Welcome to od|forum

Register now to gain access to all of our features. Once registered and logged in, you will be able to contribute to this site by submitting your own content or replying to existing content. You'll be able to customize your profile, receive reputation points as a reward for submitting content, while also communicating with other members via your own private inbox, plus much more! This message will be removed once you have signed in.

Search the Community: Showing results for tags 'opencl'.

More search options

  • Search By Tags

    Type tags separated by commas.
  • Search By Author

Content Type


  • General
    • Lounge/General chat
    • Education
    • Jobs
  • Houdini
    • General Houdini Questions
    • Effects
    • Modeling
    • Animation & Rigging
    • Lighting & Rendering
    • Compositing
    • Games
  • Coders Corner
    • HDK : Houdini Development Kit
    • Scripting
    • Shaders
  • Art and Challenges
    • Finished Work
    • Work in Progress
    • VFX Challenge
    • Effects Challenge Archive
  • Systems and Other Applications
    • Other 3d Packages
    • Operating Systems
    • Hardware
    • Pipeline
  • od|force
    • Feedback, Suggestions, Bugs

Found 15 results

  1. This operator allows you to run an OpenCL kernel as part of your SOP network. Depending on the GPU, some operators can be orders of magnitude faster than even VEX. In this case the OpenCL code is 144 times faster than VEX on GTX 970.
  2. Hi all, is it possible or a way to enable all my GPUs for OpenCL? i have 3x titan-x and a quadro m4000 total of 4 cards. thanks
  3. Hi, Please let me know if there is a simple setting in preference where we can enable or switch to GPU or OpenCL. I am not sure whether viewport uses GPU, I think it does and I would also like to know if we can directly switch a DOP node to use GPU for its calculations. Can Mantra and Arnold use GPU? Please let me know the specific settings. Thanks again.
  4. Hi My understanding is that VEX runs entirely on the CPU. But with the right hardware (ie latest Pascal based GPU) the same functionality in OpenCl could be magnitudes faster. Is anyone aware of any OTL or other project which have converted VEX code to OpenCL?
  5. Hi If you are enabling OpenCL does installing the latest CUDA 8.0 Toolkit offer any advantage over simply having the latest Nvidia drivers installed?
  6. Volume convolution on the GPU using OpenCL. For 27M voxels using 100 iterations, OpenCL is 650 times faster than C++ and 12525 times faster than VEX.
  7. Smoothing geometry on the GPU using OpenCL. In VEX it has to be done in 2 steps where the same thing can be achieved in OpenCL in one step using barriers [khronos.org]. For a 822K polygon model, it's over 500 times faster than VEX on GTX 970 for 1000 iterations, and over 600 times faster for 10K iterations. Timings are recorded separately as there seems to be some overhead when running them one after another. Special thanks to SESI.
  8. Hi, Looked all over the forums and couldnt find a solution, so opened a new thread. I am using Pyro in Houdini 14 and when I set "Use OpenCL", the Houdini Console pops up with what I believe to be an error. (see attached image) Do I need to install a driver or something for my graphics card? Am currently using a Quadro M4000. Hope someone can help out. Thx in advance. Here are other specs to the computer I am using: PC: Dell Precision T7810 OS: Windows 7 SP 1 64bits CPU: Intel Xeon ES-2630 v3 @ 2.40GHz RAM: 64 GB GPU: Quadro M4000
  9. I have to GPUs I like to utilize in Houdini, one for opengl drawing and the other solely for OpenCL computation. Using following variables: HOUDINI_OCL_DEVICETYPE=GPU HOUDINI_OCL_DEVICENUMBER=0 But I am unsure if its really separated or if the drawing card also does the opencl computation. Is there any way to figure out which device is used for what task? tried to monitor with the gpu-z but it is giving some inconsistent results Thanks Sebastian
  10. Hey using HOUDINI_OCL_DEVICENUMBER=1 in Houdini build wont work. There are no problems with the 14 builds. I really would like rule out any hardware/driver issues on my side and need a repro from you guys. Ive already passed this on to the support during the open beta. Thanks
  11. Hey, at the brink of having Octane in Houdini, and a (kinda) Nvidia vram support for 64bit opencl, I am thinking of upgrading my GPUs to two Titan X cards. I am aware that Sli wont do much with OpenCl, but alot for gpu rendering. Octane Benchmarks are batshit crazy with such a rig. What I am concerned about are some reports of the double precision capabilities with Maxwell chips, as these wont match those of the first Titan card which had aimed at the computation market. How would this affect the OpenCl computation within Houdini? I believe that the Maxwell Quadros would have the same issue.
  12. Hello! I haven't seen a resource that benchmarks the calculation speed of graphics cards in Houdini so I thought I'd post this. I hope that people can participate and post their scores using their gaming/workstation graphics cards so other people can get a rough idea of the performance of each graphics card in Houdini before they decide to buy/upgrade anything. This should only take a few minutes - 1. Start Houdini 2. Use the following units under your Hip File Options - Unit Length - 1 Unit Mass - 1 3. Frame Range 1-240 @ 24fps 4. Add a sphere at origin 5. Goto the Pyro FX shelf tool and select Explosion. Select the sphere. 6. Disable Cache Simulation in your DOP and use the following settings - Pyro Container Settings Division size - 0.075 Size - 7.5, 10, 7.5 Center - 0, 4, 0 Resize Container Settings Padding 0.3 Subtract Threshold 0.2 9. Select the Pyrosolver. In the Advanced tab, turn on OpenCL 10. Save the file and restart your computer 11. Use a stopwatch or http://www.online-stopwatch.com/ to keep track of the time 12. Open the file, start the sim and the stopwatch timer 13. Record the time after 240 frames. 14. Run the same sim again but with OpenCL disabled 15. Post the exact times with your PC Specs, Houdini version and OS in the following format (using my results as an example) - Houdini Version - OS - Windows 7 Ultimate x64 GFX - Zotac 770GTX Amp! Edition (2GB) / Driver - 347.09 Time - 1:37.222 CPU - Intel i7 3770K @ stock Time - 2:09.503 Other - ASRock z77 Pro3 Motherboard, 32 GB Corsair Vengeance RAM Thank you! PS - I've attached a hip file with the above settings using Houdini 14 that you can run right away. Thanks for the suggestion, Skybar. OpenCL_TestFile.hipnc
  13. http://www.cgchannel.com/2014/09/video-amds-new-gpu-based-raytrace-renderer-for-maya/ And 3ds Max integration: http://fireuser.com/blog/video_demo_of_firerender_interactive_photorealistic_rendering_plugin/ Does SESI consider involving in this? Looks pretty cool according to OpenCL part and I suppose AMD is quite big and stable as the developer (as far as I know only Octane 3 supports OpenCL path). What do you think?
  14. There is no better way how to start the New year than smoke your GPU BulletSOP 2.5 preview features implementation of WIP Bullet Library 3.0 with OpenCL(CPU/GPU) support. Please keep in mind that this version supports only fraction of BulletSOP features and It's only for testing and highly unstable! For OpenCL settings, check btSolver node -> Advanced tab! You can activate "old" CPU solver(which has some bugs unlike BulletSOP 2.0.10) and compare result and speed. Testing scene includes 80K boxes(of course you can try to go higher - change settings on OpenCL tab). And If you want to publish some result(s) it's better to restart Houdini(clean memory) before running final simulation. So Let's try it and let us know about your results! Best wishes, Milan Suk BulletSOP binary is free of charge, but You can make a donation and help keep the project running. It's up to you! Download BulletSOP v2.5 preview for Windows(compile on Houdini 13.0.260): bit.do/bulletsop_2_5_preview PS: This is my result on 3 years old GPU!
  15. Hi. I want to know something about Houdini Rop Cache and Computing. Im simulating 240 Frames of dust with smokesolver emitted from particles. Pyro Divisions are 0.06 - Volume source scale divisions: 0.1 Im caching and from Frame 0 to Frame 100 it was only 4 hours.. .bgeos weight from 18 Mb at early start to 200 Mb each bgeo file. So now the caching and is taking ages.. It took 17 hours to cache from Frame 100 to Frame 133. Bgeos are getting as big as 450mb each. How long it will take to cache until Frame 240? Is It normal that a bgeo weight 400mb and take 1.5 hours to compute each frame? Im using Open CL - Quadro 4000