polygonfuture Posted September 23, 2015 Share Posted September 23, 2015 Hi all, I'm an extreme newbie with houdini so please forgive my incredible ignorance. I am attempting to run a sim based off of the tutorial by Ben Watts: In my case, I am wanting to melt a geometry object from a heat source located at the ground plane. Similar to an object (like butter for instance) that has been placed on a super hot hot skillet. Right now the big problem is that the basic vex script that affects the flip fluid particles with gravity is very simple and only affects particles if their "temp" attribute is greater than 1. This gives the effect of "hot" particles melting and flowing down and outwards onto the ground plane; however, the non-heated particles outside of the range of heat, are suspended in midair. Obviously this does not give a proper effect of an object that is overall affected by geometry. Does anyone have any pointers on how I can get a more realistic result of an object being melted and affected by gravity properly? I've attached my hip file (sans geo). melting_sim_v4.hipnc Quote Link to comment Share on other sites More sharing options...
NSugleris Posted September 23, 2015 Share Posted September 23, 2015 I'd try to get more values than 1 and 0, and usually I've gotten away with linking my new attrib "your case temp" to viscosity and then I just link those in my SHOPS context to get the heating look. Also H15' has a new tool to do exactly what you want. ( albeit making this without a shelf tool is always good ) Quote Link to comment Share on other sites More sharing options...
polygonfuture Posted September 23, 2015 Author Share Posted September 23, 2015 Thanks Nico. Right after I posted this I discovered the cause of the weird simulation. I had a typo between the viscosity attribute name in the sopsolver and and the flipsolver. Once I fixed that the entire point volume began to behave like I expected. Right now though the issue is the simulation takes a long time (like overnight). Can anyone provide some tips as to ways I can optimize the sim (and the meshing of the final geo)? I uploaded a new version of the .hip file. The meshing takes place in the sim_object_meshing node. melting_sim_v6.hipnc Quote Link to comment Share on other sites More sharing options...
Guest tar Posted September 23, 2015 Share Posted September 23, 2015 (edited) Right now though the issue is the simulation takes a long time (like overnight). What's is your computer specs? EDIT: your hand geo is missing - can you lock that node so the geo is included in your .hip file. Thanks! Edited September 23, 2015 by tar Quote Link to comment Share on other sites More sharing options...
polygonfuture Posted September 23, 2015 Author Share Posted September 23, 2015 (edited) Lets try this again. Hope this works. The laptop is old. It's a 2011 17" MBP with 2.2ghz i7 and 16 gigs of ram. Updated it with the locked geo node. melting_sim_v6.hipnc Edited September 23, 2015 by polygonfuture Quote Link to comment Share on other sites More sharing options...
Guest tar Posted September 23, 2015 Share Posted September 23, 2015 Your CPU is rated at 6185, http://www.cpubenchmark.net/cpu.php?cpu=Intel+Core+i7-2720QM+%40+2.20GHz, This is very low by todays standards - when you run the performance monitor in Houdini you can see a very well threaded 'VDB from Particles' is using up all the processing time. Taking 'overnight' can imply ~6 hours, so a faster CPU will cut that time down dramatically - i.e. a dual X5680 gives ~18,000 CPU power, -> 6 hours should then take ~ 2 hours. To speed it up you could drop the number of particles to surface. Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.