compilation. Currently, it is not yet active by default, but
Genscher wanted to do some tests.
It can be used to distribute the computation load onto multiple shared-
memory CPUs by splitting the domain along the y-axis (assuming a
gravity force along z). However, there is no load balancing: so
if there's fluid only in one of the y-axis halves you will not get
a speedup for 2 CPUs.
- Added a fix for the memory allocation bugs #7120 and #6775. In
solver_init.cpp there are now several variables max___MemChunk
(line 692+), that set upper limits for various systems. The same
problem existed for mac & linux, but the limit is higher, so
it probably went by undetected. The windows limit is currently 1GB,
if the strange 700MB limit problems mentioned in the bug regports the
bugs persist, this could be further reduced. For 64bit compilations
this problem shouldn't exist anyway.
What's still missing is a display of how much the resolution was
reduced to fit into memory...
- And some minor solver code cleanup.
#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
of the fluidsim can now be disabled with the
flag: BF_NO_ELBEEM='true', e.g. for irix systems.
(The number of ifdefs from the original
patch was reduced, and the defines are now
only necessary when elbeem is switched off.)
- particle generation option is available again
- Added part of Austin's msvc8 fixes (vector::erase function
was "misused"), hopefully compiles better now.
- Ctrl-b now also bakes a selected fluidsim domain
similar to the softbodies.
- Added surface smoothing option for domains: default is
1, higher values result in a smoother surface (and probably
slightly higher comupation times), while 0 means the surface
is not modified at all.
- Added BLENDER_ELBEEMBOBJABORT environment variable in readBobj,
if >0 quits blender when a not yet existing fluidsim
frame should be loaded. Useful for rendering simulations
as far as possible from the command line.
- Surface normals pointer is now set to NULL in readfile.c
- Fixed win32 error string handling, now uses a function
to return the string from the solver.
- Fixed fluidsim particle halo scaling problem.
- Solver update
issues in parallel... So this commit contains: an update of
the solver (e.g. moving objects), integration of blender IPOs,
improved rendering (motion blur, smoothed normals) and a first particle
test. In more detail:
Solver update:
- Moving objects using a relatively simple model, and not yet fully optimized - ok
for box falling into water, water in a moving glass might cause trouble. Simulation
times are influenced by overall no. of triangles of the mesh, scaling meshes up a lot
might also cause slowdowns.
- Additional obstacle settings: noslip (as before), free slip (move along wall freely)
and part slip (mix of both).
- Obstacle settings also added for domain boundaries now, the six walls of the domain are
obstacles after all as well
- Got rid of templates, should make compiling for e.g. macs more convenient,
for linux there's not much difference. Finally got rid of parser (and some other code
parts), the simulation now uses the internal API to transfer data.
- Some unnecessary file were removed, the GUI now needs 3 settings buttons...
This should still be changed (maybe by adding a new panel for domain objects).
IPOs:
- Animated params: viscosity, time and gravity for domains. In contrast
to normal time IPO for Blender objects, the fluidsim one scales the time
step size - so a constant 1 has no effect, values towards 0 slow it down,
larger ones speed the simulation up (-> longer time steps, more compuations).
The viscosity IPO is also only a factor for the selected viscosity (again, 1=no effect).
- For objects that are enabled for fluidsim, a new IPO type shows up. Inflow
objects can use the velocity channels to animate the inflow. Obstacles, in/outflow
objects can be switched on (Active IPO>0) and off (<0) during the simulation.
- Movement, rotation and scaling of those 3 types is exported from the normal
Blender channels (Loc,dLoc,etc.).
Particles:
- This is still experimental, so it might be deactivated for a
release... It should at some point be used to model smaller splashes,
depending on the the realworld size and the particle generation
settings particles are generated during simulation (stored in _particles_X.gz
files).
- These are loaded by enabling the particle field for an arbitrary object,
which should be given a halo material. For each frame, similar to the mesh
loading, the particle system them loads the simulated particle positions.
- For rendering, I "abused" the part->rt field - I couldnt find any use
for it in the code and it seems to work fine. The fluidsim particles
store their size there.
Rendering:
- The fluidims particles use scaled sizes and alpha values to give a more varied
appearance. In convertblender.c fluidsim particle systems use the p->rt field
to scale up the size and down the alpha of "smaller particles". Setting the
influence fields in the fluidims settings to 0 gives equally sized particles
with same alpha everywhere. Higher values cause larger differences.
- Smoothed normals: for unmodified fluid meshes (e.g. no subdivision) the normals
computed by the solver are used. This is basically done by switching off the
normal recalculation in convertblender.c (the function calc_fluidsimnormals
handles other mesh inits instead of calc_vertexnormals).
This could also be used to e.g. modify mesh normals in a modifier...
- Another change is that fluidsim meshes load the velocities computed
during the simulation for image based motion blur. This is inited in
load_fluidsimspeedvectors for the vector pass (they're loaded during the
normal load in DerivedMesh readBobjgz). Generation and loading can be switched
off in the settings. Vector pass currently loads the fluidism meshes 3 times,
so this should still be optimized.
Examples:
- smoothed normals versus normals from subdividing once:
http://www10.informatik.uni-erlangen.de/~sinithue/temp/v060227_1smoothnorms.pnghttp://www10.informatik.uni-erlangen.de/~sinithue/temp/v060227_2subdivnorms.png
- fluidsim particles, size/alpha influence 0:
http://www10.informatik.uni-erlangen.de/~sinithue/temp/v060227_3particlesnorm.png
size influence 1:
http://www10.informatik.uni-erlangen.de/~sinithue/temp/v060227_4particlessize.png
size & alpha influence 1:
http://www10.informatik.uni-erlangen.de/~sinithue/temp/v060227_5particlesalpha.png
- the standard drop with motion blur and particles:
http://www10.informatik.uni-erlangen.de/~sinithue/temp/elbeemupdate_t2new.mpg
(here's how it looks without
http://www10.informatik.uni-erlangen.de/~sinithue/temp/elbeemupdate_t1old.mpg)
- another inflow animation (moving, switched on/off) with a moving obstacle
(and strong mblur :)
http://www10.informatik.uni-erlangen.de/~sinithue/temp/elbeemupdate_t3ipos.mpg
Things still to fix:
- rotating & scaling domains causes wrong speed vectors
- get rid of SDL code for threading, use pthreads as well?
- update wiki documentation
- cool effects for rendering would be photon maps for caustics,
and motion blur for particles :)
and viscosity, an example can be found here:
http://www10.informatik.uni-erlangen.de/~sinithue/temp/fluid_timeanim.mpg
- for simulation time animation the time IPO of the object is currently used,
for all three there should probably be new ipos in the fluidsim struct
- started the API in elbeem.cpp, to get rid of parser & export
via HD (it's not yet used)
- debug output now controlled globally by elbeem debug level
(BLENDER_ELBEEMDEBUG environment var), also for fluidsimBake
and read/writeBobj
- debug output is written to file for WIN32
- added "for" and "vector" etc. defines for MSVC6
(I couldnt get hold of the compiler itself, so not tested yet)
