Jump to content


  • Posts

  • Joined

  • Last visited

  • Days Won


CeeGee last won the day on September 16 2021

CeeGee had the most liked content!

About CeeGee

  • Birthday 05/24/1983

Contact Methods

  • Website URL
  • Skype

Personal Information

  • Name
  • Location
    Belgrade - Srbija

Recent Profile Visitors

12,660 profile views

CeeGee's Achievements


Newbie (1/14)

  • Week One Done Rare
  • One Month Later Rare
  • One Year In Rare

Recent Badges



  1. Some of your settings are bit strange, maybe you want to use them like that not sure. But i did some cleanup and tweaks to your initial tank, to use narrowband and adaptive solve to make it faster so you can push more details. broad_splash_02.hip
  2. Hi, I did few rnd for distribute simulation and you can do distribute simulation on single workstation, for example if you have 64core you can split 2-4slices sometihng like 32core or 16core per slice. But you need to have a lot of memory here are some stats and tests - http:// https://vimeo.com/380432448 In your case 14 core with 64GB of ram, is not good option for single workstation distribute simulation. Also simulating on linux can be a lot better then on windows, also simuilating something with flip is more of optimization to solve problem then actual brtue force. You can see what Pixomondo did on Midway https://www.youtube.com/watch?v=kJ8Hz2Bjejc&t=10962s
  3. You shark anim is realy fast, i would advise to run with bit more substep to get better collision. You can also animate shark vel maybe to reduce when shark is going out.
  4. Can you try this scene, You collsin geo was setup incorect, no velocity, or vdb for collision. flip_Sticktest_v002.hip
  5. Its realy depend of your simulation, one thing that you will notice for sure is 3x less memory usage, but speed its diffrnet thing, sometime i have maybe 50% faster, sometime 2x faster, but only if my container is deep like 5-20m depend of shot. So speed that you are getting i think its more then ok, but be sure to chk memory this is where you will have more save compare to full tank.
  6. I need to chk for scene because i don't have permission to share it, but let me try to explain. Mesh part I just split particles using speed and vorticity, and low speed/vort areas i used vdb from particle fluid with some smooth mask, same as default particle fluid mesh do and on high speed/vort values i add high res settings on vdb from particles fluid and then just combined 2 vdb volumes and convert to poly, thats all Rendering part - its just fluid mesh with displacment on top, using speed and vorticity to add bit of color, and there is simple bubbles and foam sim render as points under surface. - no this is not like whale sim, on whale i did convert surface filed to mesh and then add displacment, but because i wanted to get all details from sim, i did fluid mesh on this one.
  7. Setup is very easy, one button Everytihng is same as before with Houdini Flip, this narrow band just save a lot of memory, now you can make deeper tank sims that give you more realistic look, and in past that you are limited with depth because of particle count. I don't like to compare with Bifrost, because Bifrost is till closed you don't have that control like in Houdini, when they open there node graph stuff then its going to be easier to compare.
  8. With shark sim there is not secondary sim, i treid to get all from single sim with nice mesh details
  9. I used 2 workstations on this i7 4790 with 32GB of ram, and 2 x Xeon E5 20core per cpu with 128GB of ram. Most of sim times are 5-15min per frame, using from 25-120GB of memory,if you think with full tank most of sims are going from 400-900mil particles, but with narrow band its much easier now. Spaceship sim was 120GB of memory because at the end it was 1.1Bil particles, you have some details on single spaceship playblast, particles and voxel count. Igor
  10. I had chance to test new Flip NarrowBand workflow and wanted to share resaults with community
  11. Hello Guys, Saber Jlassi and Igor Zanic are glad to announce the formation of REBELWAY CO, an online school that will focus on producing the most high-end online training. Our goal is to teach Sidefx Houdini primarily along with high-end renderers like SolidAngle Arnold and realtime rendering technology like Unreal Engine 4.0. We are proud to announce our very first workshop aimed at teaching Procedural Assets Creation In Houdini. This workshop is for all 3d Artists who want to dive into Houdini and use it for environment creation for VFX and Games. Promo video: https://vimeo.com/sabervfx/rebelway Course link: https://rebelway.net/advanced-asset-creation-in-houdini-for-vfx-and-games/ Note: all the renders are straight from arnold I hope you guys like the content and find it interesting.
  12. Large ocean is plane with displacment, cut in middle where flip sim is. you can find similar workflow in this tutorial https://www.fxphd.com/details/?idCourse=432 , this is complete tutorial for this kind of effect. Yes in my sim splash go over, because this is just RnD, for final i would make bit bigger container and also stronger advection around edges to get better blending. Most of my splashes come from simulation, only strong impact ones are doing using cookie from ship and ocean, and used that as extra emiter thats all, but most are from flip sim.
  13. Hi, I wanted to create better setup then waveLayer and splash tank, so i manage to create my own guideLayer using mix of waveLayer and splashTank workflow All sim details, core sim and foam sim count you can see on videos. video01 video02
  14. http://ibi.ac.ir/ff92c

  • Create New...