what is the problem here???

[Ivaylo Katev]

ok, the problem is fixed

[themaxxer]

thanks guys... until now it seems to work.

[themaxxer]

it works... but after 340 frames max crashed... but that's a other problem. sadly this happen's alot. anyway... my question is... i would like to make a preview grab viewport. i only can render out the mesh without the foam and the splashes. do i have to enable something in phoenix or max making it work? in the viewport they display correct

[Ivaylo Katev]

i think it is related to max, because it does some "smart" filtering of the objects and perhaps it hides our content. try to tweak the preview objects type, perhaps particles are not allowed.
btw, trying to pause the flip solver simulation may cause hang of the max, this includes the sim time rendering as well. it's known issue, just avoid pausing or rendering without saved scene.

[tsetso]

also make sure that "Only if selected" is unchecked before you start the preview

[Svetlin.Nikolov]

The pausing issue with the FLIP solver should have been resolved on 29 jan.

[themaxxer]

hi guys, messing around with my scene... i remodeled different parts, made it wider, more interessting. i have to say unfortunatelly i still have shooting splashes. it's in areas where my stones lay on the ground in clusters. maybe to small ones. i have a question does surface tension have a impact on a flip fluid simulation? do i need it? what is a realistic parameter for foam bubble size. i mean enough to build patterns and stick together? and how long should i leave the bubbles life? my half life parameter is 0.3. is this to small. should i increase it? you know i want to make a realistic scene. thanks in advance

[themaxxer]

what i saw by testing the scene with viscosity, is that the lowest value 0.001 is far to viscous... the difference between no viscosity and it enabled is to extreme. in nature there is viscosity in every fluid, correct me if i am wrong? can't you lower the difference between those two values and make it more subtle. do you know what i mean?

[Ivaylo Katev]

the surface tension is not implemented in the flip solver.
the normal bubble size is about 5-10 mm, but the behavior is very sensitive to the bubble count, sometimes you have to use bigger bubbles in order to keep the count not too big.
really even the smallest viscosity changes the behavior rapidly, actually you should not use viscosity if you simulate pure water, there is enough undesired viscosity introduced by the numerical methods. use the viscosity only if you simulate creams, paints, chocolate etc.
about the particle bursts - other users also reported such a problem and we already have scenes, but i can't promise a fast fix, so my suggestion is to try to workaround them. export the velocity and use the viewport cut off value to determine where is the problematic point, then just change the geometry or even overlap the problematic lace with a stone or something. the usual problems are related to objects smaller than two cells in size. this may include thin objects as well, not only small objects.
another trick to find the problematic points is to export the "patches" of the geometry, select the simulator and type in the listener "$.zmr=23", then simulate. enable the smoke in the preview panel and you will see the "patches" in the grid.

[themaxxer]

hi there, everything is running ok, until this new problem here. i got foam everywhere. what's the problem here?

[Ivaylo Katev]

i suppose b2b instability, which build is this? i think you don't need b2b here, just disable it.

[themaxxer]

it's the build phoenixFD_adv_22000_max2015_vray_30_x64_25606. what i always simulated with b2b interaction to get a realistic simulation. why should i disable it now??? foam pattern need b2b. in this case you would run this simulation without b2b" can you explain me why?

kind regards david

[Ivaylo Katev]

of course we have to fix the instability (if exists), i just suggest a workaround for you, the water here is not settled and the foam does not have time to form patterns and any other b2b related effects. i'm actually not sure that really this is the reason. it may be not the b2b, because it seems stable to me, it may be the boundary conditions of the water. i observed that some geometry formations produce very high local velocity and the foam that is born near to these locations is thrown away. just disable the b2b to see what happens. no need to simulate the entire simulation again, restore it somewhere in the middle, and after few frames we will know whether or not my suspicion was correct .

[themaxxer]

thanks for your suggestion... the thing is i tested the scene with and without b2b interaction and patterns and there are differences.i modelled the rocks with pools to get this effect. i also tried the restore button but it never worked for me. i don't understand how to use it. it always begins the simulation from frame 0 again, sometimes crashes. is it the same method like in realflow the initial state? how can i use this from every simulated frame? i saw something "you can restore from here" but these are only a few frames and i can't see wich after i simulated the scene.

[Ivaylo Katev]

yes, there are few restore frames because they are bigger and will consume too much space. when you restore the simulation the nearest checkpoint frame behind the time slider will be used as starting state.
there is another way to find what causes the strange foam behavior, enable the velocity preview and set very high upper limit (let say 100000) and start to increase the lower limit. the displayed cells will start do disappear and when the lower limit is high enough only the abnormally higher velocity will be visible in the viewport. you can send me a cache file to investigate this, the cache file must have velocity channel exported, you can use some restore frame, it contains all the channels