Messages

1

General / Re: RenderManager, encapsulate sf::RenderWindow

« on: May 29, 2014, 05:44:49 pm »

If you build a wrapper that provides all the methods of sf::RenderWindow, then you're not really encapsulating it, so why not use it directly?

That was my worry. I'm going to rethink about that. If I have any questions I'll post them here.

There have been a lot of threads on separation between graphics and logics in this forum, you might be interested in them.

Yeah, that's a good idea. I'll be looking for it.

Edit: Just as an help for future viewers of this thread, I found this project on Nexus's git: https://github.com/Bromeon/TemporaryOutbreak

2

General / RenderManager, encapsulate sf::RenderWindow

« on: May 29, 2014, 05:02:18 pm »

Hi ! I'm currently thinking about a "RenderManager" for my game, it would encapsulate a sf::RenderWindow (with method to draw on it, etc, etc) and the video settings (whether the particles, vsync, fullscreen are enabled or not for example).

My question is: Am-I better to recode all access method of my sf::RenderWindow (draw, getView, setMaxFps, etc) and then when I'll draw my game I'll pass the RenderManager to the "object.draw(...)" method ? It's a lot of method to recode though, so I don't know if it's a good thing...

I'm also using TGUI for the controls and it needs the sf::RenderWindow for drawing and probably for other stuffs like cursor position on the window, so I suppose I'm better to code an access method to the sf::RenderWindow, then I'll be able to pass it directly to the tgui::gui ?

I have the felling my questions are a bit simple, but I want to be sure I'm doing it right.
If you have any suggestion, I'm open to it. Thanks !

3

Graphics / Use sf::Shape or my own classes for textured shapes

« on: May 23, 2014, 05:27:25 pm »

Hello !
I want to have a textured shape (circle and convex) and I wonder if I am better to use the derived types of sf::Shape or to implement my own classes (with vertices, yeah I know sf::Shape make use of vertices too).

I ask that because I read on another topic that there's advantages to use sf::Sprite instead of sf::RectangleShape (A little precision about that would be appreciated  ).

Thanks for your answers !

4

General / Re: Some questions about Scenegraph

« on: March 30, 2014, 10:35:42 pm »

Yeah, in this example it represent different character parts.

I think that I have all that I needed to continue. Thanks again, I'll mark my topic as solved

5

General / Re: Some questions about Scenegraph

« on: March 30, 2014, 09:48:49 pm »

Hello ! Thanks for your answer
I was very busy this week-end, so I didn't have the time to look back for the answer.

I had a doubt too on that way of doing it, but I was not sure about storing the instance that's responsible of the animation in the node.

I suppose that the best way to change the animation is to set the playing animation (animator.playAnimation(...) ) in the parent and apply it for every childs ?

6

General / [SOLVED] Some questions about Scenegraph

« on: March 28, 2014, 03:26:00 pm »

Hi ! I'm actually coding a little game and I have some problem with scenegraph implementation.
Here's my base Node class:

class Node: public sf::Drawable, public sf::Transformable
{

public:

    Node(const sf::Vector2f & origin, const sf::Vector2f & pos,
               const std::string & id, bool drawnAbove = true);

    virtual void Step(sf::Time frametime);

    //Add a new child to the node.
    bool BindNode(std::unique_ptr<Node> & node);
    //Remove a child and get it.
    std::unique_ptr<Node> UnbindNode(const std::string & id);

    //Get a child with it ID.
    Node * GetChild(const std::string & id);
    unsigned int GetChildCount() const;

    const std::string & GetId() const;

    void SetDrawnAboveParent(bool drawnAbove);
    bool IsDrawnAboveParent() const;

    virtual void OnDraw(sf::RenderTarget & target, const sf::Transform & trans) const;

protected:

    virtual void draw(sf::RenderTarget &target, sf::RenderStates States) const;

    //Childs.
    std::map<std::string, std::unique_ptr<Node> > m_childs;

    std::string m_id;

    bool m_drawnAboveParent;

};

I also implemented a spriteNode and an animationNode which have Node as base class. My question is:

If I use those classes to create an animated character (for example): Every part of the character would be a spriteNode which has as parent an animationNode (walk animation, attack animation, etc...).  How can I access the animation node and modify them effectively to the animation I want ?

Thanks for your answer !