how old is your build? the foam birth was totaly reworked the previous week in order to improve the foam birth position and control.
we are about to make particle cleaner that removes the particles from undesired positions. it will affect the birth as well.
about the gizmos and mesh mode - the gizmos are relatively hard to implement, but the compound objects should give the same result, unfortunately they do not work now, this is a bug. in one or other way we will have working gizmos.

[TSR_trix]

Thanks Ivo for the exhaustive answer!
Im using official 2.1, now Im gonna update it again to the latest nightly and make another test with strong surface mode with particles. Lets see how it works!

[Ivaylo Katev]

the boolean operations were fixed too, with the tomorrows build you can use them instead gizmo

[TSR_trix]

Ive been missing this boolean feature, can it be found under Phoenix or is it max standard boolean operation?

[Ivaylo Katev]

it's the max boolean operation.