I'm working on a cocoa wrapper for a graphics framework.
To finally draw the stuff, I'm doing this:
- (void)drawRect:(NSRect)rect
{
CGContextRef ctx = (CGContextRef)[[NSGraphicsContext currentContext] graphicsPort];
CGContextDrawImage(ctx, NSRectToCGRect(rect), image);
}
inside a subclass of NSView
.
Now I've looked at other frameworks like Gosu, Irrlicht, etc. and I saw they were always doing complicated NSOpenGL
stuff like:
// Settings, depending on fullscreen or not
NSOpenGLPixelFormatAttribute windowedAttrs[] =
{
NSOpenGLPFADoubleBuffer,
NSOpenGLPFAScreenMask,
(NSOpenGLPixelFormatAttribute)CGDisplayIDToOpenGLDisplayMask(CGMainDisplayID()),
NSOpenGLPFADepthSize,
(NSOpenGLPixelFormatAttribute)16,
(NSOpenGLPixelFormatAttribute)0
};
NSOpenGLPixelFormatAttribute fullscreenAttrs[] =
{
NSOpenGLPFADoubleBuffer,
NSOpenGLPFAScreenMask,
(NSOpenGLPixelFormatAttribute)CGDisplayIDToOpenGLDisplayMask(CGMainDisplayID()),
NSOpenGLPFAFullScreen,
NSOpenGLPFADepthSize,
(NSOpenGLPixelFormatAttribute)16,
(NSOpenGLPixelFormatAttribute)0
};
NSOpenGLPixelFormatAttribute* attrs = fullscreen ? fullscreenAttrs : windowedAttrs;
// Create pixel format and OpenGL context
ObjRef<NSOpenGLPixelFormat> fmt(
[[NSOpenGLPixelFormat alloc] initWithAttributes:attrs]);
::context = [[NSOpenGLContext alloc] initWithFormat: fmt.obj() shareContext:nil];
Why are they doing all that? Is my "simple" way okay?
I agree with ryyst. You are using CGContext which is essentially the Quartz 2D API. OpenGL is another option for graphics especially good for complex 3D rendering.