- Refactored paralax calculation code in Parallax.glsllib
- Added a boolean param to lighting material PackedNormalParallax to specify if the parallax map is stored in the alpha channel of the normal map (it was already implemented in the shader). added a dds file and a material using this
- The parallax height can now be set by users by setting the ParallaxHeight attribute. default is 0.05
- Deleted old normal map for water
- Inverted green channel of the brickwall normal texture to look good with recent change in normal calculation in lighting material
- Created a test case for parallax mapping where you can swich from classic to steep parallax using spacebar, and where you can tweak the parallax heigh by using I and K keys.
git-svn-id: https://jmonkeyengine.googlecode.com/svn/trunk@8046 75d07b2b-3a1a-0410-a2c5-0572b91ccdca
* Rewrote J3M loader
* AssetManager.unregisterLocator() must be implemented
* Added support for material default vars
* Apply NVIDIA spot light fix for TerrainLighting
git-svn-id: https://jmonkeyengine.googlecode.com/svn/trunk@8024 75d07b2b-3a1a-0410-a2c5-0572b91ccdca
* Fix bump mapping issue that was there since the beginning of time ... flipping normal.y only worked for pond_normal.png texture and IS NOT CORRECT in most cases. Now the texture itself has been adjusted to be correct while the flipping of normal.y removed
* Fix crash when playing sound in simpleInitApp()
* Fix issue where Spatial.setModelBound() would do absolutely nothing
* Skies created via SkyFactory now don't suffer from "premature culling" bug
git-svn-id: https://jmonkeyengine.googlecode.com/svn/trunk@8010 75d07b2b-3a1a-0410-a2c5-0572b91ccdca
* Enable mipmaps for packed and generated textures
* Better handling of material flags for 2.58
* Allow loading of meshes with no vertices instead of crashing
git-svn-id: https://jmonkeyengine.googlecode.com/svn/trunk@8005 75d07b2b-3a1a-0410-a2c5-0572b91ccdca
- Added Volume texture 3D loading support to the DDSLoader
- Added a test case for texture 3D loading
- fixes a minor log issue in Material.java
git-svn-id: https://jmonkeyengine.googlecode.com/svn/trunk@8003 75d07b2b-3a1a-0410-a2c5-0572b91ccdca
Added a variant of createPhysicsTestWorld with more spheres.
Changed the size of golem in TestWalkingCharacter to fit step length.
git-svn-id: https://jmonkeyengine.googlecode.com/svn/trunk@7983 75d07b2b-3a1a-0410-a2c5-0572b91ccdca
- computed the spotDirection in view space on the java side in Material and pass it as a uniform -> vec4 spotVec disapear
- this implies to decode the cos values for falloff in the frag shader instead of the vert shader, but allows to reduce light vec from vec4 to vec3
All in all it shouldn't be that much of a performance hit, since the matrix multiplication of spotVec occur once per geometry instead of once per vertex, we could go further and compute it once per frame.
On the other hand decoding of the cos values happens once for each pixel instead of once for each vertex ( a floor a mult and a fract). I guess this should be balanced in a rich scene.
git-svn-id: https://jmonkeyengine.googlecode.com/svn/trunk@7978 75d07b2b-3a1a-0410-a2c5-0572b91ccdca
plain single thread. This gives more control over the queuing
behavior as with an ThreadPoolExecutor it is difficult to implement
blocking execute() behavior. Also, this avoids creating an
extra object per-message.
Anyway, this code now implements a blocking queue instead of
a boundless queue. It's set to 16,000 messages which should only
bottom out in the worst of cases. I was seeing it in the throughput
tests where the sockets were backing up and the queues were
consuming all of the heap until an out of memory error occurred.
Outbound messaging is only throttled this way on the client.
Servers typically wouldn't do this sort of spamming anyway.
git-svn-id: https://jmonkeyengine.googlecode.com/svn/trunk@7977 75d07b2b-3a1a-0410-a2c5-0572b91ccdca
protocol. In the wild, I've only seen this crop up in
really really odd circumstances. As it turns out, the
throughput tests eventually trigger this when testing
TCP. I've lost sleep over wondering when this was going
to bite for real so I'm glad to have what is essentially
the last known bug in message transfer fixed.
This change handles the case where so far only one
byte of the two byte message size has been read from
the network. Rare, but it can happen.
git-svn-id: https://jmonkeyengine.googlecode.com/svn/trunk@7976 75d07b2b-3a1a-0410-a2c5-0572b91ccdca
Kae..pl
roo..om
rem..om
sha..rd
bre..ns
sky..ok
PSp..om
nor..67