in the game engine again.
All files in source/gameengine/ I've committed below have a comment like
/* XXX note .... */
where something needs to be done. This is pretty straightforward for Erwin,
who has been notified in person too about this. :)
So for the 2.4 release, both bullet AND sumo will be enabled. You can choose in the world buttons which physics engine is active for a scene.
sorry if this breaks your builds!
IK: needed to change abs() to TNT::abs() (thanks brecht!)
Freetype: FT_ENCODING_UNICODE typecast to FT_CharMap (thanks intrr!)
header_image: missing include for liballoc stuff
Ketsji: needed to declare m_hitObject as public in KX_MouseFocusSensor.h.
This is probably the wrong thing to do, but KX_MouseFocusSensor.cpp
compilation fails on line 279 otherwise.
including implementation. hope it works, and doesn't break to much.
it bakes physics objects transform into ipo, every frame of the running gameengine.
When you disable and run the game again, it clears the ipo's again. just for physics objects at the moment.
(perhaps some better UI in the future?)
- default the m_edgecode to 65535, the wireframe was invisible. when is the edgecode available again ?
- added an extra condition, nearsensor is not yet working for bullet, but it crashed.
I removed config.h code from the files that had them.
effect.c had nested /* so cleaned that up...
added a newline to vector.c to shut gcc up ;)
buttons_editing.c had a possible unintalized var (height) so I gave
it a default value.
removed an unused var i in interface_draw.c
removed an unused var mti in outliner.c
in BL_SkinDeformer.cpp
commented out a call to bDeformGroup->data which no longer exists so it
compiles again.
Kent
added raycast support for bullet (no triangle-mesh support, soon)
added python methods for 'getHitObject', getRayDirection, getHitPosition and getHitNormal for mouse over sensor,
which makes it easy for a shootout.blend demo :)
using bandoler's patch as a basis (thanks!).
A couple of notes:
- This is for windows only, I did not have a chance to try linux yet.
- SConscript for PHY_Bullet may need tweaking (plus support
for other platforms), but at least it's in there :)
Any problems, shout :)
>e.g. my msvc6 and gcc in linux. right?
>BM
thanks bjornmose, is was a typo
(it only disabled bullet for Visual Studio 2006, which is not even in planning stage :)
Should be fixed now
Aim was to get a total refresh of the animation system. This
is needed because;
- we need to upgrade it with 21st century features
- current code is spaghetti/hack combo, and hides good design
- it should become lag-free with using dependency graphs
A full log, with complete code API/structure/design explanation
will follow, that's a load of work... so here below the list with
hot changes;
- The entire object update system (matrices, geometry) is now
centralized. Calls to where_is_object and makeDispList are
forbidden, instead we tag objects 'changed' and let the
depgraph code sort it out
- Removed all old "Ika" code
- Depgraph is aware of all relationships, including meta balls,
constraints, bevelcurve, and so on.
- Made depgraph aware of relation types and layers, to do smart
flushing of 'changed' events. Nothing gets calculated too often!
- Transform uses depgraph to detect changes
- On frame-advance, depgraph flushes animated changes
Armatures;
Almost all armature related code has been fully built from scratch.
It now reveils the original design much better, with a very clean
implementation, lag free without even calculating each Bone more than
once. Result is quite a speedup yes!
Important to note is;
1) Armature is data containing the 'rest position'
2) Pose is the changes of rest position, and always on object level.
That way more Objects can use same Pose. Also constraints are in Pose
3) Actions only contain the Ipos to change values in Poses.
- Bones draw unrotated now
- Drawing bones speedup enormously (10-20 times)
- Bone selecting in EditMode, selection state is saved for PoseMode,
and vice-versa
- Undo in editmode
- Bone renaming does vertexgroups, constraints, posechannels, actions,
for all users of Armature in entire file
- Added Bone renaming in NKey panel
- Nkey PoseMode shows eulers now
- EditMode and PoseMode now have 'active' bone too (last clicked)
- Parenting in EditMode' CTRL+P, ALT+P, with nice options!
- Pose is added in Outliner now, with showing that constraints are in
the Pose, not Armature
- Disconnected IK solving from constraints. It's a separate phase now,
on top of the full Pose calculations
- Pose itself has a dependency graph too, so evaluation order is lag free.
TODO NOW;
- Rotating in Posemode has incorrect inverse transform (Martin will fix)
- Python Bone/Armature/Pose API disabled... needs full recode too
(wait for my doc!)
- Game engine will need upgrade too
- Depgraph code needs revision, cleanup, can be much faster!
(But, compliments for Jean-Luc, it works like a charm!)
- IK changed, it now doesnt use previous position to advance to next
position anymore. That system looks nice (no flips) but is not well
suited for NLA and background render.
TODO LATER;
We now can do loadsa new nifty features as well; like:
- Kill PoseMode (can be option for armatures itself)
- Make B-Bones (Bezier, Bspline, like for spines)
- Move all silly button level edit to 3d window (like CTRL+I = add
IK)
- Much better & informative drawing
- Fix action/nla editors
- Put all ipos in Actions (object, mesh key, lamp color)
- Add hooks
- Null bones
- Much more advanced constraints...
Bugfixes;
- OGL render (view3d header) had wrong first frame on anim render
- Ipo 'recording' mode had wrong playback speed
- Vertex-key mode now sticks to show 'active key', until frame change
-Ton-
alone with the following flags :
-Wall -Wno-char-subscripts -Wno-missing-braces.
the only one still worrying me is in rand.c line 57 :
rand.c:57: integer constant is too large for "long" type
but i have no clue about how correct cross-compiler and 32/64 bits friendly
see also my mail to commiter list for signed/unsigned issues
from brian hayward (bthayward)
Detailed description:
Currently, when an armature deformed object's mesh is replaced by the ReplaceMesh actuator, the new mesh fails to deform to the armature's movement.
My patch fixes this by properly replacing the deform controller along with the mesh (when appropriete).
For instance, if one had an animated character using any of the standard deformation techniques (armature, ipo, RVK, or AVK), that character's mesh would currently be prevented from changing mid-game. It could be replaced, but the new mesh would lack the controller which tells it how to deform. If one wanted to dynamiclly add a hat on top of the character's head, it would require storing a secondary prebuilt character (mesh, armature, logic, ect...) on another layer FOR EACH HAT the character could possibly wear, then swapping out the whole character when the hat change was desired. So if you had 4 possible hat/character combos, you would have 4 character meshes, 4 armatures, 4 sets of logic, and so on. I find this lack of flexibility to be unresonable.
With my patch, one could accomplish the same thing mearly by making one version of the character in the main layer, and adding an invisible object atop the character's head (which is parented to the head bone). Then whenever it becomes desirable, one can replace the invisible object's mesh with the desirable hat's mesh, then make it visible. With my patch, the hat object would then continue to deform to the character's head regardless of which hat was currently being worn.
*note 1*
for armature/mesh deformations, the new mesh must have properly assigned vertex groups which match one or more of the bones of the target armature before the replaceMesh call is made. Otherwise the vertices won't react to the armature because they won't know how. (not sure if vertices can be scripted to change groups after the game has started)
*note 2*
The added processing time involved with replacing the object's deform controller is negligible.
I got rid of a few warnings about blah shadows a previous declaration.
In the gameengine files I fix the following:
removed some unused vars
removed dos style line breaks
added newlines to last line in a couple of files to remove warnings.
Kent
missing files in Kester's new commits. This commit adds these two files so
that compiling can continue as it is intended to. Emilie M. found a
home for the files rather quickly, Ton recieived the files from Erwin via
email, and we all tested. Works great! Thanks to everyone that helped!
To enable it you will need to download OpenEXR and install it.
For the Makefiles you will need to set WITH_OPENEXR=true
and set NAN_OPENEXR to point to where OpenEXR is installed.
For scons you'll need to remove config.opts to get the new options
so you can enable OpenEXR, I was not able to get blender to link
with scons so the scons stuff may need to be tweaked a little but
I think it should work.
For other platform managers The OpenEXR stuff is similar to QUICKTIME
you need to define WITH_OPENEXR and setup the library stuff and
as you'll notice in this commit there are two extra files.
Kent
Basically removed #include <iostream.h>
from some files that didn't need it (and I'm assuming it caused problems)
I'll leave the scons stuff for someone else to look at since I don't really
use it.
Kent
comma at end of enumerator list
If this change causes problems, feel free to back it out.
I am trying to get gcc to complain about mixed declarations and
code since that breaks on platforms that only support C89/C90.
added a casts to:
blender/blenlib/intern/BLI_memarena.c
gameengine/Rasterizer/RAS_TexVert.cpp
Removed unused vars from:
blender/python/api2_2x/vector.c
gameengine/Ketsji/KX_PyMath.cpp gameengine/Ketsji/KX_Scene.cpp
gameengine/Physics/Sumo/Fuzzics/src/SM_Object.cpp
gameengine/Rasterizer/RAS_FramingManager.cpp
gameengine/Rasterizer/RAS_Polygon.cpp
gameengine/Rasterizer/RAS_OpenGLRasterizer/RAS_VAOpenGLRasterizer.cpp
Kent
- The patch only alters files in: blender/source/gameengine/GamePlayer/ghost
- Any blenderplayer.exe (with the patch applied) can be renamed into
something.scr, and it will behave as a Windows screen saver.
(Right-click the .scr file, and select "Install". Then, it will appear
in the screen saver properties dialog part of the display settings,
including small preview et al).
- The screen saving functionality does not affect the executable when
it is NOT named .scr
- The screen saver automatically exits when the mouse/keyboard is
moved/pressed. There is no more need to wrap blenderplayer.exe into a
bat file and use python quit controllers, or whatever else is
suggested at http://download.blender.org/documentation/oldsite/oldsite.blender3d.org/181_Blender%20news%20haloParty.html
- It has been currently been tested to work Windows XP
The camera -> world transform wasn't being calculated properly for 'ortho' cameras: getOrigin()[2] should be scaled by 100 (like the render transform in KX_KetsjiEngine)
Changed:
row_pointers = new (png_bytep) [(GetHeight() * sizeof(png_bytep))];
to:
row_pointers = new png_bytep [(GetHeight() * sizeof(png_bytep))];
Thanks emil....
Kent
Keyboard sensors can now hook escape key. Ctrl-Break can be used from within blender if you've forgotten an end game actuator.
Fixed a stupid bug preventing some actuators working (like TrackTo).
Certain actuators (Add Object/Replace Mesh/Scene/Track To) used to always require a target, or they would not influence the scene. If the actuator target is always set from Python, this could be annoying.
intern/SoundSystem/intern/SND_AudioDevice.cpp
Initalized a var that could fall through with no value.
source/blender/readblenfile/intern/BLO_readblenfile.c
source/blender/src/space.c
intern/SoundSystem/intern/SND_Scene.cpp
source/gameengine/Converter/BL_MeshDeformer.cpp
removed unused var's
intern/SoundSystem/openal/pthread_cancel.cpp
fixed a nested /*
source/blender/imbuf/IMB_imbuf.h
added static to the type returned for addzbufImBuf
source/blender/imbuf/intern/IMB_bmp.h
had a wrong prototype
source/blender/src/view.c
added newline at end of file.
source/blender/src/sequence.c
removed unused var
added #include <stdlib.h> to avoid:
warning: implicit declaration of function `abs'
initalized a var that could have been used without being set.
Kent
Better use of booleans for python
#include fixes for Windows
Python Doc fixes
Use the farthest vertex as the face position when z sorting faces. (Camera is on -z axis!)
Add Python Mapping method to CListValue
Fix Bernoulli bool distribution python method for random actuator
Fix Python IpoActuator methods setProperty and force acts local
Make data objects private
Better sort method for polygon materials - much easier to understand
Reenabled the setScript/getScript methods for python controller bricks.
Set the ray source point to MouseFocusSensor.getRaySource works.
Added: Python -> MT_Quaternion
MT_Matrix4x4, MT_Matrix3x3, MT_Vector3, MT_Point3 -> Python
Correct transform of frustum bound sphere centre point to world coordinates
Profiling revealed that the SceneGraph updated every physics object, whether it moved or not, even though the physics object was at the right place. This would cause SOLID to go and update its bounding boxes, overlap tests etc.
This callback handles the special case (parented objects) where the physics scene needs to be informed of changes to the scenegraph.
Added Python attributes (mass, parent, visible, position, orientation, scaling) to the KX_GameObject module.
Make KX_GameObject use the KX_PyMath Python <-> Moto conversion.
Depth sorting for Transparent polygons. Use ZTransp in Material buttons to enable.
This will cause an object's polygons to be sorted (back to front for alpha polygons, front to back for solid polygons.)
- Mesh Objects are sorted by depth (based on object centre.) Using object centre means the user has control over the sort.
- Polygons are not sorted.
- Polygons are not split.
- O(nlog(n))
Consider:
gameobj->getClientInfo()->m_auxilary_info = (matname ? (void*)(matname+2) : NULL);
It works if matname is "MAblah", but not if matname is "".
2. Added constructor for struct RAS_CameraData.
3. Added initializers to the struct KX_ClientObjectInfo constructor
4. Collision sensors won't detect near sensors.
5. A stack of minor tweaks, adjusting whitespace, using ++it for stl stuff.