Sound listener relativity bool inverted

[model76]

Hello Laurent,

When I was playing around with listener rotation back in May, I noticed that sf::Sound::setRelativeToListener seemed to be inverted. I didn't mention it back then because I had my head in the rotation thing, and I wanted to test it properly and provide minimal code to show the problem, if there was one.

Now I finally did, and there seems to indeed be a problem. Here is the code to show it:

#include <SFML/Graphics.hpp>
#include <SFML/Audio.hpp>


int main()
{
        sf::RenderWindow window( sf::VideoMode( 800, 600 ), "Relative to listener oddity" );
        window.setVerticalSyncEnabled( true );

        // Setup listener
        sf::Listener::setPosition( 0.f, 0.f, -1.f );
        sf::Listener::setDirection( 1.f, 0.f, 0.f );
       
        // Setup sound
        sf::SoundBuffer soundResource;
        soundResource.loadFromFile( "pling.wav" );
       
        sf::Sound soundEmitter( soundResource );
        soundEmitter.setPosition( 0.f, 0.f, 0.f );
        soundEmitter.setLoop( true );
        soundEmitter.play();

        // Listener orientation clock
        sf::Clock listenerOrientationClock;

        // Instruction text
        sf::Text instructionText( sf::String( "Press space to change listener relativity" ) );
        instructionText.setPosition( 0.f, 50.f );


        while( window.isOpen() )
        {
                // Handle events
                sf::Event event;
                while( window.pollEvent( event ) )
                {
                        if( event.type == sf::Event::Closed )
                                window.close();
                        else if( event.type == sf::Event::KeyPressed && event.key.code == sf::Keyboard::Space )
                                soundEmitter.setRelativeToListener( !soundEmitter.isRelativeToListener() );
                }

                if( listenerOrientationClock.getElapsedTime().asSeconds() > 1.f )
                {
                        sf::Listener::setDirection( -sf::Listener::getDirection().x, 0.f, 0.f );
                        listenerOrientationClock.restart();
                }

                // Draw scene
                window.clear();

                if( soundEmitter.isRelativeToListener() )
                        window.draw( sf::Text( sf::String( "Emitter is relative to listener" ) ) );
                else
                        window.draw( sf::Text( sf::String( "Emitter is not relative to listener" ) ) );

                window.draw( instructionText );

                window.display();
        }

        return EXIT_SUCCESS;
}
 

As you will notice, the sound will be affected by the listener's rotation when setRelativeToListener( false ), and vice versa.

[Laurent]

As you will notice, the sound will be affected by the listener's rotation when setRelativeToListener( false ), and vice versa.

Isn't it the expected result?

[model76]

Think about it:
If the sound has no relation with the listener, should it be affected by it? No!

[Laurent]

setRelativeToListener(false) doesn't mean "no relation with the listener". It means that the sound position is given globally, not relatively to the listener position/rotation. And if it's not relative, when the listener moves the sound doesn't, therefore you can hear the difference.

When the sound is relative to the listener, moving the listener will move the sound too so you hear no difference in your speakers.

To summarize:

sound.setRelativeToListener(false) -> the sound is fixed in the 3D world
sound.setRelativeToListener(true) -> the sound is attached to the listener, it follows it

The default is false. You would activate it for example for the sound of the gun that your player carries, because the gun follows the player.

I thought that the API doc was clear enough

[model76]

Ah, that makes perfect sense, actually. The documentation is also very clear about it, as is your explanation.

I just assumed my intuition was right on this, but assumption is the mother of all screw ups, I guess.
Thank you for clarifying.