Quote:
Originally Posted by cellaris
So what would be the real resolution, in your experience?
|
It's physics.
Well done sub pixel addressing gives better visual appearance than the colour resolution, but depends on content, but it can never get back the original resolution.
An older stripe filter (R G B) loses resolution in horizontal direction but has full resolution vertically. Because you are more sensitive to green than red or blue the sub-pixel addressing putting blue and red edges on the white surrounding black text makes the text much better.
The newer LCD (inherently mono like eink) and OLED and Kaleido 3 use a 2 x 2 patten. This means instead of 100dpi x 300 dpi for a 300 dpi panel, you have 150dpi x 150dpi, so colour resolution is better overall. However the sub pixel addressing for black text on white now isn't so advantageous. The perceived mono text quality will vary by viewer's eyesight and font. It's a scheme that suits coloured images best. But the underlying panel for eink does black and white, and only just manages 14 grey levels. Also the Kaleido is less dense and less saturated compared to Triton (conventional stripes like older LCDs) so as to make it brighter and better contrast. This desaturates the colours giving a washed out or pastel-look to the colour images, but slightly improved black text. It's simply dishonest to claim any colour panel technology using a mono panel has the original resolution in mono.
If it was true tt would be just as valid for 2 x 2 cell LCD (or QLED which is actually mono LCD, blue backlight and red + green quantum dots that's 1920 x 1080 colour to make the claim that those are 4K (3840 x 2160) in mono. They are not. Some use the same layout of R, G and B filtering as Kaleido. The dishonest claims for mono are only standing because it's a niche product with essentially a monopoly. It certainly can't compete with LCD, OLED, LED, QLED etc on colours/shades/hues (less than 4096 vs millions) or response time, or pixel addressing or contrast or greys (14 vs 254 to 1022).
If Kaleido 3 was on any mainstream product the claim that it's 300 dpi for mono would be axed. It's not physically or mathematically possible as soon as you have a colour filter of ANY kind of density and layout. Unless of course they had a 600 dpi mono panel under the filter.
This writes
Quote:
|
It takes advantage of the fact that each pixel on a color liquid-crystal display (LCD) or similar is composed of individual red, green, and blue components — subpixels
|
That "or similar" applies also to eink (but not Gallery ACEP), QLED (really LCD) and in theory to OLED and actual LED panels too.
https://en.wikipedia.org/wiki/Subpixel_rendering
The Gallery 3 panel* really has exactly same resolution for mono and colour, as does colour wheel DLP and three chip DLP projectors and colour cameras using three sensors and a splitter. Using the sub pixel addressing won't achieve the full resolution mono would have that same number of pixels if they were not R, G and B.
[* But ACEP / Gallery 3 is staggeringly slow at 1.5 s (1500 ms) for a full refresh and still worse rendition than LCD]
However at higher dpi usually sub-pixel addressing isn't used, so even apparent mono resolution = colour resolution. Grey anti-aliasing looks better above a certain resolution depending on viewing distance, and none at higher still. My desktop screen has a real colour dpi of just over 200 dpi, so I have it set to Grayscale instead of Subpixel smoothing.
Certainly if the eink panel was inherently 400dpi you'd not do any subpixel addressing. You'd have 200 dpi mono or colour, no matter image type or dark mode.
But at normal reading distance with correct reading glasses the 150 dpi is noticeably poor (Kindle DXG or other LCD screens I have).