52 comments

[ 4.4 ms ] story [ 119 ms ] thread
I don't fully understand from reading the article why this is better than EInk. Seems it just uses 1/4 of the energy but other than that is there more benefits to it?
I wouldn't downplay that, it would have a great effect on battery life.
True, it would be interesting to know how much this would improve a typical ereaders battery life considering they are very good already.
But wouldn’t eink already be a huge win anywhere battery life is concerned? Having my ereader last for 3 weeks instead of 2 weeks isn’t really all that compelling
Competition. Eink seems to have stagnated.
My understanding is that E-Ink holds all of the patents and is preventing any further competition or innovation.
That's my understanding too - if someone can create a screen that competes on energy use, then they occupy the same market and will hopefully force E-Ink to innovate/bring down prices.
Worse, it's like they are developing something but not releasing it intentionally.

Even the fast repaint algorithms are proprietary and not licensed to everyone. My Kobo Glo HD is advertising 50% faster page repaint with the new software. Amazon Kindle is using another algorithm, which is way different.

Worse, the algorithms (called waveforms) are just pieces of software which are CPU architecture specific. They even don't port these algorithms between CPU architectures.

Time for an open source implementation?
All of the stuff is possibly patented. I'm not sure how you can create an open source implementation of something patented and protected so well.

It's like the present situation of the exFAT.

Dasung seems to be pushing the envelope -- they claim (and have video as proof) their upcoming Not-eReader has a much better refresh rate than your usual eReader. I am really excited about that device -- not because of the refresh but because it's a full Android tablet and a HDMI monitor. I will use it as a secondary screen for the One Mix Yoga 2S. How handy the first Core CPU, fast (aka non-eMMC) storage 7" laptop to have a HDMI output... I always have an eReader with me because I like to read on transit, so whenever oncall hits (oncall incidents are months apart, so it's not too bad being on call) I can just pull out the tiny laptop, plug the monitor in and I have pretty decent screen real estate for a bit of emergency debugging.
I'm not sure I am reading the article correctly, but it looks to me that this new display tech uses less energy than EInk for the initial image creation, but then requires quite a bit more (0.25mJ/cm2 is greater than 0mJ/cm2, though still a minuscule amount) for continued display.

I would guess that they probably are able to get a crisper image or perhaps full color with this new display tech while still being optimized for low-energy pull.

Edit: Someone else posted this link which shows the Rdot displays do in fact have color capability - https://rdotdisplays.com/technology.

In terms of your comment that it requires a little bit more energy for continued display, this isn't really the full story. The article states that this small update is only required once every 15 minutes, but my guess is the vast majority of E-Ink applications have use cases where the screen is updated more frequently anyway (e.g. for e-readers you are rarely going to spend more than 15 mins on a page).

It is relevant for use cases like persistent artwork or ambient always-on displays, but I think for most of the common e-paper use cases it's a moot point.

I had not considered that. I definitely "turn the page" much more than once every 15 minutes when I am reading on my ereader. When you take that into consideration, these Rdot screens are roughly 4 times as efficient AND have color as well.
I couldn't tell from the link but is it only tunable once to a certain chosen colour at manufacture time? Afaik EInk can achieve 3 states (eg, black white and red)
It's possible to do full-color e-ink in multiple ways, even if releases have been exceptionally slow.
They way they phrase it it looks like you can choose the color for all the pixels (i.e. a black/white or green/white or red/white display).

I hope I'm just misreading it.

They could keep the same battery life and simply have a smaller battery.
Because it isn't eink. Eink has a monopoly and charges accordingly. Any competition would be a good thing.
How is this different from the Sharp Memory LCDs?
My calculations might be wrong, but it looks like these Rdot displays use significantly less energy than Memory LCDs.

Using numbers from Sharp[0], Memory LCDs use ~226µJ/cm2 each second when static. I'm assuming they are 60Hz displays & therefore require a refresh 60 times each second but even if they only refresh once a second, that's ~3.8µJ/cm2 a second or 13,680 times the power consumption of the Rdot displays.

The article claims these Rdot screens use 0.25mJ/cm2 every 15 minutes when static.

Again, these are my "back-of a napkin" calculations so take with some salt.

[0]https://www.sharpsma.com/products?sharpCategory=Memory%20LCD

Calculating out of my Kiester assuming 2.5cm2 display 3.0V it seems

Sharp Memory LCD is 200uA vs rdot at 0.25ua.

With AA batteries Sharp Memory is only going to last 1 year MAX. That's too short for most consumer applications. Where rdot is much less than the self discharge. So 5-10 years.

Good. E-ink displays are expensive and have a monopoly on low-energy persistent displays.

However, Rdot doesn't have nearly the same resolution as E-ink at this time, it seems: https://rdotdisplays.com/technology

Their FAQ says they are planning on offering matrix displays next year (https://rdotdisplays.com/faq):

> We expect to introduce an entirely printed passive matrix version in 2019.

It will definitely help to have some competition in this space.

Well I'll be damned. I thought kidles were cheap but you can actually get a Kindle 7" fire tablet which does a lot more for 50$ and a Kindle 6" eink reader is 70$.

So yeah, I'd have to agree with you, some competition would be nice

I wouldn't recommend any of the current generation kindles. I've been gifted every single Kindle there is, and I have to say that $50 is not worth it unless you really just want a throwaway tablet as an alarm clock or something. It's cheap for the following reasons:

1) you're stuck inside the shitty Amazon App store ecosystem. Lots of developers hate that Amazon created their own separate app store, so expect no major apps on the download section of your kindle's app store. Especially not Google apps like YouTube. Just the lowest quality of low effort apps made by foreign developers. 2) screen and speaker quality sucks now compared to previous kindles. The UI and user experience has just always been bad because you don't get anything resembling stock Android OS. You have to use whatever weird skin that Amazon decided to tack on to their weird and limited version of Android. 3) the tablet's price is subsided by ads. It's cheap because you have to see a different ad every time you unlock the Kindle screen.

The kindle's speakers used to sound amazing on the old HD, I forgot what it's called but it was the best tablet Amazon ever made. Every Kindle ever since has been significantly worse.

Id you want a tablet, find some $80-$100 Chinese tablet from somewhere that can at least doesn't blast you with ads. It's worth it.

I was just today looking at E-inks new 'JustWrite' technology where the screen pixels are activated by a stylus instead of the backing, so writing feels instant (and might be just about instant too). If this tech also has that possibility, then good on them. Edit: though I'm not aware of a product which incorporates it yet, which is a shame. Maybe I'm wrong - research on mobile is hard.

What I'd love is a notebook of the stuff that I could write on and save and swap the contents with an open protocol/app.

I'd also probably love hacking on the stuff. Make a 'spellbook' for one of my D&D characters where the contents change 'magically'. :)

> though I'm not aware of a product which incorporates it yet

This is exactly why E-Ink (the company) technology feels like vaporware to me.

I wonder how fast they change? The time to rewrite the screen is a major bummer for me when compared to flipping pages with dead-tree paper
Their FAQ says update speeds range from 50–500ms.
I used to think this too, until I videoed myself reading to test.

I discovered that dead-tree page flips take between 1.8 to 4.2 seconds, pretty much normally distributed. Granted we're doing more page flips on eink because the screen is smaller, but by timing my blinks and keeping the eyes closed a bit longer, I no longer even feel the page flips and my eyes are more rested.

The difference that we feel is because with dead trees are are actively changing the page, whereas with eink we are waiting.

I don't know how you're flipping your pages, but it's possible to flip through a book at something like 6 to 10 pages per second. Which is how flipbook animations work.

Even setting aside this specific form of flipping, with a moistened finger and a small book I can leaf through at a rate of several pages per second.

The annoying part with E Ink update speeds isn't the page turn you do when you've finished reading the page - it's the inability to rapidly visually scan the book.

I am not referring to flipping through a book's pages. I am referring to reading a book.
Considering I'm still waiting for my Indigogo backed 13.3" Good E-Reader from June 2016 I won't be holding my breath for this.
Me too, I can't believe the gall they have still taking orders.
If there's a sucker born every minute that also means every minute there's a sucker finally able to buy stuff on the internet without parental consent. Why shut down a scam under those conditions?
I love to have a reflective screen in my phone and as monitor.

For those devices energy consumption is just low priority.

Color and refresh-rate is where most reflective screens fail. So if you take on E ink I think that's where you should shine.

"Typically, the energy required for a full switch on an E Ink display is about 4mJ/cm2. The corresponding number for the Rdot display is about 1mJ/cm2 with the addition of 0,25mJ/cm2 every 15 minutes. LCD continuously consumes about 6µW/cm2."

This is some interesting numbers. So, for a 1 cm^2 screen, LCD consumes 6mJ for every 1000 seconds - i.e. just over 15 minutes. But to redraw the screen on eInk takes 4mJ. So if you do something that would cause the screen to fully redraw more than twice every 15 minutes, then eInk actually consumes more power? But wouldn't that generally be the case while reading?

What about the backlight?
That's an interesting question - the article doesn't say if it's included or not. Are you saying that it's not?

But even then, reflective passive LCD is a thing (I mean, we've been using it in calculators for ages, for one!). So this still invites the question - is it possible to make a reader that is even more energy efficient than eInk-based ones, with that tech?

While it's indeed interesting to geek out on this and I am not to say that we shouldn't be moving in this direction if we can... I don't think this is the limiting factor for this technology. Energy efficiency is already good enough with e-ink. For me more important is to try and build bigger (and cheaper) displays with higher resolution. For these I would sacrifice some energy efficiency.
On the broader market, you're right of course, as evidenced by the fact that many people read from regular phones or tablets. I'm mainly curious about how far this single parameter (power usage) could be pushed if desired.
Correct, I am under the assumption that it is not included. Some examples from real life: a phone’s battery life is often greatly improved by simply lowering the backlight. Around 2010 we started putting LEDs as backlight in devices, greatly improving battery life, and if you take a laptop with Linux you can often disable the backlight and power usage of 0 in backlight compared to screen off and using external only is the same for me (so rendering and output still occurs the same amount, but in one case there’s just LCD and no backlight and in the other it’s no LCD no backlight).
No, the screen is not usually "fully redrawn" when reading, which involves a full flash visible to the user. Often it's only a fraction of the screen that has to be redrawn as most of the pixels can remain the same between page turns
For a single page flip, sure. But how many page flips are you going to do in 15 minutes? Wouldn't you have enough changed pixels add up to at least a couple full repaints?
E-ink and similar tech seems to have tapered off the last couple of years, the only thing (as far as I know) that is doing anything new is the reMarkable. I was just about to pull the trigger on ordering one - anyone have experience with it? Is it really as paper-like as they claim?
I was thinking about ordering one, but out of the Boox note and the sony devices, I ruled out the remarkable first. I can't remember why, but you can search reviews for yourself.

I choose the boox note over the sony because it's more fully featured (e.g. better web browsing). I don't regret buying it. The only problem is that you need to use a screen protector to prevent scratching from the stylus, but the screen protector is impossible to place without air bubbles. Eventually I decided that the air bubbles weren't a big deal.