Jump to content


  • Content count

  • Joined

  • Last visited

Community Reputation

0 Neutral

About vtwinracersp2

  • Rank
  1. centos vs rh vs fedora

    i've been using fedora 17. it's a pain because of the frequent updates which make things break. simple things are just broken. i would just like to have a stable OS. i always feel like the rug is being pulled from underneath me. i was thinking about switching to centos or rh. centos is basically rh without the branding and support (with centos receiving updates a little later)? my main concern is the nvidia driver. currently centos uses an older version (290) of the nvidia driver? how is it updating the latest nvidia drivers and how well do they work? do they work as well with the older kernel that centos uses? how stable is houdini with centos?
  2. compiling sim_solverbullet errors

    hcustom -g -L /usr/lib64 -I /usr/include -lBulletDynamics -lBulletCollision -lBulletSoftBody -lLinearMath SIM_BuleltData.C hcustom -g -L /usr/lib64 -I /usr/include -lBulletDynamics -lBulletCollision -lBulletSoftBody -lLinearMath SIM_ConRelConeTwist.C hcustom -g -L /usr/lib64 -I /usr/include -lBulletDynamics -lBulletCollision -lBulletSoftBody -lLinearMath SIM_SolverBullet.C fedora 17 has bullet 2.81 installed. i do get a SIM_BulletData.o and then hcustom puts SIM_BulletData.so into /home/username/houdini12.1/dso/ there is some older version of bullet that is floating around (maybe on git) that was compiled against version 2.71 and that works(maybe because it defines the sim_bulletdata object inside of sim_solverbullet), but not the version of sim_solverbullet.c that comes with houdini in /opt/hfs12.1.125/houdini/public/SIM_SolverBullet is this the correct way to compile and link this?
  3. -i'm using fedora 17 with gcc 4.4.7 and houdini 12.1.125 -i compiled bullet 2.81 with debug options -i compiled SIM_SolverBullet.C that was in /opt/hfs12.1.125/houdini/public/SIM_SolverBullet i get an error when i try and launch houdini symbol lookup error: /home/username/houdini12.1/dso/SIM_SolverBullet.so: undefined symbol: _ZN14SIM_BulletData27getBulletDataDopDescriptionEv i was looking online for any clues. maybe the version of SIM_SolverBullet.C was compiled against a different version of bullet? i think there there might be an issue with IMPLEMENT_DATAFACTORY
  4. debugging: examining HDK objects

    seems like i have been using -g i think the issue is that the HDK is closed source. I am able to examine my own objects or c++ objects, but not any HDK objects.
  5. i'm debugging using gdb i just want to verify this. if i have a debug session going and there is any instanced object of a HDK object (say for example sim_object) i won't be able to see inside that object correct? all i get at the moment is just the address of the pointer, but i can't see what's inside. the only way is to invoke methods (for my example, sim_object, maybe i would run getIsBaked() ) and print them out. basically when i try to load what is at the address i get an error message "unable to load memory from the specified address"
  6. anyone have an example hip scene that uses the hdk example sim_solverhair.c? there is a note in the header file, but i don't think i'm setting this up right. Would EmptyObject be pointing to some wire object? /* * The simplest network to see something working with this... * * RBD Object EmptyObject * \ | * \ HairSolver * \ | * \__ Gravity * \ / * Merge * * The gravity node should be set to have a non-zero X velocity. * The HairSolver node should have the name of the rbd object * (default rbdobject1) as its Source Object. */