[Solved] Problem with creating textures using IntBuffer

[Matt Merkulov]

Please help me. I cannot understand what I doing wrong (result is white square instead of chessboard):

package examples;

import java.nio.ByteBuffer;
import java.nio.IntBuffer;
import java.util.logging.Level;
import java.util.logging.Logger;
import org.lwjgl.LWJGLException;
import org.lwjgl.opengl.Display;
import org.lwjgl.opengl.DisplayMode;
import static org.lwjgl.opengl.GL11.*;

public class Test  {
	public static void main(String[] argv) {
		try {
			Display.setDisplayMode( new DisplayMode( 800, 600 ) );
			Display.create();
		} catch ( LWJGLException ex ) {
			Logger.getLogger( Test.class.getName() ).log( Level.SEVERE, null, ex );
		}
		
		glShadeModel( GL_SMOOTH );
		glDisable( GL_DEPTH_TEST );
		glDisable( GL_LIGHTING ); 
		
		glEnable( GL_BLEND );
		glBlendFunc( GL_SRC_ALPHA, GL_ONE_MINUS_SRC_ALPHA );
		glEnable( GL_TEXTURE_2D );
		
		glMatrixMode( GL_PROJECTION ) ;
		glLoadIdentity();
		glOrtho( 0d, 800, 600, 0d, -1d, 1d );
		glMatrixMode( GL_MODELVIEW ) ;
		
		glClear( GL_COLOR_BUFFER_BIT | GL_DEPTH_BUFFER_BIT );
		glClearDepth(1);
		
		
		IntBuffer buffer = ByteBuffer.allocateDirect( 4 * 64 * 64 ).asIntBuffer();
		for( int y = 0; y < 64; y++ ) {
			for( int x = 0; x < 64; x++ ) {
				int color = 0xFFFFFF;
				if( ( y + x ) % 2 == 1 ) color = 0;
				buffer.put( x + y * 64, color | 0xFF000000 );
			}
		}
		
		
		int textureID = glGenTextures();
		glBindTexture( GL_TEXTURE_2D, textureID );
		glTexSubImage2D( GL_TEXTURE_2D,  0,  0,  0,  64, 64, GL_RGBA, GL_BYTE,  buffer );
		
		
		glBegin( GL_QUADS );
		
		glTexCoord2d( 0d, 0d );
		glVertex2d( 200, 100 );
		glTexCoord2d( 1d, 0d );
		glVertex2d( 600, 100 );
		glTexCoord2d( 1d, 1d );
		glVertex2d( 600, 500 );
		glTexCoord2d( 0d, 1d );
		glVertex2d( 200, 500 );
		
		glEnd();
		
		Display.update();
		
		while( !Display.isCloseRequested() ) Display.processMessages();
	}
}

[broumbroum]

Integers conversion is always a problem, because Java uses unsigned int's signed int's and the C code can have both signs and unsigned integers. so that the first significant bit is the sign bit as of the two's complement form. Similarly, I also avoid to use GL_BYTE and go with GL_UNSIGNED_BYTE,  for all textures.

[quew8]

Ie. use "0x000000" instead of "0." But don't do it anyway.

If you're interested, I have a Colour class which stores a colour as 4 floats but it has a nested (is it inner or nested? whichever one isn't static) class called ByteColour (guess what it does) which you can then use to directly put into a ByteBuffer and use with GL_UNSIGNED_BYTE. It is pretty simple but as an example:

public class Colour {
    private float red;
    private float green;
    private float blue;
    private float alpha;
    
    @Override
    public String toString() {
        return "Colour:" + red + ", " + green + ", " + blue + ", " + alpha;
    }

    public class ByteColour {
        private byte red, green, blue, alpha;
        public ByteColour() {
            this.red = (byte)((int)Colour.this.red * 255);
            this.green = (byte)((int)Colour.this.green * 255);
            this.blue = (byte)((int)Colour.this.blue * 255);
            this.alpha = (byte)((int)Colour.this.alpha * 255);
        }

        @Override
        public String toString() {
            return "Byte Colour:" + (red&0xFF) + ", " + (green&0xFF) + ", " + (blue&0xFF) + ", " + (alpha&0xFF);
        }
    }

(Selective Quoting)

Edit: It's "you're" not "your"

[broumbroum]

Ie. use "0x000000" instead of "0." But don't do it anyway.

public class Colour {
  ...
    public class ByteColour {
        private byte red, green, blue, alpha;
        public ByteColour() {
            this.red = (byte)((int)Colour.this.red * 255);
            this.green = (byte)((int)Colour.this.green * 255);
            this.blue = (byte)((int)Colour.this.blue * 255);
            this.alpha = (byte)((int)Colour.this.alpha * 255);
        }
    }

...

Hi, I think you might add a debugging output here, that's worth the clue.
1.Because you cast float to an (int), where the float is always 0...1 for colors, your values get stuck onto the 1*255 .. (which turns out into a full brightness
2.Okay so let's choose one format for clarifying this. The easier way is to forget a bit about the bytes and use only floats or int, but not mixing, it's confusing !
3.The way you want to get the conversion from a nested class is an immutable pattern . If the Colour wants to change over time, bytecolour should as well. Basically a casual function like Colour.this.toBytes() is lighter.
4. uhh, no four.. ^^
cu

edit : I'm sorry I thought about answering to Matt, but it was quew8 !!! Surely forget about the bytes..

[quew8]

Last I checked, this worked. I'm pretty sure that (red * 255) is what is cast to int rather than red itself. (I know there aren't brackets but I think that's how it works). I suppose it could be that I have never tried it with decimal values for the RGBAs (I haven't used this in a very long time) but I would say that it is unlikely and am fairly sure I remember using this with a lovely tone of green.

You quite right about the whole nested class vs METHOD (since we're being technical (it could be a function but that's probably bad oop) ) however I did say it was simple (simple minded rather than simple implementation) and it suited my then purposes well. Basically I wanted to be able to manipulate the colour easily which for me means floats then the only time I needed the byte values was to stick it into a texture so I passed the Colour into the texture loading function and created a ByteColour instance in the function.

To conclude, you're right, but I'm not going to listen to you. Although you have got me worried about that whole casting issue so I'm going to check up on that when I have time. I'm actually horrified at myself to find I didn't put clarifying brackets in in the first place. It was a VERY long time ago.

[Matt Merkulov]

Found solution in Slick-util code. I should set texture filters:

glTexParameteri( GL_TEXTURE_2D, GL_TEXTURE_MIN_FILTER, GL_LINEAR ); 
glTexParameteri( GL_TEXTURE_2D, GL_TEXTURE_MAG_FILTER, GL_LINEAR );