Jump to content

Leaderboard

Popular Content

Showing content with the highest reputation since 05/26/2025 in Posts

  1. Hi @EdwinFumano, you could calculate a bunch of eikonal distances using attribute fill inside a for loop and remap them as animation: float speed = chf('speed'); float t = @Time * speed; int i = floor(t); float u = frac(t); int n = npointsgroup(0, 'start'); string att = 'dist_' + itoa(i%n); float dist = point(0, att, i@ptnum); float mask = chramp('shape', dist - u + 1.0); float fade = 1.0 - smooth(0.5, 1.0, u); f@mask = mask * fade; setdetailattrib(0, 'active_dist', att, 'set'); infection.hip
    4 points
  2. Here are some ideas using this on curves (if I understood correctly) shift_points_along_curve.hipnc
    3 points
  3. Studio ??? Talk to your manager right away and give these guys a job @hannes603 @konstantin magnus if they need a job.. Find Qiita website . !Advent Houdini 2020! you have Files for Infect and very Nice tricks for making those effects and much more using same Logic . Then(TIP) When you have Full Code that Works in Houdini you can Make anything in chatgpt(reverse and investigete) Now days. Have a nice Day .
    2 points
  4. I think I have found the perfect solution. This seems 100% correct. Takes into account the camera motion as well. screen_space_velocity_2.hiplc
    2 points
  5. This could be due to the internal order of the point group. try shuffling starting points by randomly picking a point number out of the array: int seed = chi('seed'); int pts[] = expandpointgroup(0, 'start'); int iter = detail(1, 'iteration', 0); int pick = pts[int(rand(iter, seed) * len(pts))]; i@group_boundary = i@ptnum == pick; infection_multiple_01.hip
    2 points
  6. This is what I could come up with. Uses 1 qLib node (Camera Plane ql). screen_space_velocity.hiplc
    2 points
  7. You have a vex function called toNDC where you can use your velocity information to transform it to normalized raster coordinates. Even if it says it's only well defined in shading context it seems to work in sop as well.
    2 points
  8. https://forums.odforce.net/topic/52360-heightfields-guided-by-pyro/#comment-238377 you have endless Way to make this .
    2 points
  9. Yeah, but only works on straigth line @Aizatulin's file is awesome!
    2 points
  10. I do not think a discord is the right thing for odforce, as Marc mentioned, the info get lost easily. Even though you can search etc still is not as nice as having it in a forum.
    2 points
  11. Are you interested in creating custom velocity fields and custom forces for your Houdini FLIP, Pyro and Smoke Simulations?? Are you interested in building custom DOP networks and custom solvers? This course on Udemy is for you! Enroll it now from HERE! https://www.udemy.com/course/mastering-houdini-simulation-custom-forces-velocity-fields/?referralCode=374420356510272B5A71 Course Overview: Take your Houdini simulations to the next level by mastering the art of custom velocity fields and forces. This advanced course is designed for artists and technical directors looking to gain full control over their simulations. Through real-world examples, we will explore a variety of techniques to create stunning motion effects using velocity volumes, custom forces, volume VOPs, and point clouds. By the end of this course, you will have the ability to craft complex, Houdini dynamic simulations that behave exactly as you envision. Whether you're working with particles, smoke, FLIP water simulations or other volumetric effects, these techniques will help you stand out in the world of VFX and motion design. What You’ll Learn: How to create custom velocity fields for unique water, particle and smoke effects Techniques for designing an ocean vortex with beautiful banding effect Crafting custom curve forces to guide particle and smoke simulations Utilizing complex geometry to generate unique force interactions Creating a particles-to-word effect using object forces Developing axis-related forces for controlled simulations Applying velocity volumes in various ways, including: POP Advect by Volume Wind forces Directional drag and full drag Advanced techniques with Point Clouds and Volume VOPs for precision control Create a curve force and axis force digital assets that can be used to control smoke and particle simulations Course Breakdown: Module 1: Ocean Vortex Effect Designing a realistic vortex simulation Using velocity banding for artistic control Fine-tuning the look of fluid motion Module 2: Curve-Driven Forces for Smoke & Particles Creating custom curve forces Directing smoke and particles along a path Controlling force strength and falloff Leveraging Point Clouds for volume manipulation Customizing Volume VOP networks for precise effects Debugging and optimizing simulations Module 3: Forces & Complex Geometry Using geometry to influence simulations Crafting intricate force interactions Case study: Real-world examples of geometric forces Module 4: Particles to Word Effect Transforming particles into text shapes Managing object forces for controlled transitions Enhancing readability and motion flow Module 5: Axis-Based Forces for Dynamic Simulations Applying forces relative to an axis Rotational and directional force applications Controlling turbulence and flow behavior Understanding POP Advect by Volume Blending velocity for natural movement Wind, directional drag, and full drag techniques Who This Course is For: Intermediate to advanced Houdini users VFX artists looking to master simulation control Motion designers who want more precision in particle and smoke effects Technical artists eager to dive into advanced Houdini workflows Why Take This Course? Learn industry-standard techniques used in professional VFX studios Gain hands-on experience with practical, project-based lessons Unlock the ability to create your own unique simulations Elevate your skills and portfolio with high-end effects using Houdini Course Requirements: Basic knowledge of Houdini A computer capable of running Houdini simulations Passion for creating stunning, controlled simulations Are you ready to take full control of your simulations? Enroll now and start mastering custom velocity fields and forces today!
    1 point
  12. A couple things I see at first glance: 1. Make sure to subdivide the tubes, noodles. Add more detail in the bends so the curves are smoother. 2. Make them thinner. 3. Perhaps more subsurface scattering? Nice work, though. This is a great start.
    1 point
  13. I'm working on a personal project starring a pasta monster, I had to create a bolognese sauce for it. I tought it could be fun for others (at least to look around in COPs). Have fun with it! bolognese_cops.hiplc I followed this article and tried to replicate what he did in Substance Designer, but using COPs. Copernicus is indeed amazing!! https://80.lv/articles/making-fusilli-pasta-material-with-substance-3d-designer-houdini The ingredients:
    1 point
  14. How would you calculate the velocity of a moving geometry (consistent topology, static camera) in screen space?
    1 point
  15. collision_passes_100.mp4 collision_passes_10.mp4 col_passes_vellum.hip col passes 100 vs 10 jfyi
    1 point
  16. Thanks everyone — this thread helped clarify a lot. Just to add for anyone else running into this: yes, even for static objects, if you’re using a VDB Proxy Volume, Houdini still seems to evaluate the proxy each frame unless you explicitly manage when and how it’s updated. Disabling the Collision Guide (as Ryan pointed out) makes a huge difference, especially in viewport playback — it's not just cosmetic, it actually controls whether Houdini recalculates that VDB every frame. Also, diving into the Static Object DOP and checking out the internals is super useful. You can often lock in your VDB or pre-cache it manually if you're hitting perf issues. Highly recommend enabling the Performance Monitor too — it really does help track where the time is going if you're not sure what’s dragging down the sim. TL;DR: It's not a bug, just a gotcha. VDB-based proxy collisions are powerful but require a little setup awareness, even for static colliders.
    1 point
  17. Put the relevant parts of the code inside a loop to offset the time and increment the mask number with each iteration. float speed = chf('speed'); int num = chi('infections'); float t = @Time * speed; int n = npointsgroup(0, 'start'); float mask_collect = 0.0; for(int k = 0; k < num; k++){ t += 1.0 / float(num); int i = floor(t); float u = frac(t); string att = 'dist_' + itoa(i+k%n); float dist = point(0, att, i@ptnum); float mask = chramp('shape', dist - u + 1.0); float fade = 1.0 - smooth(0.5, 1.0, u); mask_collect = max(mask_collect, mask * fade); } f@mask = mask_collect; infection_multiple.hip
    1 point
  18. Hey friends, Had a fellow 3d person stop by my youtube channel where I mostly sculpt random Zbrush stuff and ask for this model to be donated to the Houdini community: Which I think is a brilliant idea as I dabble with Houdini myself and love it. So here you go, have fun, would love to see some awesome things created with my humble sculpt Use it for whatever, personal, commercial. A link back to me/my channel always massively appreciated, but no obligations. Cheers! EoL PS hope this is the right section of forum to post this, apologies if not. xportPigHead_v01.fbx xportPigHead_v01.obj
    1 point
  19. Cache out the initial sim then use this cache as an animated collision object in ANOTHER sim? --> yes, this way you can control the primary animated geo more easy
    1 point
  20. @sipi maybe //box ndc vector offset = {0.5, 0.5, 0}; vector cameraScale = set(ch("camera_X"), ch("camera_Y"), 1); if (@group_near) @P = set(@P.x, @P.y, ch("near")*-1); if (!@group_near) @P = set(@P.x, @P.y, ch("far")*-1); vector xformpos = @P * maketransform(0,0, offset,{0,0,0},cameraScale,{0,0,0}); @P = fromNDC(chs("camera_path"), xformpos);
    1 point
  21. Hi Jim, you could integrate text from Copernicus which is also a 2D voxel field. heightfield_text.hip
    1 point
  22. I would like to increase the curviness of a given curve in post. Do you know other methods to this problem? This seems to work nicely, but maybe there are better ways to do it.(?) increase_curvature_in_post.hiplc
    1 point
  23. Thank you for your explanation, that's great. I realized you changed your setup in the last file.
    1 point
  24. interesting one, though it doesn't seems to be 100% correct for example here the normals don't follow the curvature properly adjusting the Search Radius can help, but it has to be larger than the length of a segment in the line (this can be problematic if the distribution of points are not uniform in the line (and if you can't apply resample for some reason)) but thanks, it's an interesting and simple approach indeed.
    1 point
  25. if you mean Those Swirls .its just Texture that its being advected and Rest uv by movement of particles(to volumes) that you can map to -P.y . just volumes . @korbendalas
    1 point
  26. It is cool! I will explore how you solve it, your setup is completly different than mine, your setup is totally different from mine. Could you explain a bit about your workflow?
    1 point
  27. Hi, Here's a hipfile for this: drive_ocean_amplitude.hiplc
    1 point
  28. Maybe animate the geo or volume on subframes using simple modulo expression? frozen_motionblur.hiplc (hate to put pighead in such a vulnerable situation)
    1 point
  29. This is pretty much the same as it always has been: Who's Online 10 Members, 0 Anonymous, 89 Guests However this is a good question, what can we do to generate more excitement on the forums? Is anyone interested in running challenges? Unfortunately I have a studio to run now, which is quite all encompassing, but we're happy to support any good ideas! M
    1 point
  30. Hi everyone, this is my low level curve modeling tool where you can draw, sculpt, edit and refine in bunch of ways. If you want to report bug or suggest a feature you can connect to discord from gumroad. I'd be happy to hear what you think and if you find it useful. https://9099688963566.gumroad.com/l/dibiho
    1 point
  31. To fine-tune IK in APEX, lock specific axes by adjusting the constraints and setting limits on joint rotations. Use minimum and maximum angle settings for range control. Adjust axis weights to prioritize certain movements. Test the rig’s motion and refine the limits until it behaves as needed. Check APEX documentation for specific instruction
    1 point
  32. So I fixed it. The issue was quite simple. There was no collision shape for this specific bone. I just had to click on it to create a coll shape, I didnt even resized or moved it. All those shapes doesnt look very optimized though, but it works well for what I need. Ragdoll_0102.hip
    1 point
  33. nice. i put everything in a for loop , works for me . Blur_smooth_noise_in_matnet_material_foreach.hip
    1 point
  34. Here is a basic example: smooth_shader_by_derivatives.hiplc
    1 point
  35. Hi art3mis, We used all those options before and it depends on what you are exporting. Alembic for vellum cloth is not great for optimisation later on and VAT was constantly loaded in memory which gave us problems during optimisation as well. We ended up using dembones for our baked cloth simulation and in the end was a lot lighter on the system. RBD to FBX was used extensively through out our production. That was great since we can optimise the the simulation right before it was baked and it doesn't change the rbd sim at all. Another issue with Alembic is that it doesn't handle a 2nd UV channel. Glen
    1 point
  36. Houdini was getting bored doing hours of simulation, so I gave him something to play with. - Done in Houdini, script in VEX, rendered using Karma.
    1 point
  37. Hi @virgile, you could blend positions based on their distance towards the surface: In a point wrangle with first input mesh, second input sphere: float d_min = chf('min_distance'); float d_max = chf('max_distance'); int prim; vector uvw; float d = xyzdist(1, v@P, prim, uvw); vector pos = primuv(1, 'P', prim, uvw); float mask = smooth(d_min, d_max, d); v@P = lerp(pos, v@P, mask); conform_to_surface.hipnc
    1 point
  38. attached is an example of how it could be done. Keep in mind this is intended to be used with lower contrast images, for bump mapping etc and not a fully blown high contrast brick image. but it gets the point across I think. restratio_flipblend.hiplc
    1 point
  39. and I made it work inheritVel_Test_v02.hip
    1 point
  40. During the last 3 weeks, a did some Rnd and published my results on vimeo . Some people asked me to share my files here, so here we are i hope it will help!
    1 point
  41. Also found a (rather low quality) recording of the talk I am refering to (roughly from minute 02:10 on) In this talk he states that for the connection each pore looks for the nearest point in the direction of the two nearest curve and then one beyond. So my approach right now (testing this takes time as I still struggle with vex syntax) is: get nearest point using the "nearestpoint" function, build vector from current point to each of the nearpoints. Then build direction vector from the nearest two curve points and do a dot product to check if the point matches the directionvector best, and connect these points with a curve once that is done. All in a point wrangle. After that I resample the curve, project to the surface using minpos and subtract as a VDB. Does that make sense or is there a better way?
    1 point
  42. If you add a Cluster SOP after your Cluster Points SOP, it has a checkbox to output Cluster Centers. -b
    1 point
  43. Also make sure to look at the cluster points node which even offers an "output: average points" function.
    1 point
  44. Hi, You can just average out all the points found in the point cloud lookup where you are coloring them, and either create new points at the computed center or store it as an attribute.
    1 point
  45. Hey Masoud /; check hip file it's that what you want ? it's not clean way but maybe work ... / I think your start Method can be better for distributions Points Cluster .. CenterOfEachColoni_01.hip
    1 point
  46. couple of things, I focused only at SDf force as that's the way to go: - you are using Isooffset to generate SDF, which is slow and may introduce artifacts, consider using VDB From Polygons - your volumegradient is not sampling the SDF, but the default cube geo, you have to point it to your geo - your volumegradient is sampling using v, it should sample at P position - you are adding P to result of your gradient, while the gradient direction is already parallel to the tube normal, so you shouldn't mess with it more, just negating and normalizing is fine - VOP Force is old and may be slow, in your case when using RBDs consider using POP Force Instead here is the file with all the changes also has both VOP Force and POP Force approach Custom_Gravity_fix.hip
    1 point
  47. Carve SOP. Fist U : 0 Second U : 1 Breakpoints : Cut At all Internal U Breakpoints
    1 point
  48. i wired a vop shop. it should be what you want ramp1.hipnc
    1 point
×
×
  • Create New...