v0.7.4b: Initial LibGDX commit! more details below:

Large sections of game logic are now working through libgdx instead of
android libraries. There is still work to do but this is the first
major step. Big changes include:
- Graphics code is now through LibGDX (except for text rendering)
- Initialization and screen-handling logic is now mostly through LibGDX
- Audio is now through LibGDX
- Input handling is now through LibGDX
- Most misc functions are now through LibGDX

SPD-classes/src/main/java/com/watabou/glwrap/Matrix.java

@@ -21,6 +21,8 @@
 
 package com.watabou.glwrap;
 
+//TODO LibGDX offer matrix classes as well, which might give better performance.
+//should investigate using them
 public class Matrix {
 
 	public static final float G2RAD = 0.01745329251994329576923690768489f;
@@ -91,7 +93,6 @@ public class Matrix {
 		m[5] *= y;
 		m[6] *= y;
 		m[7] *= y;
-	//	android.opengl.Matrix.scaleM( m, 0, x, y, 1 );
 	}
 	
 	public static void translate( float[] m, float x, float y ) {
@@ -100,6 +101,64 @@ public class Matrix {
 	}
 	
 	public static void multiply( float[] left, float right[], float[] result ) {
-		android.opengl.Matrix.multiplyMM( result, 0, left, 0, right, 0 );
+		final float ax1 = left[0];
+		final float ay1 = left[1];
+		final float az1 = left[2];
+		final float aw1 = left[3];
+		
+		final float ax2 = left[4];
+		final float ay2 = left[5];
+		final float az2 = left[6];
+		final float aw2 = left[7];
+		
+		final float ax3 = left[8];
+		final float ay3 = left[9];
+		final float az3 = left[10];
+		final float aw3 = left[11];
+		
+		final float ax4 = left[12];
+		final float ay4 = left[13];
+		final float az4 = left[14];
+		final float aw4 = left[15];
+		
+		final float bx1 = right[0];
+		final float by1 = right[1];
+		final float bz1 = right[2];
+		final float bw1 = right[3];
+		
+		final float bx2 = right[4];
+		final float by2 = right[5];
+		final float bz2 = right[6];
+		final float bw2 = right[7];
+		
+		final float bx3 = right[8];
+		final float by3 = right[9];
+		final float bz3 = right[10];
+		final float bw3 = right[11];
+		
+		final float bx4 = right[12];
+		final float by4 = right[13];
+		final float bz4 = right[14];
+		final float bw4 = right[15];
+		
+		result[0] = ax1 * bx1 + ax2 * by1 + ax3 * bz1 + ax4 * bw1;
+		result[1] = ay1 * bx1 + ay2 * by1 + ay3 * bz1 + ay4 * bw1;
+		result[2] = az1 * bx1 + az2 * by1 + az3 * bz1 + az4 * bw1;
+		result[3] = aw1 * bx1 + aw2 * by1 + aw3 * bz1 + aw4 * bw1;
+		
+		result[4] = ax1 * bx2 + ax2 * by2 + ax3 * bz2 + ax4 * bw2;
+		result[5] = ay1 * bx2 + ay2 * by2 + ay3 * bz2 + ay4 * bw2;
+		result[6] = az1 * bx2 + az2 * by2 + az3 * bz2 + az4 * bw2;
+		result[7] = aw1 * bx2 + aw2 * by2 + aw3 * bz2 + aw4 * bw2;
+		
+		result[8] = ax1 * bx3 + ax2 * by3 + ax3 * bz3 + ax4 * bw3;
+		result[9] = ay1 * bx3 + ay2 * by3 + ay3 * bz3 + ay4 * bw3;
+		result[10] = az1 * bx3 + az2 * by3 + az3 * bz3 + az4 * bw3;
+		result[11] = aw1 * bx3 + aw2 * by3 + aw3 * bz3 + aw4 * bw3;
+		
+		result[12] = ax1 * bx4 + ax2 * by4 + ax3 * bz4 + ax4 * bw4;
+		result[13] = ay1 * bx4 + ay2 * by4 + ay3 * bz4 + ay4 * bw4;
+		result[14] = az1 * bx4 + az2 * by4 + az3 * bz4 + az4 * bw4;
+		result[15] = aw1 * bx4 + aw2 * by4 + aw3 * bz4 + aw4 * bw4;
 	}
 }