Add log warnings when users try to set depth/stencil textures which have no depth color format.
git-svn-id: svn://svn.code.sf.net/p/irrlicht/code/trunk@5845 dfc29bdd-3216-0410-991c-e03cc46cb475
Delay switching back to previous active texture until mipmap generation is done.
Avoids 4 GL calls in most cases (not really noticable for speed, but makes reading api-traces of GL calls a bit easier).
git-svn-id: svn://svn.code.sf.net/p/irrlicht/code/trunk@5842 dfc29bdd-3216-0410-991c-e03cc46cb475
Before it updated only on move events. That could lead to troubles when the camera was actived by click as it then started the rotation with the position of the last move event it had.
git-svn-id: svn://svn.code.sf.net/p/irrlicht/code/trunk@5835 dfc29bdd-3216-0410-991c-e03cc46cb475
Just making it easier to see when DepthStencil can be set (we never have a depth texture format when we have no texture).
git-svn-id: svn://svn.code.sf.net/p/irrlicht/code/trunk@5833 dfc29bdd-3216-0410-991c-e03cc46cb475
SW_SHOWMINIMIZED isn't a bitflag as was probably assumed.
git-svn-id: svn://svn.code.sf.net/p/irrlicht/code/trunk@5827 dfc29bdd-3216-0410-991c-e03cc46cb475
This mostly happens because we merge vertices by position in the meshloader. But such triangles tend to cause troubles and won't render, so kick them out.
git-svn-id: svn://svn.code.sf.net/p/irrlicht/code/trunk@5826 dfc29bdd-3216-0410-991c-e03cc46cb475
GenerateTextureMipmap had been forgotten in initialization when they got added recently.
Re-ordering initialization to avoid compiler warnings.
git-svn-id: svn://svn.code.sf.net/p/irrlicht/code/trunk@5823 dfc29bdd-3216-0410-991c-e03cc46cb475
Thx @ criss and devsh for this patch (got applied with minor adaptions).
New functions are about functions using new direct state access functions from GL 4.5
And array textures.
Not yet used internally in the engine (but possible to access it with some hacks from apps, we probably should make the extension handler public at some point).
Also now using glActiveTexture as default instead of glActiveTextureARB.
git-svn-id: svn://svn.code.sf.net/p/irrlicht/code/trunk@5818 dfc29bdd-3216-0410-991c-e03cc46cb475
Deprecate PolygonOffsetFactor and PolygonOffsetDirection in SMaterial.
Replace it by PolygonOffsetDepthBias and PolygonOffsetSlopeScale.
Old values still work for now (as well as they did), but will be removed after Irrlicht 1.9.
The old implementation was based a lot on the way Direct3D8 had worked.
- We only had values -1 and 1 for the slope bias before, but sometimes other values are necessary.
- An int value for PolygonOffsetFactor couldn't worked for Direct3D9 which (unlike D3D8) uses a value range of -1 to 1.
Thx @ Criss and devsh for implementing some code which showed that different slope scaling is sometimes needed.
git-svn-id: svn://svn.code.sf.net/p/irrlicht/code/trunk@5817 dfc29bdd-3216-0410-991c-e03cc46cb475
(was still unix-style as I just copied this header from khronos).
git-svn-id: svn://svn.code.sf.net/p/irrlicht/code/trunk@5816 dfc29bdd-3216-0410-991c-e03cc46cb475
Note: Khronos made glext.h now depend on a new file KHR/khrplatform.h
The idea seems to be that some common types can be shared with other gl headers (for gles versions).
But I run into some problem with that as it was included with <> instead of "" so Visual Studio wouldn't find
it without adding include paths to all projects for new KHR folder. Didn't want that (can't test that for example on iOS),
so I modified official khronos header by changing the include to using "".
git-svn-id: svn://svn.code.sf.net/p/irrlicht/code/trunk@5815 dfc29bdd-3216-0410-991c-e03cc46cb475
While COpenGLCoreCacheHandler wasn't meant to be accessed externally, it is possible with some hacks (a few casts and including internal headers) and sometimes necessary.
And that's one of the state-flags people have a need to know from outside the engine.
git-svn-id: svn://svn.code.sf.net/p/irrlicht/code/trunk@5814 dfc29bdd-3216-0410-991c-e03cc46cb475
Before they could get pretty large as each node used the maximum memory it started with.
Also some changes to avoid unnecessary memory re-alloction when creating it.
Thanks @Squarefox for reporting the problem and proposing a patch.
See discussion at: http://irrlicht.sourceforge.net/forum/viewtopic.php?f=7&t=52484
git-svn-id: svn://svn.code.sf.net/p/irrlicht/code/trunk@5813 dfc29bdd-3216-0410-991c-e03cc46cb475
Thx@ Squarefox for reporting (http://irrlicht.sourceforge.net/forum/viewtopic.php?f=7&t=52484)
Basically the loop was split into 2 loops so children do their allocations after the parent node
has finished his. Otherwise we got fragmentation by chaotic allocation order where parents/childs
switched all the time.
Also indention of loop changed (sorry, should have done that before).
git-svn-id: svn://svn.code.sf.net/p/irrlicht/code/trunk@5812 dfc29bdd-3216-0410-991c-e03cc46cb475
In r5805 they got broken when adding compile-flags for them.
Copy-paste error, define was missing a '_' at the end in some places, sorry.
git-svn-id: svn://svn.code.sf.net/p/irrlicht/code/trunk@5810 dfc29bdd-3216-0410-991c-e03cc46cb475
(probably was added accidental, as it was always enabled that way which kills it's original purpose).
Thx@ LunaRebirth for report (http://irrlicht.sourceforge.net/forum/viewtopic.php?f=1&t=52471)
git-svn-id: svn://svn.code.sf.net/p/irrlicht/code/trunk@5807 dfc29bdd-3216-0410-991c-e03cc46cb475
Following flags got added:
_IRR_COMPILE_WITH_OCTREE_SCENENODE_
_IRR_COMPILE_WITH_TERRAIN_SCENENODE_
_IRR_COMPILE_WITH_SHADOW_VOLUME_SCENENODE_
_IRR_COMPILE_WITH_BILLBOARD_SCENENODE_
_IRR_COMPILE_WITH_WATER_SURFACE_SCENENODE_
_IRR_COMPILE_WITH_SKYDOME_SCENENODE_
_IRR_COMPILE_WITH_CUBE_SCENENODE_
_IRR_COMPILE_WITH_SPHERE_SCENENODE_
Also the corresponding _NO_IRR_COMPILE_WITH... flags certainly.
More such flags might be added in future, but will test first if it really makes a difference for WebGL apps which were the main reason to add this.
git-svn-id: svn://svn.code.sf.net/p/irrlicht/code/trunk@5805 dfc29bdd-3216-0410-991c-e03cc46cb475
Names like "my%20texture.png" will now load "my texture.png". Collada filenames are in xs:anyURI format.
xs:anyURI is used in more places, but we don't support any other file-loading inside Collada so far, so that was the most important place to fix.
Also added/fixed a few comments.
git-svn-id: svn://svn.code.sf.net/p/irrlicht/code/trunk@5793 dfc29bdd-3216-0410-991c-e03cc46cb475
CColladaMeshWriter::pathToURI tries to create now a correct name in xs::anyURI format (was more like xs::NCName before)
Note that the Collada loader in Irrlicht doesn't handle escaped characters yet, so right now can't load textures with space in names anymore written by Irrlicht.
That needs another fix in loader (it fails also when the .dae was written by other applications).
git-svn-id: svn://svn.code.sf.net/p/irrlicht/code/trunk@5789 dfc29bdd-3216-0410-991c-e03cc46cb475
Done similar as it was done in CWGLManager::activateContext, thought not tested as I currently have no test-case for this on X11.
git-svn-id: svn://svn.code.sf.net/p/irrlicht/code/trunk@5785 dfc29bdd-3216-0410-991c-e03cc46cb475
The reset was prevented before, likely to make the use in beginScene easier. But it's necessary for using an OGL context from another thread.
Only implemented for WGL so far, GLX implementation will follow soon.
git-svn-id: svn://svn.code.sf.net/p/irrlicht/code/trunk@5784 dfc29bdd-3216-0410-991c-e03cc46cb475
If it ever has same texture twice in it's cache that would be a bug and this slows texture removing down unnecessarily.
Also wasn't correct anyway as it would have missed it if there were 2 times in a row the same texture as the loop index wasn't adapted after the erase.
git-svn-id: svn://svn.code.sf.net/p/irrlicht/code/trunk@5778 dfc29bdd-3216-0410-991c-e03cc46cb475
Basically the destructor tried to call a function to drop() itself, but that certainly prevented the constructor from being called in the first place as there still was a reference.
No need to backport to 1.8, this was caused by some rewrites in Irrlicht 1.9 back when TextureCache for active materials got added to the driver.
git-svn-id: svn://svn.code.sf.net/p/irrlicht/code/trunk@5777 dfc29bdd-3216-0410-991c-e03cc46cb475
Last change (in r5766) fixed GL and messed up D3D. This needs to use a parameter to tell what the matrix does.
Thx @DevSH for reporting.
git-svn-id: svn://svn.code.sf.net/p/irrlicht/code/trunk@5776 dfc29bdd-3216-0410-991c-e03cc46cb475
Fixes wrong near-plane values with OpenGL (showed too much before).
git-svn-id: svn://svn.code.sf.net/p/irrlicht/code/trunk@5772 dfc29bdd-3216-0410-991c-e03cc46cb475
Note: It would likely be nicer if it wouldn't switch back to a projection matrix at all but stay with ortho-matrix. But until I get to that - this will at least return the correct flag.
git-svn-id: svn://svn.code.sf.net/p/irrlicht/code/trunk@5767 dfc29bdd-3216-0410-991c-e03cc46cb475
We also did just read that box and then dropped the info immediately, so it wasn't used anyway. Not sure what that code was about, I hope it was just accidental and not about support for some strange Collada files in the wild.
git-svn-id: svn://svn.code.sf.net/p/irrlicht/code/trunk@5759 dfc29bdd-3216-0410-991c-e03cc46cb475
It already worked correct when Z_UP was set (we switched x/z in that case which also flips coordinate system)
But for the default (Y_UP) it had been wrong before (that one is still right-handed in Collada).
We do now convert all z to -z (so z coordinates do change, but it prevents objects from showing up mirrored).
Note: For readLookAtNode I also added code to regard Z_UP, but have no test-case yet.
Note: I have no test-case for readSkewNode, so not sure about that one.
Note: No fix for readBboxNode as I'm going to kick that out next.
I tested with Blender (which has no textures) and Collada (which needs manual adaptions in .dae files as Irrlicht can't read libraries order independent, but that's another problem).
git-svn-id: svn://svn.code.sf.net/p/irrlicht/code/trunk@5758 dfc29bdd-3216-0410-991c-e03cc46cb475
Writing might be slightly slower, but has to write less on disk on the other hand.
So loading the files later on will be faster.
git-svn-id: svn://svn.code.sf.net/p/irrlicht/code/trunk@5721 dfc29bdd-3216-0410-991c-e03cc46cb475
Note: Right now our Collada loader can no longer load the files we write correctly as it also has to be fixed.
git-svn-id: svn://svn.code.sf.net/p/irrlicht/code/trunk@5709 dfc29bdd-3216-0410-991c-e03cc46cb475
(just making it easier to read to prepare for some other changes).
git-svn-id: svn://svn.code.sf.net/p/irrlicht/code/trunk@5703 dfc29bdd-3216-0410-991c-e03cc46cb475
(first one is probably correct once we fix the rest of coordinates to a right-handed system)
git-svn-id: svn://svn.code.sf.net/p/irrlicht/code/trunk@5700 dfc29bdd-3216-0410-991c-e03cc46cb475
A workaround for broken Collada importers which insist on specific names.
Seems SketchUp insists on UV's being called "S" "T" or it won't show textures.
git-svn-id: svn://svn.code.sf.net/p/irrlicht/code/trunk@5695 dfc29bdd-3216-0410-991c-e03cc46cb475
