E-Ink's Gallery 3: 300dpi

[kyteflyer]

https://www.eink.com/news.html?type=...ar=2022&page=1

I'm not sure how this directly relates to e-readers, as they speak of paper... can someone explain?

[kyôdai]

Hey there,

E-Ink is adding yet another color display to their portfolio.
Yesterday, they announced the "Gallery 3 Color", based on their ACeP technology, will join the recently updated "Gallery Plus Color".

While the "Plus" version, like its predecessor, is only geared for commercial signage and public information displays, the "Gallery 3 Color" is intended for use in ebook-readers.

Here you can find the press release:

-> https://www.eink.com/news.html?type=...ar=2022&page=1

E-Ink says the new display has a 300ppi color resolution, showing upto 50.000 colors.

They have significantly improved the refresh rate of the display compared to the first generation (where a full screen refresh could take upto 10 secends).

The new display has multiple refresh modes. It takes 350ms in black/white mode, 500ms in "fast color" mode, 750-1000ms in "standard color" mode and 1500ms in "best color" mode.

[pdurrant]

This looks really interesting. B&W refresh at 300ms is very good. A maximum refresh for best colour of 1.5s is acceptable, given that there's a slightly lower quality 0.5s refresh rate for colour.

And 300dpi colour... it's a big improvement on the colour displays that use a colour filter in front of a B&W display.

I would be very tempted if a Kindle came out using this display, even if just for the book covers!

[jhowell]

Originally Posted by pdurrant

I would be very tempted if a Kindle came out using this display, even if just for the book covers!

I would be tempted also, but I doubt that the added cost would be worth it for me to actually go ahead and make the purchase.

[Pajamaman]

https://www.eink.com/news.html?type=...ar=2022&page=1

They have examples of foldable, bendable color e-ink panels. I can see this stuff revolutionizing panels, TVs and reading material in the future.

[kyôdai]

E-Ink released another new video today.
They are demonstrating all their new color displays at their booth at this years "Touch Taiwan 2022" exhibition.

[Quoth]

I always wrote that colour epaper needed a different technology to R G B filters on a mono eink screen, which is inherently poor due to intrinsic physics and maths. I suggested a four layer C Y M K (Cyan, Yellow, Magenta. K= Black, but can mean white with black/grey) was the solution. Early colour photography used R G B dots but was only viable years later using C Y M layers.

It will be interesting to see how it really works as each layer has to be transparent or coloured or somewhere between. Traditional eink isn't transparent.

It's pretty slow so won't do video/tv/animation.

It will be interesting to see how bright the whites are in ambient light compared to regular eink, contrast ratio, how many levels of grey and saturation and gamut.
In theory the lighting layer can be below the three C Y M layers and above the "white".

An issue is viewing angle. This is limited by the thickness of each layer vs size of the pixels.

Also I think there isn't a huge point to much less than a 10" screen if you look at size of colour illustrated books, magazines, graphic novels and comic. Most of my 3000+ books are monochrome and most just text. Most of the coloured books I have are quite large; larger than 10" diagonal

The price and battery life will be important. Are the C Y M cells bistable like regular eink?

[pdurrant]

There are no layers - all the colours are in the sample capsule. There's some clever tweaking of electrical properties that allows one of the colours to be displayed in preference to the others, or white. Yes, they're bistable (or quad-stable) like regular B&W e-ink.

eInk mention 300dpi B&W and 150dpi colour, so I think they're not able to mix colours within a capsule, so full colour needs to be a sub-cell of four dots. But with much better contrast and saturation than from colour filters in front of B&W cells.

[Quoth]

R G B can only use parallel sub cells. Additive colour model.
C Y M can ONLY use layered sub cells. Subtractive colour model.
It's basic physics.
Additive mixing (sub pixels in parallel). Dots or cells are side by side.
G + B = C
R + G = Y
R + B = M
R + G + B = W
Vary proportions for other hues and saturations.

Subtractive Colour Mixing uses cells or pixels in layers.
C blocks Red
Y blocks Blue
M blocks Green
Thus Layers
White - C - Y = Green
White - C - M = Blue
White - Y - M = Red
White - Y - C - M = Black. Because print pigments are not perfect black is also used or you get a dark muddy colour.

Using C Y M print the pigments or transparent cells are superimposed.
R G B needs the dots side by side.

Projectors & cameras currently use either R, G & B added in a prism, OR red, green & blue filter or spinning R G B clear wheel on monochrome. Some displays might use also yellow for brightness. Some cameras and displays don't stripe R G B but use a Bayer pattern.

You could use a single cell with mix of multiple Cyan, Yellow and Magenta balls if there was a way of separating them. It was a lie-for-children in art class that you used red, yellow and blue primary paints. The red was closer to magenta and the blue closer to cyan or you couldn't get very many colours by mixing.
It's not impossible that they have cyan, yellow and magenta in one cell and mix like paint. I can't imagine how that's done compared to three thin layers of cyan, yellow & magenta balls in a clear liquid that can be move to sides of cell or face of cell.

No doubt eInk Corp will explain more later.

[pdurrant]

A fair point about the difference between additive and subtractive. But I think you're missing something.

Cyan reflects Green and Blue light.
Magenta reflects Red and Blue light.

Red, Green and Blue light makes white.

So putting dots of Cyan and Magenta next to each other will give Red. Similarly for other combinations. So I think it is theoretically possible to do a display using CMY pigments with side-by-side cells, not layered.

We'll have to wait for people to examine the display in close-up to find out for sure.

[jhowell]

Originally Posted by pdurrant

Cyan reflects Green and Blue light.
Magenta reflects Red and Blue light.

Red, Green and Blue light makes white.

So putting dots of Cyan and Magenta next to each other will give Red.

Could you explain this?

My understanding is that putting cyan and magenta on top of each other will produce saturated blue since red and green light are absorbed by one of the pigments. Putting them next to each other gives unsaturated blue because some red and green light is reflected along with the blue.

[Quoth]

So putting (reflective) dots of Cyan and Magenta next to each other will give Red.

No, it gives a very desaturated blue.
You can't get saturated primary colours that way.

We'll have to wait for people to examine the display in close-up to find out for sure.

Yes. A bit pointless to argue. However R G B filtering of reflected monochrome (filter on regular eink) is at best 1/6th brightness unless pastel filter and/or clear pixels too (which also desaturates).
This was known over 120 years ago (early colour magic lantern slides by photography) and Cyan, Yellow, Magenta layers, where the layer can be anywhere from transparent to a complementary colour (Subtractive) was a major breakthrough and why colour prints, slides and filmed movies worked.
Phosphor (CRT), Plasma, electroluminescent, LED, DLP and OLED can only be a single layer so has to use additive R G B.
LCD can be multiple layers, but it works by polarisation, so though the crystal could have a colour, the mode coloured to black, not the coloured to transparent, so all multicolour LCDs actually are monochrome clear to black with filters. An LCD that could be clear or coloured doesn't yet exist, unless that's what eink Corp have discovered.

Mirasol used diffraction. It's fast but the contrast and colour isn't great. Two Asian brands used the same electronics and panel. It may have died due to Qualcomm's infamously awkward royalty schemes.

There are colour cells that squeeze the pigment in and out of tear or comma shaped cell, but too slow and impracticable. Never any commercial product.

I've always said (since about 1980) C Y M in layers with transition to clear would be the ultimate colour display, and suitable for posters, adverts and reading if too slow for video. Side by side C Y M would be nearly worthless. The photos of screens don't look as desaturated as that would be.

[Quoth]

Originally Posted by jhowell

Could you explain this?

My understanding is that putting cyan and magenta on top of each other will produce saturated blue since red and green light are absorbed by one of the pigments. Putting them next to each other gives unsaturated blue because some red and green light is reflected along with the blue.

Absolutely correct. A pale blue. Nearly worthless. Pastel R G B filter on monochrome would be as bright and give a better colour range.

[pdurrant]

Originally Posted by jhowell

Could you explain this?

My understanding is that putting cyan and magenta on top of each other will produce saturated blue since read and green light are absorbed by one of the pigments. Putting them next to each other gives unsaturated blue because some red and green light is reflected along with the blue.

My silly mistake. It gives blue, of course. (1 x red, 1 x green 2 x blue == blue with white)

[kyôdai]

E-Ink explains their ACeP technology as such:

-> https://www.eink.com/color-technology.html

"Advanced Color ePaper (ACeP™) is a high quality, full color reflective display. In the ACeP™ system, the ink can produce full color at every pixel, without the use of a color filter array (CFA).

ACeP™ achieves a full color gamut, including all eight primary colors, using only colored pigments. The display utilizes a single layer of electrophoretic fluid which is controlled using voltages compatible with commercial TFT backplanes. The fluid can be incorporated into either microcapsule or Microcup® structures. The richness of the colors is achieved by having all the colored pigments in every picture element (pixel) rather than the side-by-side pixel colors achieved with a CFA. This eliminates the light attenuation, which can be quite significant. ACeP™ maintains the ultra-low-power and paper-like readability under all lighting conditions of regular E Ink ePaper."

The attached image is from an editorial about all the different color display technologies from E-Ink. It illustrates how the ACeP screen displays colors.

-> https://spectrum.ieee.org/how-e-ink-...l-color-epaper