Jump to content

Pazuzu

Members
  • Content count

    369
  • Donations

    0.00 CAD 
  • Joined

  • Last visited

  • Days Won

    22

Posts posted by Pazuzu


  1. Regards speed, I think it mainly depends on the method, but the vdb one is the best in terms of control, because you can continue to filter and reshape the field using the vdb arsenal. Sometimes the gas reinitializesdf microsolver is faster, but it can also become a weird bottle neck with huge boundaries, for example if you want to compute a gradient out of an sdf and you need the gradient to cover the whole volume container or to have a better extrapolation, in this case vdb can be a pain because of his sparse nature, of course you can expand or extrapolate your sdf using the amazing vdbactivatesdf, but that can be even slower than to use the microsolver directly.

    • Like 1

  2. I think is also because of the details. I assume this is a small scale thing, right? If so, you should use surface tension, right now it looks like a large scale one with lack of splashing detail, if you want more detail at large scale try to use the "swirly kernel" on the solver, this will preserve even more the momentum of the sim but will smooth a bit the violence and chaos behavior in comparison with the "splashing kernel", but will give you much better splashing details.

    Then on the skinning stage, you can try an old school method: with a VDB pipeline use: Dilate - Smooth - Erode. 

    Cheers!

    Alejandro

     


  3. 14 minutes ago, scorpes said:

    Really nice Pazuzu! How do you create the individual droplets coming from the crown, is that a separate particle emission?

     

     

    Thanks for the comment!

    Its part of the same simulation. That WIP shows a new option functionality of my small scale pipeline to control how much do you want the droplets to detach from the main fluid body. Its a very important artistic control in a simulation.

    Cheers!

    Alejandro


  4. This video will give you very good foundations on using the pop fluid! That way you don't need to "Abuse" anymore curvature on FLIP method to have "Beautiful" crown splashes!! ;)

    Cheers!

    Alejandro

    • Like 1

  5. I know that I posted this video in another post as an mp4.

    So here is a vimeo video that shows all the stages for this simulation result!

    I hope you like it!

    Thank you!

    Alejandro

    • Like 3

  6. 5 hours ago, moneitor said:

    Brutal viejo alejo, yo tengo ganas de hacer un ripple parecido para deformaciones de metal a gran escala, pero seguramente lo hare como un postproceso.

    Gracias viejo Hernan! Para eso hay una muy buena referencia, y podrias hacerlo si quieres de esta forma, el problema esque necesitarias tets (seria deformarlos de cierta forma para asi enganhar el tool ya que no salen directamente de FEM) y lastimosamente es una pesadilla en cierta geometria, pero creo que vale la pena pillarlo! Aqui te dejo un link:

    https://www.sidefx.com/forum/topic/48343/?page=1#post-219429

    Saludos,

    Alejandro


  7. 27 minutes ago, HowardM said:

    Thanks everyone, cool stuff!  Great shading on your last render Pazuzu!  Mind checking your messages?  Cheers!

     

    Hey Howard! Sorry about that! Here is the scene, I polyreduced a bit one of the frames of the sim to have a less heavy file to share!

    By the way, that mesh still has a freeSurface attribute; So you can use that gradient for nice shading effects.

    I forgot to say that this scene uses Redshift. v2.6.32; H17.459.

    Cheers!

    Alejandro 

    FLIP_crown_v005_share.hip

    • Like 4

  8. 6 minutes ago, moneitor said:

    Hey man, great job, are those micro ripples in the simulation itself, or you are exporting attributes to drive them in the meshing stage?

    Hablame en espanhol viejo Hernan, por lo menos pa que practique!! :)

    Estoy usando el gradiente del atributo que genero dentro de la simulacion para detectar el FreeSurface; Los MircroRipples en este preview son calculados en simulacion, pero a veces no se contrastan lo suficiente asi que los contrasto a nivel de mesh tambien para que se noten mas aun, ya que al rasterizar las particulas se pierden si no se usa una resulucion muy alta, por ende opte tambien en tener una solucion que no necesite resoluciones muy altas para hacerlos mas notorios.

    En este preview podes notarlos directamente en las particulas!

    Saludos

    /,,/ /,,/

    crown_v004.mp4


  9. 6 minutes ago, Atom said:

    Nice work, does this make use of the new H17 retime node?

    Thanks Atom!

    For this retime I used an old tool that I developed years ago to do almost the same as the new retime tool, the main differences are the nice interpolation options for attributes that the new H17 one has, mine just have linear as interpolation option, so I'll migrate very soon to that one.

    Alejandro


  10. Here is another test using a last iteration of my tools to do small scale stuff. This time, now is possible to art direct how much do you want to break a sheet into drops and tendrils, and there are some Micro Ripples on the sheets that are relative to the fluid speed and curvature at some portions of the main fluid body!

    oil_details_shading_cam01_v001.mp4

    Thank you!

    Alejandro

    • Like 3

  11. 2 hours ago, Jesper Rahlff said:

    big question for me would be if Vray is still faster rendering volumes.. Not sure it would be. Mantra is not doing to bad when it comes to volumes

    I think Mantra is faster when rendering volumes, but the main issue is when you need multiple-scatter; Right now the path tracing alternative to point clouds is extremely slow in comparison with Arnold for example. Lets hope that H17 surprise us with some improvements for Mantra in volumetric rendering.


  12. 3 minutes ago, toadstorm said:

    It's actually possible to use op:/ syntax to point at SOPs in a shader, as long as the SOP is the output (or upstream of the output) of an object listed on the Mantra ROP. If you don't actually want that object to render, you can add the vm_renderable flag to the container and set the value to 0.

    Yeah I know! I was meaning that not directly, you need to tweak first to make it functional. Even when you tried to render that's what the command line tells you! I didn't explain my self correctly! :)

    • Like 1

  13. 42 minutes ago, logix1390 said:

    hello,

    Is there a way to group the particles that are only effected by the pop advect by volumes pop? I can do it by proximity but it creats a perfect circle around the effected area, which is not what i want.

    Thank you

    Hi,

    You can create a group, testing the speed of the sampled velocity. So particles with a speed threshold below some value get out of the group.

    Cheers,

    Alejandro


  14. I'm working on something related to art direct the swirly motion of gases; Its an implementation of a custom buoyancy model that let you art direct very easily the general swirly motion of gases without using masks, vorticles, temperature sourcing to have more swirly motion in specific zones, etc. Also it gets rid of the "Mushroom effect" for free with a basic turbulence setup. :)

    Here are some example previews. Some with normal motion, others with extreme parameters values to stress the pipeline.

    For the details is just a simple turbulence + a bit of disturbance in the vel field, nothing complex, because of this the sims are very fast (for constant sources: average voxel count 1.8 billions, vxl size 0.015, sim time 1h:40min (160 frames), for burst sources, vxl size 0.015, sim time 0h:28min).

    I'm working on a vimeo video to explain more this new buoyancy model that I'm working on.

    I hope you like it!

    Cheers, 

    Alejandro

    constantSource_v004.mp4

    constantSource_v002.mp4

    burstSource_v004.mp4

    constantSource_v001.mp4

    burstSource_v002.mp4

    burstSource_v003.mp4

    burstSource_v001.mp4

    constantSource_v003.mp4

    • Like 5
    • Thanks 4

  15. Hey Ezequiel!

    No hay problema!!  El grupo es porque, al usar los boundary layers hay un reeseding (o creacion dinamica de particulas) en la zona de padding, asi que se debe crear un grupo o definir un atributo para asi diferenciar las particulas justo antes del reseding; Asi al transferir attributos solo las nuevas lo recibiran concervando asi el color sin sobreescribirlo en las particulas ya existentes.

    Saludos,

    Alejandro

    • Like 1
×