Originally Posted by tommyv2
There are no subpixels in a 3-LCD display, there are only single-coloured pixels overlaid on each other, making white when they're all in a row.
Very good point. My bad. I have edited my post to reflect what is seen is the individual LCD panel element (red) which makes up each white pixel "point", not a "sub
pixel". Check. [I knew that actually, but must have been having a brain fart and worded it poorly. Oops.]
What you're seeing in that photo is poor convergence.
I personally disagree with that assessment of it being "poor", though. What I am seeing, and is shown in the photo, is such spectacularly good
convergence, AFAIC, such that neither vertically nor horizontally, even when examined so closely that individual pixel "points" are discernible, do the white lines show any
degree of red, green, or blue fringing, at all
[ignoring CA]. They are just white lines. The "poor" misconvergence, as you deem it, might be just
barely discernible if I had a test pattern with diagonal, instead of horizontal and vertical lines, I suppose [considering the individual white pixel "point" seems slightly red in its upper left], but for a straight out of the box projector with smooth screen and no
user adjustable panel alignment facilities, I've never seen better. If you have pictures of a superior convergence, under said conditions
, I'd be more than happy to examine it.
Considering the individual white lines are indeed just white lines, and not "clearly fringed with red, green or blue on one edge, at least when examined at extremely
close distances using a torturous test pattern of alternating, single pixel width, white and black lines", I'm quite
happy with my "poor" convergence.
Some purple fringing will come from the lens, but that's normal.
The CA might also be partly from the lens on my cheap point and shoot camera. It balks at extremely high contrast situations like this.