Annon Posted May 24, 2011 Share Posted May 24, 2011 There must be something wrong with my scene (and all other ones I create). If I make just a simple sphere and get it to emit flip particles from the particleFluidEmitter and collide with the floor then it's totally fine. Adding in some proper geometry with a group to emit from slows it down to unusable speeds for me, especially at these low settings. Here's my scene, I hope it's just something I'm doing wrong and there's a way to speed this up. Any help is much appreciated as always. Christian P.s. I created a ROPOutput to make things faster, but it doesn't touch the sides, so you can just bypass it... ldev_ocean_cjw_v003_t01_flipTests.hip Quote Link to comment Share on other sites More sharing options...
anim Posted May 24, 2011 Share Posted May 24, 2011 your volume is too dense go into configureflip1 and multiply Division Size with reasonable number. Now it's multiplied by 2 so you may try 5 or 10 it depends on detail you want your volume velocity to have (of course you may consider using just dummy particles instead of flip for this) Quote Link to comment Share on other sites More sharing options...
Adam Ferestad Posted May 25, 2011 Share Posted May 25, 2011 I saw a video somewhere where the guy did a very low quality fluid simulation and used a wedge ROP to simulate it several times to disk, then merged the several cached files to make a higher quality final. He had the wedge set to increment the seed value on the simulation. I don't really know if this applies to your situation as I have not worked with Flip Fluids myself yet, but I hope it helps. Quote Link to comment Share on other sites More sharing options...
Jason Posted May 25, 2011 Share Posted May 25, 2011 You might want to download a more recent Daily Build; SESI have just made an improvement which could help you. Monday, May 16, 2011 Houdini 11.0.747: Optimizations to the liquid fluid solver and FLIP solver to better handle large empty areas. Mostly empty volumes have a much faster setup time. Quote Link to comment Share on other sites More sharing options...
Annon Posted May 25, 2011 Author Share Posted May 25, 2011 You might want to download a more recent Daily Build; SESI have just made an improvement which could help you. Thanks everyone, I was originally trying it with SPH and getting slow results, so swapped to FLIP to check it out and forgot to sort out my domain like an idiot! Thanks for pointing it out to me. I'm back to SPH and have got things running at a reasonable speed at the minute, but will get them to update to the newer version and see if that helps... Thanks. Christian Quote Link to comment Share on other sites More sharing options...
Jason Posted May 25, 2011 Share Posted May 25, 2011 Ah, I see However, with those new empty-space changes, I would think that FLIP could more easily handle the type of sparse fluids that you'd usually need to switch to SPH for. Might be worth another spin? Jason. Quote Link to comment Share on other sites More sharing options...
Annon Posted May 25, 2011 Author Share Posted May 25, 2011 Cool thanks, I'll get them to install the newer version tomorrow and get back on my findings. On a quick note, I've moved back to SPH and found that the pesky little "Accurate Births" checkbox was to blame for low speeds. At the scale these particles are going to be at I don't need them, and usually use the trailSOP to fake it anyway. Things are now moving at a decent speed with 50x more particles than before.... Cheers Christian Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.