Added occlusion culling capability in the BGE.
More info: http://wiki.blender.org/index.php/Dev:Ref/Release_Notes/2.49/Game_Engine#BGE_Scenegraph_improvement
MSVC, scons, cmake, Makefile updated.
Other minor performance improvements:
- The rasterizer was computing the openGL model matrix of the objects too many times
- DBVT view frustrum culling was not properly culling behind the near plane:
Large objects behind the camera were sent to the GPU
- Remove all references to mesh split/join feature as it is not yet functional
- Most mesh particle effectors can now have their effection point taken per particle as the nearest point on the mesh surface.
- This is activated with the "surface" button in the effector field panel.
- Activating the option adds a "surface" entry to the modifier stack where the state of the mesh is read from.
For an example of usage see http://www.youtube.com/watch?v=3XkO1EAmJks.
User guide:
http://wiki.blender.org/index.php/Dev:Source/GameEngine/Fisheye_Dome_Camera
Fixed two bugs from original patch:
- deleting a text will clear the warp field from Game framing settings
- removed spurious black dots along the edge of the cube map in the gameplayer
Known limitation:
- resizing of the screen doesn't work in the gameplayer
Known bugs:
- Texture with reflexion are not rendered correctly
- Spurious problems with light
This commit contains a number of performance improvements for the
BGE in the Scenegraph (parent relation between objects in the
scene) and view frustrum culling.
The scenegraph improvement consists in avoiding position update
if the object has not moved since last update and the removal
of redundant updates and synchronization with the physics engine.
The view frustrum culling improvement consists in using the DBVT
broadphase facility of Bullet to build a tree of graphical objects
in the scene. The elements of the tree are Aabb boxes (Aligned
Axis Bounding Boxes) enclosing the objects. This provides good
precision in closed and opened scenes. This new culling system
is enabled by default but just in case, it can be disabled with
a button in the World settings. There is no do_version in this
commit but it will be added before the 2.49 release. For now you
must manually enable the DBVT culling option in World settings
when you open an old file.
The above improvements speed up scenegraph and culling up to 5x.
However, this performance improvement is only visible when
you have hundreds or thousands of objects.
The main interest of the DBVT tree is to allow easy occlusion
culling and automatic LOD system. This will be the object of further
improvements.
- Explode didn't use the dmcache index for getting the particles emitter position.
- One "tri or quad"-comparison tested the wrong index. Leading to one quad converting into a tri.
If the density texture was taken into account with hair parent particles there were cases when there weren't any parents left to interpolate children from. Now a density texture is only taken into account for hair child particles.
-Virtual parents were not randomly selected due to optimization in child particle distribution code.
-Wave-kink had a wrong matrix multiplication.
-Amount of virtual parents wasn't scaled properly to the amount of children rendered.
-Calculating virtual parents is now thread safe.
ffmpeg format switched from RGBA32 to BGR32, it's no longer needed to swap color planes.
NOTE: this commit also attempts to fix the big endian case, but since I don't have a machine to test it, I'd appreciate if someone else would.
own fault, broke rev16702.
Curves created by fonts didnt have their radius set. Forgot do do this when making radius calculated with the curve (like tilt)
[18429] Typo in IPO Actuator
[18377] Crash fo yofankie (G.curscreen==NULL)
/* No screen, happens when saving a blendfile in background mode,
* then loading in the game engine
* just assume we need the mesh info */
particleInstanceModifier_applyModifier uses getVertCo,which is not implemented by ccgDM.
getVertCo was used to simply perform a min/max, so it was fixed to use the apropriate min/max
dm->getMinMax(dm, min_r, max_r);
Maybe the fact ccgDM doesn't implements getVertCo/getVertNo is a bug, since it implements every
other DM function and its strange those are left out.
Without thread locking the function that allocates new threads, black dots appear in renders.
This wont affect composite nodes,
Ton/Brecht - this is possibly too many lock/unlocks but I timed the render from the bug report and it didn't make a noticeable difference.
Support for jpeg2000 and writing DCI Cinema standard files.
Notes
* 12 and 16bit channel depths are converted from/to blenders float buffer.
* Grayscale/RGB with alpha supported.
* Theres an option to save color channels as YCC rather then RGB.
* Quality 100 saves lossless
* The UI is a bit weired because of the DCI standards need to be given to the encoder.
Additions/Features...
- Painting onto multiple images at once.
- Clone with an offset similar to the gimp, also an option to clone between UV layers, (not both offset and between layers atm)
- Stencil layer, similar to The Gimp's quickmask.
- Face mask mode - (painting onto only selected faces).
- Texture brushes (so image and texture-node brushes too)
- Seam bleed so you don't see gaps when painting over UV seams (new option set in the panit panel)
- Occlusion (only point onto the first face under the paintbrush, can also be used for masking out parts of the model to paint onto) - (new option set in the paint panel)
- Muti-threaded, each thread operates on its own bucket.
- Support partial updates to speedup OpenGL texture refreshing.
- Support tile based multiple undo's
- Support float image buffers.
- Backface culling (new option set in the paint panel)
- All color blending modes - Add/Multiply.. etc.
- Airbrush rate. (used to be broken)
- Improve the way paint is mixed when painting, so painting with 0.2 alpha will not accumulate above 0.2 alpha on the image as you paint.
- Option to use normals to fade out strokes on the edge of a surface to avoid ugly edges. (same as Vertex Paints "Normal" option)
Known Problems
- Not directly related to painting but changes in selection flags are not updated on the final derived mesh.
- Smear tool has some distortion when in perspective mode.
- While in perspective view any faces with 1 vert behind the viewpoint, will be completely ignored.
- Painting can be initially slow while buckets are being initialized, a number of solutions exist.
- Size 1 brushes don't work yet.
- When painting onto large faces its still possible to get artifacts.
- Spatial optimizations are needed to skip occluded faces and test pixel occlusion. (will look into using blenders BVH )
* pixels with <= the current opacity are not painted onto, speeds up painting especially with low spacing. (only when airbrush is disabled)
* use qsort rather then own crufty sorting function
* grouped projection painting functions together.
* float buffer support for all painting operation and undo.
* only run brush_sample_tex() for textures brushes.
* redraw the brush outline even when nothing is painted.
Robin (Frrr) Allen did a decent job on this, so we can also welcome him
as a member in the svn committers team to maintain it!
I do the first commit with some minor fixes:
- get Makefiles work
- fix rounding issue with tiles on unit faces
- removed UI includes from tex node
A nice doc in wiki is here:
http://wiki.blender.org/index.php/User:Frr/TexnodeManual
On the todo for Robin is:
- When using one or more Texture-input nodes, you cannot edit them by activating
(as works now for Material nodes).
- The new "output node" option fails on the default case, when only one
output node is active. It then shows often a blank menu. Will get fixed asap.
- When using a NodeTree-Texture as input node, the menu for 'active output'
should not show. NodeTree should ignore other nodetrees to keep things sane
for now.
- On a future todo is proper usage of "Dxt" and "Dyt" texture vectors for
superior antialising of checkers/bricks.
General note; I know people are dying to get a full integrated shader system
with nodes. In theory we could merge this with Material Nodetrees... but I
rather wait for a solid and very well thought out design proposal for this,
also including design ideas for unifying with a shader language (GPU, CPU).
For the time being this is a nice extension of current textures. :)
