* Update GLImageFormats.java
* First implementation to solve the glTexSubImage issue
* Setting proper format to the image
* Removed the Image.convertToFormat method and its usage from GLRenderer.modifyTexture. Added a warning instead for the potential failing cases
Fixing the issue in JmeBatchRenderBackend.loadImage transforming it to RGBA8 if required
because user code should never throw NullPointerException.
Also made trying to add a child to itself an error instead of a no-op. Attempting
to do something like guiNode.attachChild(guiNode) will now throw an IllegalArgumentException.
a user later chooses to call updatePoints() after having created the line with
JME constants... since updatePoints() would actually call set() instead of just
replacing the references. The constructor and updatePoints() should match and I chose
to err on the side of caution and make them both operate on local instances.
to see if they've changed in the current frame. SimpleApplication will configure
this by default to use asserts but the application is free to change it to throw
regular exceptions or just log the error as desired.
* Update GLImageFormats.java
* Added basic support for openGL ES 3.0 and GLSL300 for android
Added instancing support for android which is core feature in GLES3.0
Added fix for shadows in android
* Fixed required types of GLES30.glDrawElementsInstanced
* Properly check GLES3 version and setting precision for shaders for all GLSL ES versions
* Added support for Opengl ES 3.1 and 3.2 and their shader versions
Added proper checkings for new OpenGL ES versions
Fixed some issues with shadow rendering in GLES 3.0 or better (strict type checking, OpenGL ES specific extensions and percision definition)
Added GLSL320 and GLSL300 to Unshaded and Lighting materials
* Added depth texture 24bit for OpenGL ES
* Added geometry shaders and tessellation to GLES 3.1 as extension and GLES 3.2 as core
Modified test materials to use GLSL310
* Added FB blit for android GLES30
* Added GlSL300 or better to all effects not able to run in GLSL100
Mods on Shadows.glsllib to try to fix android shadow rendering
Fixed ssaoBlur.frag that was using a reserved word
Fixed precision type missmatch from vertex to fragment shader
* Partial multisample support
Framebuffer MRT support
* Temporarily removed texture multisampling (not being able to compile using GLES31)
* Updated android.jar to api28
Enabled multisampling
* Better checking for GLES3
* Removed insert precision for all shaders. This avoided some random precision missmatch error but created rendering issues
* Removed border check for GLES, caused filtered shadows not to render on GLES2.0 devices
Added texture compare mode for GLES3.0 and better to be able to use sampler2DShadow
Modified base materials (Lighting and Unshaded) to use only GLSL100 for post shadow pass
* Corrected texture comparison function that were incorrectly changed
* Added precision to samplers for GLESSL300 also
* Created a subclass of Glsl150ShaderGenerator (Glsl300ShaderGenerator) to support ShaderNodes in GLES SL30 and included it's usage in DesktopAssetManager. Nowadays just copied and overwrote the version string. Needs a full review
Added image format RGB16F to be used as texture format only in GLES30 (cannot be used as FB format)
Added GLSL300 and/or GLSL310 to different materials
* Cleaned GLSL300 generator
Fixed shadows glsl lib removing use of GL_EXT_gpu_shaders5
* Fixed usag of textureCubeLod in fragment shader of envMapping which is unsupported in GLES2.0
Modified PBRLighting frag to properly compile on GLES3
* Added GLSL300 and 310 to PostShadowFilter material definition
* Make post shadow filter use GLSL100 for GLSL300
Set highp as default precision to fix glitched shadows caused because not enough precision
* Changed water_normalmap.dds to uncompressed RGB8 because DXT1 is mostly unsupported in GLES
* Fixed GLES30 compilation issues in Water shaders and added GLSL300 to LightScattering material. ** Not fully tested
* Added RGB8 dds formatted FullskiesSunset to be used in any platform. Previous file moved to _dtx1.dds
* Set GL_TEXTURE_MAX_LEVEL in GLES30 contexts fixing issue with not fully mipmapped images and PBR rendering
* Fixed compilation of jme3-android-examples and upgraded sdk version to use
* Fixed Geometry Shader and Tessellation Shaders in testdata package to properly compile on GLES320
* Added non-compressed textures to be able to run all examples on android
Enabled texture arrays and 3d textures on opengl es 3.0. Peding to implement wrappers to android gl functions (currently getting NullPointerException)
* More GLES30 functions included to properly support texture arrays, 3d textures and multisampling matching jme core code
* Fixed 3D texture tests for GLES30
Fixed texture 3D and texture array example shader compilation on GLES30
Fixed some random missing precision compilation errors on GLES30
* Fixed TestTextureArray for GLES30
* Some multisample additions and removal of GLES3 multisampling enable caps
Added RGBA16F as valid format for GLES32 and using it by default in env camera instead of RGB16F. kept fall back to RGB8
* added float and half float formats for gles30
added rgb10a2 format for gles30
fixed usage of glTexImage2D
removed EnvironmentCamera fallback image format as it's useless now
* Added more image formats for GLES30
Fixed rgb8 if having Caps.Rgba8
* Enabled RadialBlur for gles30
* Fixed luminance texture formats for gles
* Added more depth image formats for gles3
* Enabled multisampling in gles3
* Added GLES30 functions to GLDebugES
Reverted TestTexture3D and TestTextureArray to use RGB8 format
Updated EnvMapping100.frag to use EXT_shader_texture_lod in gles sl 100 if available
* Removed all aditional not used compressed dds files
Removed also Pond and rock png files previously used in TestTextureArray
* Removed compressed water_normalmap_dxt1.dds
* Changed GLES_30 and AndroidGL implementing GL2 to avoid duplicated code checking GL2 vs GLES30
* Added aditional checking to avoid gles30 calls from gles20 only devices
* address issue #1119 (serialization with protected/private constructor)
* remove an unnecessary step in findNoArgConstructor()
* use getDeclaredConstructor() in place of the for-loop
* simplify by throwing the exception in findNoArgContructor()
(Specifically the stuff that deals with IDs.)
Also added the ability to use groovy tests for those who like a little
testing but aren't complete masochists.
for retrieving an AppState by ID. Also modified attach() to throw an
IllegalArgumentException if the ID is already registered.
Updated TestAppStateLifeCycle to add a small ID demonstration.
This is a breaking change for direct implementers of AppState as they must
add a getId() method. (Which is also a good time to evaluate if they should
be extending AbstractAppState or BaseAppState and/or if their life cycle is
implemented correctly. e.g.: BulletAppState is breaking the AppState contract
a bit.)
* Add support for gltf morph names.
* Fix formatting stuff
* Add morph name to be saved
* Testing changes to gltf plugin
* Review changes
* Fix comments
* Fixes for review
* Remove getMorphNames from Geometry class
* jme-core: IndexBuffer fixes & API improvements
As discussed in:
https://hub.jmonkeyengine.org/t/indexbuffer-suggestions-improvements-consistency/42022/4
* [Fix] createIndexBuffer(int, int): return IndexByteBuffer as well for
vertexCount < 128
* [Enhancement] Make put(int i, int value) fluid interface
* [Enhancement] Add relative put(int i) to allow easier chaining (parity
with other buffer implementations)
* [Enhancement] Add getFormat() to allow setting an IndexBuffer to a
Mesh directly without type inspection.
* [Fix] Fix WrappedIndexBuffer
API changes make it possible that IndexByteBuffer is now a valid type
for outBuf, leverage the new getFormat() method to set the buffer to the
mesh regardless of its type.
* Update VirtualIndexBuffer to @72f8019566fa4d1379caa820c0c9dc000f489444
* #1121-updateBound() call for PQTorus geometry updates
* World bounds test added for #1121
* World bounds test enhancements for #1121
* Updated test comment
Previously if particlesPerSec was constantly set timeDifference could never grow large enough to emit a particle as timeDifference was reset to zero each time. This fixes that problem while still ensuring that timeDifference can't have grown very very large (leading to huge numbers of particles being emitted) by capping the timeDifference at the point where its just about to emit a particle.
Discussed at https://hub.jmonkeyengine.org/t/particle-emitter-cannot-emit-particles-if-its-emissions-per-second-is-updated-every-frame/41930/2
* Attempt to fixing 1074
* Added method implementations in GlRenderer
* Added test.
* Added test to jme-examples
* Some changes to resolve merge conficts
* Fixes
* Indent-fixes
* Documented the class and added stop()
* Fixed a minor error
* Fixed build gradles
* Removed line ending
* AnimComposer: save globalSpeed on export
* Updated TestModelExportingCloning to test AnimComposer.globalSpeed
* Fixed issue with AnimComposer.Layer cloning
* returns an unmodifiable set when retrieving the available animation names and available animation clips instead of an unmodifiable collection. fixes: #1070
* remove unnecessary set wrapper.
* Remove the unnecessary creation of a new hashset everytime getAnimClips() is called.
Riccardo Balbo
joliver82
Stephen Gold
Paul Speed
Matthew Universe
Riccardo Balbo
Ali-RS
Ali-RS
Trevor Flynn
Sebastian Teumert
Remy
Lou H
richardTingle
Karan Nehra
James Khan