blender/intern/elbeem/extern/elbeem.h

274 lines
7.9 KiB
C
Raw Normal View History

2011-02-25 10:51:01 +00:00
/** \file elbeem/extern/elbeem.h
* \ingroup elbeem
*/
2006-02-27 11:56:04 +00:00
/******************************************************************************
*
* El'Beem - Free Surface Fluid Simulation with the Lattice Boltzmann Method
* All code distributed as part of El'Beem is covered by the version 2 of the
* GNU General Public License. See the file COPYING for details.
- bugfixes #4742 exported normals are now correct #4821 & 4956 for complex movements in/outflows can now also use the animated mesh option - new features * isosurface subdivision: directly creates a finer surface mesh from the simulation data. this increases simulation time and harddisk usage, though, so be careful - usually values of 2-4 should be enough. * fluidsim particles: extended model for particle simulation and generation. When isosurface subdivision is enabled, the particles are now included in the surface generation, giving a better impression of a single connected surface. Note - the particles are only included in the final surface mesh, so the preview surface shows none of the particle effects. * particle loading: different types of particles can now be selected for display: drops, floats and tracers. This is a bit obsolete due to the extensions mentioned above, but might still be useful. Floats are just particles floating on the fluid surface, could be used for e.g. foam. * moving objects impact factor: this is another tweaking option, as the handling of moving objects is still not conserving mass. setting this to zero simply deletes the fluid, 1 is the default, while larger values cause a stronger impact. For tweaking the simulation: if fluid disappears, try increasing this value, and if too much is appearing reduce it. You can even use negative values for some strange results :) - more code cleanup, e.g. removed config file writing in fluidsim.c, added additional safety checks for particles & fluidsim domains (these currently dont work together). I also removed the "build particles" debug message in effects.c (seemed to be unnecessary?). Some more info on the new features: Here are two test animations showing the difference between using the particle generation with isosurface subdivision. This is how it would look with the old solver version: http://www10.informatik.uni-erlangen.de/~sinithue/blender/fluid6_fl6manc4_1noparts.mpg and this with the new one: http://www10.informatik.uni-erlangen.de/~sinithue/blender/fluid6_fl6manc4_2wparts.mpg Both simulations use a resolution of 64, however, the version with particles takes significantly longer (almost twice as long). The .blend file for a similar setup can be found here: http://www10.informatik.uni-erlangen.de/~sinithue/blender/fluid6_testmanc4.blend (Minor Tips for this file: dont enable subdivions of characters until rendering, thus leave off for simulation, as it uses the rendering settings! For making nice pictures switch on subdivion, and OSA.) And here's a picture of old vs. new (for webpage or so): http://www10.informatik.uni-erlangen.de/~sinithue/blender/fluid6_manc4compare.png
2006-11-05 16:30:29 +00:00
* Copyright 2003-2006 Nils Thuerey
2006-02-27 11:56:04 +00:00
*
* API header
*/
#ifndef ELBEEM_API_H
#define ELBEEM_API_H
// simulation run callback function type (elbeemSimulationSettings->runsimCallback)
// best use with FLUIDSIM_CBxxx defines below.
// >parameters
// return values: 0=continue, 1=stop, 2=abort
// data pointer: user data pointer from elbeemSimulationSettings->runsimUserData
// status integer: 1=running simulation, 2=new frame saved
// frame integer: if status is 1, contains current frame number
typedef int (*elbeemRunSimulationCallback)(void *data, int status, int frame);
#define FLUIDSIM_CBRET_CONTINUE 0
#define FLUIDSIM_CBRET_STOP 1
#define FLUIDSIM_CBRET_ABORT 2
#define FLUIDSIM_CBSTATUS_STEP 1
#define FLUIDSIM_CBSTATUS_NEWFRAME 2
2006-02-27 11:56:04 +00:00
// global settings for the simulation
typedef struct elbeemSimulationSettings {
/* version number */
short version;
/* id number of simulation domain, needed if more than a
* single domain should be simulated */
short domainId; // unused within blender
2006-02-27 11:56:04 +00:00
/* geometrical extent */
float geoStart[3], geoSize[3];
/* resolutions */
short resolutionxyz;
short previewresxyz;
/* size of the domain in real units (meters along largest resolution x,y,z extent) */
float realsize;
/* fluid properties */
double viscosity;
/* gravity strength */
float gravity[3];
/* anim start end time */
float animStart, aniFrameTime;
/* no. of frames to simulate & output */
short noOfFrames;
/* g star param (LBM compressibility) */
float gstar;
/* activate refinement? */
short maxRefine;
/* probability for surface particle generation (0.0=off) */
2006-02-27 11:56:04 +00:00
float generateParticles;
/* amount of tracer particles to generate (0=off) */
int numTracerParticles;
2006-02-27 11:56:04 +00:00
/* store output path, and file prefix for baked fluid surface */
char outputPath[160+80];
/* channel for frame time, visc & gravity animations */
int channelSizeFrameTime;
float *channelFrameTime;
int channelSizeViscosity;
float *channelViscosity;
int channelSizeGravity;
float *channelGravity; // vector
/* boundary types and settings for domain walls */
- bugfixes #4742 exported normals are now correct #4821 & 4956 for complex movements in/outflows can now also use the animated mesh option - new features * isosurface subdivision: directly creates a finer surface mesh from the simulation data. this increases simulation time and harddisk usage, though, so be careful - usually values of 2-4 should be enough. * fluidsim particles: extended model for particle simulation and generation. When isosurface subdivision is enabled, the particles are now included in the surface generation, giving a better impression of a single connected surface. Note - the particles are only included in the final surface mesh, so the preview surface shows none of the particle effects. * particle loading: different types of particles can now be selected for display: drops, floats and tracers. This is a bit obsolete due to the extensions mentioned above, but might still be useful. Floats are just particles floating on the fluid surface, could be used for e.g. foam. * moving objects impact factor: this is another tweaking option, as the handling of moving objects is still not conserving mass. setting this to zero simply deletes the fluid, 1 is the default, while larger values cause a stronger impact. For tweaking the simulation: if fluid disappears, try increasing this value, and if too much is appearing reduce it. You can even use negative values for some strange results :) - more code cleanup, e.g. removed config file writing in fluidsim.c, added additional safety checks for particles & fluidsim domains (these currently dont work together). I also removed the "build particles" debug message in effects.c (seemed to be unnecessary?). Some more info on the new features: Here are two test animations showing the difference between using the particle generation with isosurface subdivision. This is how it would look with the old solver version: http://www10.informatik.uni-erlangen.de/~sinithue/blender/fluid6_fl6manc4_1noparts.mpg and this with the new one: http://www10.informatik.uni-erlangen.de/~sinithue/blender/fluid6_fl6manc4_2wparts.mpg Both simulations use a resolution of 64, however, the version with particles takes significantly longer (almost twice as long). The .blend file for a similar setup can be found here: http://www10.informatik.uni-erlangen.de/~sinithue/blender/fluid6_testmanc4.blend (Minor Tips for this file: dont enable subdivions of characters until rendering, thus leave off for simulation, as it uses the rendering settings! For making nice pictures switch on subdivion, and OSA.) And here's a picture of old vs. new (for webpage or so): http://www10.informatik.uni-erlangen.de/~sinithue/blender/fluid6_manc4compare.png
2006-11-05 16:30:29 +00:00
short domainobsType;
float domainobsPartslip;
/* what surfaces to generate */
int mFsSurfGenSetting;
2006-02-27 11:56:04 +00:00
/* generate speed vectors for vertices (e.g. for image based motion blur)*/
short generateVertexVectors;
/* strength of surface smoothing */
float surfaceSmoothing;
- bugfixes #4742 exported normals are now correct #4821 & 4956 for complex movements in/outflows can now also use the animated mesh option - new features * isosurface subdivision: directly creates a finer surface mesh from the simulation data. this increases simulation time and harddisk usage, though, so be careful - usually values of 2-4 should be enough. * fluidsim particles: extended model for particle simulation and generation. When isosurface subdivision is enabled, the particles are now included in the surface generation, giving a better impression of a single connected surface. Note - the particles are only included in the final surface mesh, so the preview surface shows none of the particle effects. * particle loading: different types of particles can now be selected for display: drops, floats and tracers. This is a bit obsolete due to the extensions mentioned above, but might still be useful. Floats are just particles floating on the fluid surface, could be used for e.g. foam. * moving objects impact factor: this is another tweaking option, as the handling of moving objects is still not conserving mass. setting this to zero simply deletes the fluid, 1 is the default, while larger values cause a stronger impact. For tweaking the simulation: if fluid disappears, try increasing this value, and if too much is appearing reduce it. You can even use negative values for some strange results :) - more code cleanup, e.g. removed config file writing in fluidsim.c, added additional safety checks for particles & fluidsim domains (these currently dont work together). I also removed the "build particles" debug message in effects.c (seemed to be unnecessary?). Some more info on the new features: Here are two test animations showing the difference between using the particle generation with isosurface subdivision. This is how it would look with the old solver version: http://www10.informatik.uni-erlangen.de/~sinithue/blender/fluid6_fl6manc4_1noparts.mpg and this with the new one: http://www10.informatik.uni-erlangen.de/~sinithue/blender/fluid6_fl6manc4_2wparts.mpg Both simulations use a resolution of 64, however, the version with particles takes significantly longer (almost twice as long). The .blend file for a similar setup can be found here: http://www10.informatik.uni-erlangen.de/~sinithue/blender/fluid6_testmanc4.blend (Minor Tips for this file: dont enable subdivions of characters until rendering, thus leave off for simulation, as it uses the rendering settings! For making nice pictures switch on subdivion, and OSA.) And here's a picture of old vs. new (for webpage or so): http://www10.informatik.uni-erlangen.de/~sinithue/blender/fluid6_manc4compare.png
2006-11-05 16:30:29 +00:00
/* no. of surface subdivisions */
int surfaceSubdivs;
2006-02-27 11:56:04 +00:00
/* global transformation to apply to fluidsim mesh */
float surfaceTrafo[4*4];
/* development variables, testing for upcoming releases...*/
float farFieldSize;
/* callback function to notify calling program of performed simulation steps
* or newly available frame data, if NULL it is ignored */
elbeemRunSimulationCallback runsimCallback;
/* pointer passed to runsimCallback for user data storage */
void* runsimUserData;
/* simulation threads used by omp */
int threads;
2006-02-27 11:56:04 +00:00
} elbeemSimulationSettings;
// defines for elbeemMesh->type below
/* please keep in sync with DNA_object_fluidsim.h */
#define OB_FLUIDSIM_FLUID 4
#define OB_FLUIDSIM_OBSTACLE 8
#define OB_FLUIDSIM_INFLOW 16
#define OB_FLUIDSIM_OUTFLOW 32
#define OB_FLUIDSIM_PARTICLE 64
#define OB_FLUIDSIM_CONTROL 128
// defines for elbeemMesh->obstacleType below (low bits) high bits (>=64) are reserved for mFsSurfGenSetting flags which are defined in solver_class.h
#define FLUIDSIM_OBSTACLE_NOSLIP 1
#define FLUIDSIM_OBSTACLE_PARTSLIP 2
#define FLUIDSIM_OBSTACLE_FREESLIP 3
#define FLUIDSIM_FSSG_NOOBS 64
#define OB_VOLUMEINIT_VOLUME 1
#define OB_VOLUMEINIT_SHELL 2
#define OB_VOLUMEINIT_BOTH (OB_VOLUMEINIT_SHELL|OB_VOLUMEINIT_VOLUME)
2006-02-27 11:56:04 +00:00
// a single mesh object
typedef struct elbeemMesh {
/* obstacle,fluid or inflow or control ... */
2006-02-27 11:56:04 +00:00
short type;
/* id of simulation domain it belongs to */
short parentDomainId;
2006-02-27 11:56:04 +00:00
/* vertices */
int numVertices;
float *vertices; // = float[n][3];
/* animated vertices */
int channelSizeVertices;
float *channelVertices; // = float[channelSizeVertices* (n*3+1) ];
2006-02-27 11:56:04 +00:00
/* triangles */
int numTriangles;
int *triangles; // = int[][3];
/* animation channels */
int channelSizeTranslation;
float *channelTranslation;
int channelSizeRotation;
float *channelRotation;
int channelSizeScale;
float *channelScale;
/* active channel */
int channelSizeActive;
float *channelActive;
/* initial velocity channel (e.g. for inflow) */
int channelSizeInitialVel;
float *channelInitialVel; // vector
/* use initial velocity in object coordinates? (e.g. for rotation) */
short localInivelCoords;
/* boundary types and settings */
short obstacleType;
float obstaclePartslip;
- bugfixes #4742 exported normals are now correct #4821 & 4956 for complex movements in/outflows can now also use the animated mesh option - new features * isosurface subdivision: directly creates a finer surface mesh from the simulation data. this increases simulation time and harddisk usage, though, so be careful - usually values of 2-4 should be enough. * fluidsim particles: extended model for particle simulation and generation. When isosurface subdivision is enabled, the particles are now included in the surface generation, giving a better impression of a single connected surface. Note - the particles are only included in the final surface mesh, so the preview surface shows none of the particle effects. * particle loading: different types of particles can now be selected for display: drops, floats and tracers. This is a bit obsolete due to the extensions mentioned above, but might still be useful. Floats are just particles floating on the fluid surface, could be used for e.g. foam. * moving objects impact factor: this is another tweaking option, as the handling of moving objects is still not conserving mass. setting this to zero simply deletes the fluid, 1 is the default, while larger values cause a stronger impact. For tweaking the simulation: if fluid disappears, try increasing this value, and if too much is appearing reduce it. You can even use negative values for some strange results :) - more code cleanup, e.g. removed config file writing in fluidsim.c, added additional safety checks for particles & fluidsim domains (these currently dont work together). I also removed the "build particles" debug message in effects.c (seemed to be unnecessary?). Some more info on the new features: Here are two test animations showing the difference between using the particle generation with isosurface subdivision. This is how it would look with the old solver version: http://www10.informatik.uni-erlangen.de/~sinithue/blender/fluid6_fl6manc4_1noparts.mpg and this with the new one: http://www10.informatik.uni-erlangen.de/~sinithue/blender/fluid6_fl6manc4_2wparts.mpg Both simulations use a resolution of 64, however, the version with particles takes significantly longer (almost twice as long). The .blend file for a similar setup can be found here: http://www10.informatik.uni-erlangen.de/~sinithue/blender/fluid6_testmanc4.blend (Minor Tips for this file: dont enable subdivions of characters until rendering, thus leave off for simulation, as it uses the rendering settings! For making nice pictures switch on subdivion, and OSA.) And here's a picture of old vs. new (for webpage or so): http://www10.informatik.uni-erlangen.de/~sinithue/blender/fluid6_manc4compare.png
2006-11-05 16:30:29 +00:00
/* amount of force transfer from fluid to obj, 0=off, 1=normal */
float obstacleImpactFactor;
/* init volume, shell or both? use OB_VOLUMEINIT_xxx defines above */
short volumeInitType;
2006-02-27 11:56:04 +00:00
/* name of the mesh, mostly for debugging */
const char *name;
/* fluid control settings */
float cpsTimeStart;
float cpsTimeEnd;
float cpsQuality;
2008-07-07 14:36:33 +00:00
int channelSizeAttractforceStrength;
float *channelAttractforceStrength;
int channelSizeAttractforceRadius;
float *channelAttractforceRadius;
int channelSizeVelocityforceStrength;
float *channelVelocityforceStrength;
int channelSizeVelocityforceRadius;
float *channelVelocityforceRadius;
2006-02-27 11:56:04 +00:00
} elbeemMesh;
// API functions
#ifdef __cplusplus
extern "C" {
#endif // __cplusplus
2006-02-27 11:56:04 +00:00
// reset elbeemSimulationSettings struct with defaults
void elbeemResetSettings(struct elbeemSimulationSettings*);
// start fluidsim init (returns !=0 upon failure)
int elbeemInit(void);
// frees fluidsim
int elbeemFree(void);
// start fluidsim init (returns !=0 upon failure)
int elbeemAddDomain(struct elbeemSimulationSettings*);
// get failure message during simulation or init
// if an error occured (the string is copied into buffer,
// max. length = 256 chars )
void elbeemGetErrorString(char *buffer);
2006-02-27 11:56:04 +00:00
// reset elbeemMesh struct with zeroes
void elbeemResetMesh(struct elbeemMesh*);
// add mesh as fluidsim object
int elbeemAddMesh(struct elbeemMesh*);
// do the actual simulation
int elbeemSimulate(void);
// continue a previously stopped simulation
int elbeemContinueSimulation(void);
2006-02-27 11:56:04 +00:00
// helper functions
// simplify animation channels
2006-02-27 11:56:04 +00:00
// returns if the channel and its size changed
int elbeemSimplifyChannelFloat(float *channel, int *size);
int elbeemSimplifyChannelVec3(float *channel, int *size);
// helper functions implemented in utilities.cpp
/* set elbeem debug output level (0=off to 10=full on) */
void elbeemSetDebugLevel(int level);
/* elbeem debug output function, prints if debug level >0 */
void elbeemDebugOut(char *msg);
/* estimate how much memory a given setup will require */
double elbeemEstimateMemreq(int res,
float sx, float sy, float sz,
int refine, char *retstr);
2006-02-27 11:56:04 +00:00
#ifdef __cplusplus
}
#endif // __cplusplus
2006-02-27 11:56:04 +00:00
/******************************************************************************/
// internal defines, do not use for initializing elbeemMesh
// structs, for these use OB_xxx defines above
2006-02-27 11:56:04 +00:00
/*! fluid geometry init types */
// type "int" used, so max is 8
2006-02-27 11:56:04 +00:00
#define FGI_FLAGSTART 16
#define FGI_FLUID (1<<(FGI_FLAGSTART+ 0))
#define FGI_NO_FLUID (1<<(FGI_FLAGSTART+ 1))
#define FGI_BNDNO (1<<(FGI_FLAGSTART+ 2))
#define FGI_BNDFREE (1<<(FGI_FLAGSTART+ 3))
#define FGI_BNDPART (1<<(FGI_FLAGSTART+ 4))
#define FGI_NO_BND (1<<(FGI_FLAGSTART+ 5))
#define FGI_MBNDINFLOW (1<<(FGI_FLAGSTART+ 6))
#define FGI_MBNDOUTFLOW (1<<(FGI_FLAGSTART+ 7))
#define FGI_CONTROL (1<<(FGI_FLAGSTART+ 8))
2006-02-27 11:56:04 +00:00
- bugfixes #4742 exported normals are now correct #4821 & 4956 for complex movements in/outflows can now also use the animated mesh option - new features * isosurface subdivision: directly creates a finer surface mesh from the simulation data. this increases simulation time and harddisk usage, though, so be careful - usually values of 2-4 should be enough. * fluidsim particles: extended model for particle simulation and generation. When isosurface subdivision is enabled, the particles are now included in the surface generation, giving a better impression of a single connected surface. Note - the particles are only included in the final surface mesh, so the preview surface shows none of the particle effects. * particle loading: different types of particles can now be selected for display: drops, floats and tracers. This is a bit obsolete due to the extensions mentioned above, but might still be useful. Floats are just particles floating on the fluid surface, could be used for e.g. foam. * moving objects impact factor: this is another tweaking option, as the handling of moving objects is still not conserving mass. setting this to zero simply deletes the fluid, 1 is the default, while larger values cause a stronger impact. For tweaking the simulation: if fluid disappears, try increasing this value, and if too much is appearing reduce it. You can even use negative values for some strange results :) - more code cleanup, e.g. removed config file writing in fluidsim.c, added additional safety checks for particles & fluidsim domains (these currently dont work together). I also removed the "build particles" debug message in effects.c (seemed to be unnecessary?). Some more info on the new features: Here are two test animations showing the difference between using the particle generation with isosurface subdivision. This is how it would look with the old solver version: http://www10.informatik.uni-erlangen.de/~sinithue/blender/fluid6_fl6manc4_1noparts.mpg and this with the new one: http://www10.informatik.uni-erlangen.de/~sinithue/blender/fluid6_fl6manc4_2wparts.mpg Both simulations use a resolution of 64, however, the version with particles takes significantly longer (almost twice as long). The .blend file for a similar setup can be found here: http://www10.informatik.uni-erlangen.de/~sinithue/blender/fluid6_testmanc4.blend (Minor Tips for this file: dont enable subdivions of characters until rendering, thus leave off for simulation, as it uses the rendering settings! For making nice pictures switch on subdivion, and OSA.) And here's a picture of old vs. new (for webpage or so): http://www10.informatik.uni-erlangen.de/~sinithue/blender/fluid6_manc4compare.png
2006-11-05 16:30:29 +00:00
// all boundary types at once
#define FGI_ALLBOUNDS ( FGI_BNDNO | FGI_BNDFREE | FGI_BNDPART | FGI_MBNDINFLOW | FGI_MBNDOUTFLOW )
2006-02-27 11:56:04 +00:00
#endif // ELBEEM_API_H