Commit Graph

91 Commits

Author SHA1 Message Date
Robert Osfield
7e43decf4c From Wojciech Lewandowski, build fixes for handling Matrix::value_type == float 2008-10-06 14:18:27 +00:00
Robert Osfield
45085f3eea Converted tabs to four spaces 2008-10-06 08:58:50 +00:00
Robert Osfield
87e8f06522 Ran dos2unix on new shadow implemenations 2008-10-06 08:53:25 +00:00
Robert Osfield
bf15614df6 From Wojciech Lewandowski, a range of new ShadowMap implementations include View Dependent Shadow Maps - LispSM
From Robert Osfield, refactored Wojciech's submission to integrated with osgShadow and use coding sytle more consistent with rest of OSG.
2008-09-19 19:51:27 +00:00
Robert Osfield
1d328ba0d4 From Christopher Blaesius,
"Soft shadow mapping is basically the same as hard shadow mapping beside that
it uses a different fragment shader.
So for me it makes sense that osgShadow::SoftShadowMap is derived from
osgShadow::ShadowMap, this makes it easier to maintain the two classes.
Additional SoftShadowMap also provides the same Debug methods as ShadowMap."
2008-09-18 14:48:28 +00:00
Robert Osfield
9cca510d6c From Chris Denham, added missing ccopy of polyOffset in copy constructor 2008-09-18 13:09:21 +00:00
Robert Osfield
40723ca8ac From Chris Denham, "
I think I may have discovered a bug in osgShadow/ShadowMap.cpp that results in incomplete shadows being generated.
The problem seems to caused by an incorrect interpretation of the spot light cutoff angle. The valid ranges for spot cutoff are 0-90 and 180, i.e half the 'field of view' for the spotlight. Whereas the shadow map code seems to assume the the spot cutoff is equal to the field of view. This results in the shadows generated by the spotlight getting clipped at half the spot cutoff angle.

I have fixed this in my copy of ShadowMap.cpp:
===============================
//Original code from OSG 2.6:
      if(selectLight->getSpotCutoff() < 180.0f)   // spotlight, then we don't need the bounding box
      {
          osg::Vec3 position(lightpos.x(), lightpos.y(), lightpos.z());
          float spotAngle = selectLight->getSpotCutoff();
          _camera->setProjectionMatrixAsPerspective(spotAngle, 1.0, 0.1, 1000.0);

_camera->setViewMatrixAsLookAt(position,position+lightDir,osg::Vec3(0.0f,1.0f,0.0f));
      }
===============================
// My modifications:
      float fov = selectLight->getSpotCutoff() * 2;
      if(fov < 180.0f)   // spotlight, then we don't need the bounding box
      {
          osg::Vec3 position(lightpos.x(), lightpos.y(), lightpos.z());
          _camera->setProjectionMatrixAsPerspective(fov, 1.0, 0.1, 1000.0);

_camera->setViewMatrixAsLookAt(position,position+lightDir,osg::Vec3(0.0f,1.0f,0.0f));
      }

This change seems correct for spot cutoff in the range 0, 90, but since OpenGL doesn't claim to support cutoffs >90 && <180, I'm not sure how shadow map should deal with those cases, but ignoring spot cut off greater than 90 here seems reasonable to me.
"
2008-09-18 13:05:24 +00:00
Robert Osfield
ff119b78cd From Adrian Egli, "i came around reviewing my code, and found now finally a solution to remove the polygon offset issue. as we all know the polygon offset has a different behaviour on different GPU system (ATI, NVidia) and this make the use of
polygon offset complicate. so i looked for a solution to remove this offset.

i changed the shader, also the filtering (default: on) use now a correct 3x3 filter:
 1 0 1
 0 2 0
 1 0 1

div: 6

of course a better one would be
 1 2 1
 2 4 2
 1 2 1

div: 16
but this isn't as performant as the simple filter above is. because we need only 5 texture lookups instead of 9, and the result is still good, if you wish we can add a enum to change the pcf filter type once, if there is a need.


testet on NVidia Quatro 570M and on ATI Radeon X1600

"
2008-09-17 18:54:23 +00:00
Robert Osfield
19d02d8c78 From Adrian Egli, "I changed the PSSM shadow map implementation, if we have filtered turned on, it should be now correct. The implementation is more robut on different scene. i tested it on NVIDIA card against a park scene, a chess board and a terrain.
unfort. i couldn't test it on any ATI system. may there will be still another problem there. if there are still some artefacts. we should try out better fZOffSet value
"
2008-09-17 18:51:17 +00:00
Robert Osfield
7f8183f64c Merged changes to OpenSceneGraph-2.6 into trunk using :
svn merge -r 8729:8734 http://www.openscenegraph.org/svn/osg/OpenSceneGraph/branches/OpenSceneGraph-2.6 .
2008-08-05 11:17:48 +00:00
Robert Osfield
3e96a9d448 Missing check-ins 2008-07-15 09:55:33 +00:00
Robert Osfield
0faaf93dc2 From Adrian Egli, "Improvements to the PSSM implementation" 2008-06-19 14:45:54 +00:00
Robert Osfield
1047f970bb From Adrian Egli, improvements to ParallelSplitShadowMap implementation 2008-05-28 14:11:22 +00:00
Robert Osfield
78964b4baa Reverted the changes for ill fated tweaks for Apple build 2008-04-24 11:34:43 +00:00
Robert Osfield
8ae30c9bfc Refactored the inclusion of the headers in the ADD_LIBRARY so that they aren't
added under OSX.
2008-04-23 20:50:53 +00:00
Robert Osfield
bbf4564103 From Sebastian Messerschimdt, Added polygon offset controls 2008-03-19 12:09:20 +00:00
Robert Osfield
207a0e8610 Revert the Camera::setInheritanceMask(0x) calls that were added as workaround to a bug introduced by a bug fix to the way the inhertiance mask was handled in CullVisito::apply(Camera&) 2008-03-19 11:41:21 +00:00
Robert Osfield
945846a3ed Set the CullSettings::InhertianceMask to 0x0 as temporary measure till we
work out the best scheme to use w.r.t inhertiance of CullSettins.
2008-03-18 17:44:21 +00:00
Robert Osfield
dd2b4ed83e From Wojciech Lewandowski, "Attached is modified osgShadow::ShadowMap. I changed following things:
1:
Shadow map camera sets ABSOLUTE_RF_INHERIT_VIEWPOINT refernce frame. 
 
2:
Light Direction  by matrix multiplications replaced with transform3x3 multiplication.
 
3:
I made DebugingHUD functional by adding special draw callback. Former version was simply drawing pale square. 
 
4:
I was tempted to make 4 th change but decided to not do it. Instead I put it whith #if VIEW_DEPNDENT_TEXGEN. If you decide you may let it go.
 
When objects are not centered at 0,0,0 coord but in some distant location (for example at surface of earth ellipsoid) shadow texgen suffers from inadequate precision of float matrices. I changed that by premultiplying Texgen matrix (using OSG double matrices) with inverse modelview and applying it later with ModelView identity matrix. This tweak may be appropriate for OverlayNode texgen as well.
 
I left former version because I suspect that this change will make osgShadow::ShadowMap view dependant. Currently texgen matrix remains the same no matter what View displays it. With my change it wuld be different for each view. This touches the subject of View Dependent Shadow Techniques that J-S asked recently."
2008-03-04 16:22:47 +00:00
Robert Osfield
1cfba8787a From Paul Martz, Hi Robert -- "as you and J-S and I have been discussing on osg-users, this submission changes the clamp mode for the fake white texture applied to non-textured geometry. CLAMP_TO_EDGE will keep the geometry a consistent color across the entire texture coordinate range." 2008-01-18 20:55:16 +00:00
Robert Osfield
9a2bc98d3a From Paul Obermeier, "Please find enclosed the following 2 bug fixes:
File osgShadow/Version.cpp, Line 25:

const char* osgShaodowGetLibraryName()

should be:

const char* osgShadowGetLibraryName()


File CMakeModules/OsgMacroUtils.cmake, Line 224:

SET_TARGET_PROPERTIES(${TARGET_TARGETNAME} PROPERTIES DEBUG_POSTFIX ${CMAKE_DEBUG_POSTFIX})

should be:

SET_TARGET_PROPERTIES(${TARGET_TARGETNAME} PROPERTIES DEBUG_POSTFIX "${CMAKE_DEBUG_POSTFIX}")

Otherwise setting CMAKE_DEBUG_POSTFIX to an empty string instead of "d" in
the main CMakeLists.txt does not work under Linux.
"
2007-12-17 18:38:21 +00:00
Robert Osfield
b76b85770f From Wojiech Lewandowski, "There was a bug in ShadowMap camera view matrix computation. View matrix was wrong when light was directional and shadowed scene was not centered at zero coord. I fixed that and also modified cast distance to much smaller value. With former range it was possible to generate shadows with lowest LODs. " 2007-12-16 17:41:46 +00:00
Robert Osfield
8b21b87d96 Added setting of ambient bias uniform 2007-12-15 15:22:00 +00:00
Robert Osfield
2ab66d9948 Tweaked the abmient lighting contribution so that the OpenGL vertex lighting has
the ambient light source switched off, and use the fragment shader to add this
contribution back in.
2007-12-15 15:17:51 +00:00
Robert Osfield
f4afa427a7 From Roland Smeenk, "Attached you will find a large set of small typo fixes (mainly in the comments)." 2007-12-10 17:30:18 +00:00
Robert Osfield
6ff9810be1 From Mihai Radu, "After someone asked for a fix for non-textured object appearing black
with the shader for textured objects. This works very well in cases
where there could be a mix of textured and non-textured objects in the
scene, and it makes the initialization more robust.
The idea is from PSSM, to add a 1pixel texture to the main rendering  as
to provide white for any objects missing textures."
2007-12-09 16:06:14 +00:00
Robert Osfield
b3ecb8f38f From Mahai Radu, improvements to handling of spotlights 2007-10-03 12:48:23 +00:00
Robert Osfield
1dc06b4553 From Mihair Radu, "Most of the additions are small utility methods:
- set the resolution of the shadow map; it calls dirty() to
re-initialize at next update
- keep a list of Shader objects to use instead of the default ones, if
the list is empty, the default shaders are used
- explicitly create the Uniform variables, so that subsequent additions
that require more Uniforms can put them in a central place
- set a Light or LightSource to use explicitly for shadow casting,
allows multiple lights in the scene, with one casting shadows

There are two additions that do not ( yet ) function correctly, but in
the present usage they do not interfere with the regular usage of the
techique:
- support for using spotlights, it's using Light.spotCutoff to determine
if it's a spot-light and not point-light,
   there is an error in the setup of either the shadow camera or the
texgen, most likely due to the direction of the spotlight, since the
position is being used just like in point or directional lights.
- creation of a debugHUD
   the hud is created properly, ( the example included shows it ), but
it displays only white, there has been some discussion of displaying the
shadow map, but I could not find it, the addition of a simple fragment
shader with the appropriate color transform should get this going."
2007-10-02 21:45:09 +00:00
Robert Osfield
6df0110d0c From Adrian Egli, "update the PSSM, i did some bug fixes and added new features to move the camera virtual against the view direction by applaying a simple distance factor (a), which is calculated camera eye - camera center distance. and we can move the "light camera" against the light direction (b).
(a) some objects behind the camera can cast shadow
(b) object aboive the camera can cast shadow

then i fixed the shadow map orientation, now screen x coordinate alinged which improve the quality"
2007-09-27 12:47:34 +00:00
Robert Osfield
d11d5b819a From Adrian Egli, further work on PSSM implementation. 2007-09-24 15:24:23 +00:00
Robert Osfield
82b0f004e9 From Adrian Egli, update to ParallelSplitShadowMap. 2007-09-22 16:46:38 +00:00
Robert Osfield
55a9ea19d7 From Anders Backman, added bias and texture size methods.
From Robert Osfield, tweaks to the above for method parameters and indenting.
2007-09-20 10:12:58 +00:00
Robert Osfield
cf69352873 From Adrian Egli, ParallelSplitShadowMap implmentation.
Changes by Robert Osfield - CMakeLists.txt and name change to ParallelSplitShadowMap.
2007-09-19 11:56:44 +00:00
Robert Osfield
5caf64e5c7 From Paul Martsz, changed the Texture clamping mode to CLAMP_TO_EDGE 2007-09-13 10:32:05 +00:00
Robert Osfield
3c82058c68 Updated wrappers 2007-08-16 15:29:50 +00:00
Robert Osfield
fe66501c47 From Stephan Huber, "the softshadowmap-implementation does not compile on os x, to be more
precise, the shader does not compile on os x because of some
type-conflicts ala "can not convert from const int to const float"

So I changed the offending lines to force the type of the vars. It works
now on OS X (albeit very slowly, 3fps on a 7300), perhaps you find the
changes useful. Note: perhaps there is a better way in shaders to
cast/convert from int to float and viceversa."
2007-08-15 15:07:56 +00:00
Robert Osfield
e4e89101b6 Replaced uint with unsigned int 2007-08-14 08:41:43 +00:00
Robert Osfield
932608f804 From Bruno Hebelin,
"I have adapted to osgShadow the soft shadow map technique described in  "Efficient Soft-Edged Shadows Using Pixel Shader Branching" by Yury Uralsky, Chapter 17 of GPU Gems 2 (Matt Pharr ed. Addison-Wesley).

Here is my code in attachment: basically, it works in the same way as osgShadow/ShadowMap (core code is copied from it) but implements a specific GLSL shader for the soft rendering of penumbra.

I have tested it under Linux with a NVidia graphic card, but there should be no dependency on platform nor on the graphics driver (as far as they support GLSL 2). Screenshots attached show the current results (frame rate bound to v-sync, but the shader takes actually not much time)."
2007-08-12 12:12:55 +00:00
Robert Osfield
eac3dc1963 From Paul Melis, "Here is a list of fixes to misspelled APIs. Unfortunately, more than one
of these are public APIs and therefore will break linkage to existing
shared libraries."

Note from Robert Osfield, updated wrappers.
2007-06-27 20:36:16 +00:00
Robert Osfield
bd6a001dcc Added default values for ambientBias 2007-06-15 14:57:17 +00:00
Robert Osfield
08d3b13430 Added ShadowMap::s/getAmbientBias, updated NEWS and wrappers 2007-06-15 12:59:22 +00:00
Robert Osfield
7016db5f26 Removed ParallelSplitShadowMap as its currently a non op. 2007-06-13 09:56:12 +00:00
Robert Osfield
a9d9d387df Fixed spelling of Receives 2007-06-04 08:38:11 +00:00
Robert Osfield
58fc56aabd Made the local shaders definitions static const char to avoid multiple definiations 2007-05-23 15:14:18 +00:00
Robert Osfield
3809d0dad0 Moved the OpenThreads link locally to each lib 2007-05-03 10:06:38 +00:00
Robert Osfield
f242570269 Unified the setup of version numbers so that they all are based on the version
number setup in the include/osg/Version header file.
2007-04-22 20:19:43 +00:00
Robert Osfield
68a6882d04 Bumbped the version number up to 1.9, in preperation for using odd release numbers
for development releases, odd versions for stable releases.
2007-04-16 06:03:22 +00:00
Robert Osfield
2e6ee4bb39 Removed old gmake and VisualStudio build files 2007-04-10 13:10:25 +00:00
Robert Osfield
31ac332b1f Add setting of ShadowVolumeGeometry draw mode 2007-04-06 14:33:55 +00:00
Robert Osfield
b419fa93ef From Luigi Calori, "when we link against something that comes out from a Find... we ususally have a variable <LINK_VAR_NAME>available like OPENTHREADS_LIBRARY,
so I' ve set up a macro that uses the variable name expanded for linking, and  test if a variable ${LINK_VAR_NAME}_DEBUG
like OPENTHREADS_LIBRARY_DEBUG exists and in case uses it for linking in debug mode.
I' ve also set up FindOpenThreads to set up these variables.
I had to edit the core libraries CMakeLists to add the calls to the macros used.
I' ve tested under MSVC"
2007-03-26 13:02:38 +00:00