cleaning process in slow motion

[Ivaylo Katev]

the problem is resolved, but is still not uploaded to the builder , hopefully until the end of the week will be.

[Jassir]

That's great. Thanks a lot for this.

[Jassir]

Hi,

is the problem fixed in a current version?

[Ivaylo Katev]

yes, it should be

[Jassir]

Hello,

I am back to this project. I simulated the water again with a higher resolution. Now my render times are significantly higher. I am not sure if this is a vray or phoenix issue.
When I started the project the render times were fine. So I am kind of confused.
The scene renders in ~2 minutes (frame "108") if I hide the water. If I unhide the water it takes ~45 minutes? The material is rather simple.

I uploaded the testscene if anybody wants to take a look testscene.zip

[Ivaylo Katev]

i will see the scene, but it's very probably this is expected behavior

[Jassir]

Is there a way to increase render speed without loosing to much quality?

[Ivaylo Katev]

well, the first tip is to use mesh rendering mode, not implicit surface that is used in your scene.
but the most important is the particle rendering. you have one particle shader rendering the splash of the simulation. when the scene is relatively complex, the foam shader may lead to very big render times, because it is volumetric shader and slows down each ray passing near to the particles, even these that are not actually affected. the best solution is to render the splash in separate pass and add it in compositing,

[Jassir]

I managed to get reasonable render times, thanks for your help.
I am back to particle flow and PhoenixForce now. I distributed particles over my object and added a deflector that follows the water jet. If the deflector hits the particles they are transmitted to a new particle flow event that contains a PhoenixForce operator with the water jet as Phoenix node. I was expecting that the particles therefore are driven by the water simulation. So they would stop if I would freeze the frame of the water jet by activating direct frame mode after some frames. Unfortunately they won't. Was my assumption wrong? Do I have to freeze them in another way?

[Ivaylo Katev]

they are driven by the velocity simulation in the current frame, that means even if you have "frozen" by the input simulation, they still follow the velocity found in the currently loaded frame. the velocity is the same as when you play the animation with normal speed. well, this scenario was predicted, tha's why the option "ignore time scale" was added, but unfortunately the play speed was introduced later, and this option gives in account only the simulation time scale, not the play speed. well, we will fix this

[Jassir]

I'll find a workaround for now, a fix would be great though.