I'm curious to know if there are specific optimizations that depend on the pixel layout of a screen for how an RGB grid is actually mapped onto these different screen types. Is that the responsibility of the display driver to worry about those details or does it help to actually consider the pixel layout when doing rasterization?
ToiletCommander
Ive heard that there is a specification in screen that is called "effective resolution". And Samsung’s displays are actually patented because it gives a bigger "effective resolution"
Staffjamesfong1
@austinapatel There are! These optimizations are referred to as "subpixel rendering" techniques. These methods are commonly used to draw extremely fine details, such as lines in text.
In order to implement subpixel rendering properly, both the display driver and the rasterizer need to cooperate. The display driver needs to let the rasterizer know what the subpixel layout is, and the rasterizer needs to know how to exploit that layout. This added complexity is why most applications just make the same "colored square full-color pixel" assumption we make in this class.
I'm curious to know if there are specific optimizations that depend on the pixel layout of a screen for how an RGB grid is actually mapped onto these different screen types. Is that the responsibility of the display driver to worry about those details or does it help to actually consider the pixel layout when doing rasterization?
Ive heard that there is a specification in screen that is called "effective resolution". And Samsung’s displays are actually patented because it gives a bigger "effective resolution"
@austinapatel There are! These optimizations are referred to as "subpixel rendering" techniques. These methods are commonly used to draw extremely fine details, such as lines in text.
In order to implement subpixel rendering properly, both the display driver and the rasterizer need to cooperate. The display driver needs to let the rasterizer know what the subpixel layout is, and the rasterizer needs to know how to exploit that layout. This added complexity is why most applications just make the same "colored square full-color pixel" assumption we make in this class.