Jump to content

gillsp

Members
  • Content count

    43
  • Donations

    0.00 CAD 
  • Joined

  • Last visited

Everything posted by gillsp

  1. Pyro shockwave

    Hi, I have been experimenting with shockwaves, but they are rather huge and takes up a lot of memory.(travels hundreds or thousands of meters) I have been trying with just a torus expending outward, the results are okay, but the most detail I could get was 80 x 30 x 80 with 0.1 resolution(to get fine details). This is so small for a shockwave, but if I just scale the sim later, the voxels become very visible. impulsewave.hip
  2. Pyro shockwave

    okay, sorry. here it is. But I think I will have to settle with a more simplier shockwave than the one in the link. Because I only got 32 G of memory. So any approach of cheating the effect will be fine also.
  3. Pyro shockwave

    anyone?I have searched up the forum...but none of them has the solution. Any help is appreciated.
  4. Pyro Shockwave

    Hi, I have been experimenting with shockwaves, but they are rather huge and takes up a lot of memory.(travels hundreds or thousands of meters) I have been trying with just a torus expending outward, the results are okay, but the most detail I could get was 80 x 30 x 80 with 0.1 resolution(to get fine details). This is so small for a shockwave, but if I just scale the sim later, the voxels become very visible. impulsewave.hip
  5. large scale mist

    I have come to the last step of my sim, I have about 3mil particles to source the mist. Though using the shelf tool, my memory gets full pretty quickly in about 48 frames(already cached the sources), not to mention I cannot see the mist(or the pyro/smoke) by just the using the mist shelf tool. Anyone has a better way in doing mist? Reference is here: Small trails of mists along with the fall, and large mist/smoke from the bottom.
  6. large scale mist

    Sure, marty. Basically, I took a look at the mist solver, it's basically a whitewater shelf tool, but with emitters that will emit pyro(smoke) and particles will push around them ( advect). The mist source is essentially the same as the whitewater source, but build with pyro sourcing. Mist shelf produces finer spray than the normal whitewater shelf tool(which is more ideal to produce just foam and bubble in my opinion). The whitewater shelf tool will create spray that is better used in smaller scale. I figured the thing that was eating my memory was the emitters that creates finer spray(particles). So I just took that out and I was fine just to simulate with the pyros.
  7. large scale mist

    nvm, figured it out myself...
  8. So I was doing some flip simulation, and the highest I managed to get is around 50mil to 60 mil particles with one simple flip solver.(already deleting some attributes) with 32GB of RAM. But when I was doing some other tests, with a few more microsolvers, I was only able to get to not even 20mil(and my computer tells me not enough memory). My question is what is the main methods to optimize the scene except particle counts and deleting useless attributes? I knew Igor was able to optimize and get around 40mil(if I remember correctly) with only 16GB of RAM?Please help me, of coz, all other advises and help will also be greatly appreciated.
  9. Hi Igor, I was actually the guy who PM you on Vimeo, because I was not sure if you will be either one, we can handle the question with the vimeo pms or emails, thanks.
  10. WaterFall Scale

    Hi, I have been having struggles on how to make the right scale on a personal project. The setup is in the file. Basically is a waterfall running down a mountain(34 meters tall), somehow I think either my texture is giving a mountain bigger scale feeling or somewhat, the scale of the waterfall is not right(too fast). I have tried the mass scale and spatial scale, both gave me very little results that is hardly notiable. The only thing seems to work is timescale of 0.5 or reducing gravity, but the reducing gravity is not ideal in any case because it doesn't look right. Adjusting time scale is just like being in slow motion though. My question is how can I only adjust to speed of the running down waterfall without changing gravity or something else.(the emitter emits another flip), the targeted speed of the runoff is about half or 2/3 of the current simluation, without changing the overlook of the waterbed. Obj: https://drive.google.com/file/d/0B0-QsiyptV2gTFFpN0Q1QXNGTVk/view?usp=sharing setupv5highopttestv2.hip
  11. WaterFall Scale

    Well, the thing about breaking it into two sims is the collision between them. In houdini it doesn't allow you to have two different fluid objects that collide with each other, right?
  12. WaterFall Scale

    Any suggestions?
  13. WaterFall Scale

    Hi everyone, looks like timescale does the trick: https://drive.google.com/file/d/0B0-QsiyptV2gWWh4M3J0eVdwZ1dRTWcwSE5RVGxEdzZKT0Vv/view?usp=sharing However, because I want more break up of the fall, and the geometry of mine is a mountain that doesn't have any boundaries to control the fluids. So it tends to be thin and spreaded out. I tried to get more volume in it, but this is the highest detail I can get. (0.1 separation, 32GB ram, around 40mil particles at first frame, 0.09 will give me memory error on later frames).
  14. WaterFall Scale

    anymore thoughts?
  15. WaterFall Scale

    I just did some quick test, didn't seem to work, unless I have to push the values to some extreme(100 or 1000), but viscosity is a number that I don't really want to mess with because it will give weird simulation than water I think, as honey or mud. The best bet now is timescale.
  16. WaterFall Scale

    Hi, sorry I was at work. Let me try adding viscosity and frictin to see if that help. Well, the moutain is at a right scale of 34 meters tall, it's just maybe the texture of it is giving it a look of larger which makes the whole simulation looks small scale, perhaps the simulation is at right scale. But anyways, I did a test on timescale, I think it works better, though I am not sure if that is what I want to do. Im doing tests on viscosity and friction, see if those can help. Thanks.
  17. WaterFall Scale

    Any help will be appreciated, awaiting online.
  18. Rendering Foam particles

    Make your particles pscale small, drop a mantrasurface shader, set 0.7 for the diffuse intensity, leave the color white, and turn on reflection 0.25....Of coz you can feel free to edit it later. If you feel it is too bright.
  19. I have cached up the geometry, but when I render the foam(590Gb), about 40mil, the mantra: calculating geo/foam starts to eat up the ram, up to 100%(32GB). Anyone has any idea to optimize? Thanks
  20. rendering eating up ram

    But packprimitives only stores transform, would it be problematic when it comes to render motion blur which requires velocity?
  21. rendering eating up ram

    Thanks, I haven't really tried packed primitives or the delayed load, I will take a look into these later. Appreciated.
  22. Particle influencing pyro

    How to import the particle velocity so it can act to affect the pyro?
  23. Particle influencing pyro

    Hi Ryan, thanks I got it by the first method. Just by increasing the force scale.
  24. Particle influencing pyro

    Yes, thanks that was what I was looking for. But somehow both didn't work in my scene. Ryan's first method, I could have the guides showing that there is a velocity field going on with the particles, but it is just not affecting the pyro or it is affecting by very little. The 2nd method which was Grzegorz's method right? That one took a long time to calculate, and has no effect whatsoever.
  25. Particle influencing pyro

    anyone help?
×