XRM

Hello everyone, I was looking at distributed whitewater in houdini. When you click the Particle Fluids -> Distributed Particle Fludis, it creates the necessary nodes needed for clustering whitewater sim. As far as houdini help file, that is what the way is. BUt when I am doing that, it is creating a subnet node inside the whitewater sourcing node named partition_emission. It is throwing an error. There is a node inside it named intersect_box which is a VDB activate node. It has a fit and detail expression. That expression is causing the error. You can see in the screenshots what I am talking about. Has anyone faced this issue ? Houdini - 17.5.425. OS : Ubuntu 19.10

Thank you Weish!

The issue is related to UV. Add uvproject to both meshes. Should fix it.

Do you have a solution ? I am getting the same error.

Houdini bullet uses packed primitives. I guess you have 'create packed geometry' checked on the assemble node. So your geometry now is represented by point which uses the name attribute to reference the original geometry. When you import these points inside a top using a RBD packed Object node, it creates proxy mesh for your geometry. This is actually the base of bullet. This makes things faster as you only have convex hull for your geometry. When you use an unpack node, it recreates the initial geometry back to your points referencing the name attribute. Cheers!

Oh. That sounds bad. Unfortunately I do not have houdini at home. So I won't be able to share a file. By the way, why don't you just use one frame for the static water?

Hi. Is there a chance you can upgrade your houdini? The latest houdini is quite stable for flip sims. I think that would help you. I tried with houdini 16.5 and 17. Recreated the same scenario as you. It works!

Maybe your source node has time limited by expression. Check your source node. If that is not the case share your file here. I can have a look. Ranadeep FX TD @dNeg

Hi everyone,I was trying to achieve a bullet passing through high viscous gel. I started with a keyframed bullet which passes through a flip cube. I am using the suction force option to make the flip particles retain shape. I can't get it right at all. Always particles are bursting out. I am attaching an youtube link to the reference. Ranadeep FD TD Dneg

Hi fellow FX artists. I was recently doing a sim which required viscous liquids. I wanted to try the opencl option with helps to calculate pressure and viscosity on a graphics card. I am getting this error c:\workarea\16.15.2211\stream\opencl\compiler\clc2\ocl-headers\build\wNow64a\B_rel/opencl12_builtins.h:14603:36: note: candidate function not viable: no known conversion from 'const __global fpreal *' (aka 'const __global float *') to 'const __local half *' for 2nd argument half __attribute__((overloadable)) vload(size_t offset, const __local half *p); ^ c:\workarea\16.15.2211\stream\opencl\compiler\clc2\ocl-headers\build\wNow64a\B_rel/opencl12_builtins.h:14609:36: note: candidate function not viable: no known conversion from 'const __global fpreal *' (aka 'const __global float *') to 'const half *' for 2nd argument half __attribute__((overloadable)) vload(size_t offset, const __private half *p); ^ error: Clang front-end compilation failed! Frontend phase failed compilation. Error: Compiling CL to IR clBuildProgram (-11) OpenCL Exception: C:\Users\Ranadeep\AppData\Local\Temp\\OCL9128T41.cl:286:27: error: no matching function for call to 'vload' fpreal dq_blend_weight = vload(i, dqLW); My version of Houdini is 15.0.244.16 This use to work fine on houdini 14. Graphics info is below OpenCL Platform AMD Accelerated Parallel Processing Platform Vendor Advanced Micro Devices, Inc. Platform Version OpenCL 2.0 AMD-APP (2004.6) OpenCL Device Hawaii OpenCL Type GPU Device Version OpenCL 2.0 AMD-APP (2004.6) Frequency 947 MHz Compute Units 40 Device Address Bits 64 Global Memory 4096 MB Max Allocation 2880 MB Global Cache 16 KB Max Constant Args 8 Max Constant Size 64 KB Local Mem Size 32 KB 2D Image Support 16384x16384 3D Image Support 2048x2048x2048 OpenCL Platform Extensions: cl_khr_icd cl_khr_d3d10_sharing cl_khr_d3d11_sharing cl_khr_dx9_media_sharing cl_amd_event_callback cl_amd_offline_devices OpenCL Device Extensions: cl_khr_fp64 cl_amd_fp64 cl_khr_global_int32_base_atomics cl_khr_global_int32_extended_atomics cl_khr_local_int32_base_atomics cl_khr_local_int32_extended_atomics cl_khr_int64_base_atomics cl_khr_int64_extended_atomics cl_khr_3d_image_writes cl_khr_byte_addressable_store cl_khr_gl_sharing cl_khr_gl_depth_images cl_amd_device_attribute_query cl_amd_vec3 cl_amd_printf cl_amd_media_ops cl_amd_media_ops2 cl_amd_popcnt cl_khr_d3d10_sharing cl_khr_d3d11_sharing cl_khr_dx9_media_sharing cl_khr_image2d_from_buffer cl_khr_spir cl_khr_subgroups cl_khr_gl_event cl_khr_depth_images cl_khr_mipmap_image cl_khr_mipmap_image_writes Please if someone has a workaround let me know. Thank you

Check the attached image. Turn that option on. It shows the collision sdf volume. I am sure the volume is not correct. Cheers!

Asus AMD R9 290 4GB opencl - 01:09:51 Intel i7 4930k - 01:37:53 Rest configuration 256Gb SSD Gigabyte x79 UP5 64 GB DDR3 2133MHz

you were right edward. My graphics drivers were set to autoupdate. Reverting back to the previous drivers fixed it.

The scene files are opening fine in the other systems. There is not problem with it I am sure of it. Even if I just add a box and then connect it to a vdb to polygons node, I can see this problem. I am on windows 7, Houdini 13.0.547, AMD R290 with latest drivers. It was working fine. I can only think of formatting and installing everything fresh again.

Nothing. It just started happening all of a sudden. I even tried restoring and also deleted the houdini13 preferences folder from my documents folder. I even re installed the graphics drivers. All in vain.