ares1214 said:
Hello everybody, I'm asking about something I've wondered for a very long time. Pixels these days use square/rectangle shapes, 4 sided, something to that effect. What i was wondering is that if 2 Triangles make up a square/rectangle, or even 4 if you divide it once more and so on, why do we just keep on using squares/rectangles? If every pixel was cut into half on a 1920x1080 monitor, wouldn't it lead to a monitor with a resolution of 3840x2160, and therefore a 8.294 Megapixel screen, vs. the 2 Megapixel screens of today? I'm sure other things might be improved, however is there anything wrong with this? Even again, cut those triangles in half, and you get 7680x4320, at over 33 megapixels. Of course video cards would start crying, and 1920x1080 is already using up all the bandwidth TV's have to offer, but besides that, is there anything wrong with triangular pixels? Maybe even other shapes?
To get things clear, this is what i always thought to be true about colour display technology:
- pixel's are the square 2D shapes that are rendered on your screen output that you see, it is the output at the end of the process, your screen translates them in to a block composed of 1 red and/or 1 green and/or 1 blue (or none in case of black), assuming native screen resolution and video card output are the same.
- the virtual 3D shapes rendered by the graphics card are made from triangles.
- CRT monitors use circular dots of red green and blue to make a pixel, CRT TV's use rectangular r, g, b dots as do LCD's.
![]()
DLP projectors use single microscopic square mirrors, colours are made by a spinning colour wheel made of 3 colours and normally a white section also. So before the mirrors, the three colours are all created separately, but at a rate so fast our brain thinks its one colour. Its a concept where the harder you think about it, the more impossible it seems :S. Its basically an illusion, but it works.
Which is it you think could possibly be another shape? It makes sense that at the moment, the rgb colours should optimally be rectangular, but together make a square, to give best coverage of the monitor area, and give best performance for the square pixels that are rendered. they match to give a sharp image, so whatever the shape, the output of the graphics card and physical dot shape of the monitor should be the same. A hexagonal shape for example, i think would take more processing, although it would make an image appear smoother on screen. it may make the dot less noticeable as a visible square, but not quite a circle which would waste more screen real estate as they have no edges to align to. You also want less black area between the dots, so there is no contrast to tell there is shape there. So to change the technology, both the output of the virtual image, and physical shape of the display would have to change.