Jump to content

Nerox

Members
  • Posts

    290
  • Joined

  • Last visited

  • Days Won

    7

Nerox last won the day on February 25 2014

Nerox had the most liked content!

About Nerox

  • Birthday 06/16/1985

Contact Methods

  • Website URL
    http://www.nicknimble.com
  • Skype
    nvanzutphen

Personal Information

  • Name
    Nick
  • Location
    @The Farm
  • Interests
    Houdini, Python, series, 3d animation, the farm, visual effects, mountain-biking , a good movie and music.

Recent Profile Visitors

6,989 profile views

Nerox's Achievements

Newbie

Newbie (1/14)

21

Reputation

  1. Hi Neon, Well I assume you're trying to make it look like if the train raises up from deep water but don't want to simulate it all. I think the easiest thing to do is to emit more water from the train parts that are above the water level. This is relatively easy to do, make you train a source and mask the part that you want to source from. You can also make the train a bit sticky (on the collision tab of the FLIP solver), which slows down fluid that is close to collision objects (train). How ever, if you're going to emit a lot of water you're going to run into trouble with your container that is filling up. There are methods of working out the total volume of your fluid end if its more than it inititially was you can drive the 'divergence' to compress the fluid and get rid of the surplus, but that is rather advanced stuff. Good luck :-)
  2. How are your New Year's Resolutions coming along? I think that in the upcoming year quit a few things will be come more obvious, clearer but just not quite yet ;-) '2014 The year where things become clearer'
  3. I think that is the point, because you would need have separate geo ROP's for that and when they are executed they'll recache the sim. Now I'm thinking of it, it might be possible that when you submit a sim using HQueue, the H13 Simulation submitter has a new 'frame by frame' option, which should allow (when connected in the right order) to do a sim and split out various outputs through multiple Geo ROP's without recalculating the sim every time.
  4. I have found similar stalling behavior with the FLIP solver in a setup where you supply a proxy collision field in a static object. It seems to stall when you feed it an empty SDF. In particular its the pressure solve that keeps on going for ever. This occurs in 13.0.237, the previous production build.
  5. As far as I know are VDB volumes not suported in DOP in general. The only exception that I have seen of VDB in DOP's (H13) is volume sampling it in a POP Vop.
  6. Hi, Great work! Suggestions I would like to make: Light up a ground plane by the fire Have the camera track along the horse, with a bit of handheld motion to it. Like if the camera man is on a quad or something. Advanced suggestion Try to make the horse out of flames as well. A starting point would be to turn the horse in a volume and have that emit fire with velocity from the horse his motion. Good luck :-)
  7. Using a procedural (noise) is the easiest using a vopsop on the geometry it self, or apply it in a shader. This is easy in a sense that you can avoid the texture map lookup, its quite advanced stuff. In order to make it work you have to manualy create a rest attribute on the particles that you can transfer onto the mesh to use as a P for the noise. I would use a sopsolver on the second input of the FLIP solver and create your rest attribute there. I advise to untoggle the reseeding since this makes things even more complicated.
  8. My next stept in creating CGI Food, Liquid Chocolate, who doesn't want that? :-)
  9. I love that finite solver :-) 8min a frame is pretty slow, but eh it looks awesome! Sounds like the cloth solver on steroids. What about 'Bending and Tearing FX'! We're going to have so much fun with that I find the idea of having particles interact with a flip sim very clever, because in this case you would need to have a huge flip container to sim all the drops that come down the stairs. But now I can imagine, that if all the water that comes down are particles, the FLIP container is practically flat to the ground, which must have a massive influence on the sim speed.
  10. Hi Freaq, Thanks for your reply, what do you mean with 'floaty'? Its plays in slow-motion, this might pull you off. It has been simulated at 'realtime' using all the real forces and then its timewarped afterwards.
  11. For those who have missed this course last January, this term (Oct 2013) the course HOU206 FX TD Fundamentals will rerun and I'll be available in the forums for questions. For more information check out the course listing
  12. I've made a second attempt at creating realistic Potato-Chips. The mayor innovation in this version is my layered noise .otl, that works in the SHOP and SOP context joined by a preview node for easy previewing and isolating individual layers for tweaking. It allows one to layer 3 different layers of noise, have float and vector outputs, apply various operations to combine the layers and offset the noise patterns by an id input. I've introduced the idea of a float id attribute, where the rounded number and the fraction can be handled differently. For instance, the whole number can be constant per object or piece, where the float value can be different on a per point basis with only minor differences, allowing for a wide range of artistic control. I'm using different sets of layered noise for channels like diffuse, spec, displacement and refraction. Part of the trick is to have some sets that share similar patterns. For instance the larger displacement pattern (single layer) is also used to drive the base layer of the diffuse. To get the snowy peaks idea, but then topped up with two different layers to make it look more natural. I never used the exact same patterns to drive different channels of the shader, people are really good at spotting patterns. Another feature of the potato-chips shader is that it takes a 'edge' attribute, basically an attribute that has a 0-1 value from the center to the edge. This allows to scale down displacements that are closer to the edge, add burned edges and that sort of stuff. I found that per piece random slight color offsets help to sell the shot as well. It makes it easier to believe they are real. Apart from that it was a tremendous amount of testing, I found it so difficult to get the right look. In the process I've created a ton of pringles, those are easy ;-). The pieces don't feature the 'popped bell' things, because A I couldn't get it right on a procedural basis (probably need to create hero pieces for that) B. This is aimed at clients in the television commercial business and I don't think Lays wants any of these things in its commercials. You can find a detailed article (though aimed on potential clients) on my website
  13. Sounds like a $1 sales opportunity on Orbolt ;-)
  14. Ghaha, yep it 'insiders only' jargon I suppose. Welcome to the 'club' ;-)
  15. What you can do is add an 'id' attribute, this is a toggle on the FLIP solver. Write out the particles as .bgeo in SOP's. read them back in, add a Timeblend SOP and then a time warp. Using the time warp your can animate scale ramp what ever you like and it looks very smooth. I wend from a 25fps to 2000fps (like how you would shoot it in real life) slowmotion and smooth as it can be. After the timewarp you can do your meshing.
×
×
  • Create New...