E-paper display reaches the realm of LCD screens
spectrum.ieee.org
spectrum.ieee.org
EInk needs a lot of power to move the heavier ink particles around. If you are doing that more and more rapidly, then even more power is drawn.
By 75Hz, I'm almost certain that LCD is far more power efficient. The LCD pixel (aka the liquid crystal) is a glorified capacitor, it takes some power to charge but it's exceptionally 'light' compared to eink.
That's why LCDs can go faster and faster. It's just physics. A capacitor / twisted crystal uses less power to turn on or off than EInk.
---------
EInks advantage is that if you turn off power, the ink stays put. So you spend a ton of power moving the ink around and then save lots and lots of power over the next seconds, minutes or more.
That's why EInk is ideal for once-a-day updates of prices (or other retailer tasks). The less you update, the less power used.
https://www.crowdsupply.com/modos-tech/modos-paper-monitor/u...
> E-ink screens are quite power hungry when it comes to peak current. Modern high-resolution panels can consume >20 W peak.
This is where I was wondering and yeah, 20+W is pretty hefty to support a relatively small 8" EInk screen or something.
All those updates cost all that power as long as updates are occurring. Maybe you can optimize many of them away (if some parts of the screen don't move, especially if software was rewritten to optimize for the display).
More importantly, it sounds like you've created a full custom FPGA controller over the voltages that go into an EInk display? That's impressive in its own right even if I don't think 75Hz is a good idea lol.
--------
FPGA or Full Blown Microprocessor are the only choices here. A high power SIMD/NEON arm64 probably could do the job, but I think the Spartan6 is a good choice as well and has more obvious and straightforward parallelism (and probably all the pins required to control the screen. Even a big microprocessor won't have as many low latency pins as an FPGA).
Yes, that’s definitely something we want to work toward. As the community grows, we hope to tackle these kinds of optimizations together.
> A high power SIMD/NEON arm64 probably could do the job, but I think the Spartan6 is a good choice as well and has more obvious and straightforward parallelism
Yes, precisely for the reasons you stated. We also talk more about it below:
- https://www.youtube.com/live/okjJURIejIY
- https://github.com/Modos-Labs/Glider?tab=readme-ov-file#desi...
We personally couldn't make it work with low power but this seems promising!
> “Traditionally, the [e-paper display] controller used a single-state machine to control the entire panel, with only two states: static and updating,” says Modos cofounder Wenting Zhang. “Caster treats each pixel individually rather than as a whole panel, which allows localized control on the pixels.”
So in practice would it ever actually hit 20W unless you're doing something that requires constantly updating the whole screen, like playing a video game or watching a video? Surely updating only a few pixels at a time can't be that power hungry, right?
My Kindle has much less contrast than my iPad, phone or computer - albeit I have brightness turned up.
I only use the Kindle as its battery will last over a day if reading my iPhone will not and also if reading in bright sunlight.
If you think of the refresh rate not as a constant frequency but as variable with user input, there are some cases where driving eink quickly in short bursts could make sense? It seems like this project offers a foundation for such a controller, where e-reader controllers are strictly optimizing for low refresh rates. E-ink is not going to be competitive for playing a video game or watching a video, but you can create a more responsive experience with less eye strain for typical tasks like marking up documents.
Part of the challenge is deciding what belongs in hardware and what should sit higher up in the OS or software stack.
Hopefully, as more people get the kits and the community grows, we’ll be able to think through these questions together and explore where the right balance between hardware and software should be.
Identifying when to increase the refresh rate may be a challenge but I can see it pretty doable for this kind of "limited" scenarios where you either read or navigate a storage app.
E-ink's other advantage is that it reads like paper. In a desktop context I could not possibly care less about the power consumption, but being able to read a forum thread, chat channel, HN discussion, etc. without a backlight would make my eyes very happy.
¹ https://shop.dasung.com/products/dasung-25-3-e-ink-monitor-p...
I'm aware of a lot of anecdotal evidence in favor of e-ink displays being easier on the eyes than normal LCDs in some way, my own personal experience included, but I will happily admit I'm wrong if there are studies indicating otherwise.
I like my Kindle and DIY e-ink weather display but I'm not religious about it. I wouldn't be shocked to find out it was just a weird placebo thing because it's different.
[1] suggests that LCD even increases processing speed compared to e-ink
[0] https://pubmed.ncbi.nlm.nih.gov/22762257/
[1] https://bop.unibe.ch/JEMR/article/view/2338/3534
there's a study from Havard concluding "e-ink is 3 times better for eye health than LCD" but it feels rather dubious from the claims (blue light stressing more the retina... like i couldn't use a glass or apply a filter on my screen), light intensity (again...), in-vitro study and who funded the study (a great e-ink screen producer) -> https://sid.onlinelibrary.wiley.com/doi/10.1002/jsid.1191
i probably read hundred books since i sold my e-reader and moved to my smartphone. i really like having a single device. battery is fine. physical books with images still rocks but maybe becuse i don't have a tablet :)
(i.e. setting the "frontlight" brightness to 0% on a Kindle, which would also eliminate color temperature control other than room ambient light).
It does seem hard to believe that e-ink + a reflected frontlight would be any easier on the eyes than an LCD backlight (particularly since it's probably also using PWM). But an e-ink display on its own at least removes an additional light source pointing directly at the eyes, which could provide a potential mechanism for different effects on the eyes/brain.
Not really, since LCD/OLED aren't an additional light source, but absorb and thus replace the ambient light that would be coming from their direction.
I started as a teenager with cathodic tubes, which were killing my eyes and bringing daily headaches; moved on to LCDs which stopped the headaches but still tire my eyes significantly (some of them literally make me cry after a few minutes); and then found eInk and it's so much better, I will definitely move to that once prices of large color monitors at 60hz get into my price range. I honestly don't care about power draw one bit.
Quick experiment to show the effect: Go into a room with low 2700K or lower-temp lighting. Take an LCD, set its colour temperature same as the external lighting, then display an all-black screen. Since the screen is displaying #000, the software colour temp adjustment can't do anything, and you'll see the screen as emitting blue light, the colour of its backlight.
OLEDs don't have this issue, which makes them great for night-time use when configured properly, but they also generally use low-frequency PWM dimming on low brightness.
Don’t they actually have to over power the reflection you would see with the screen off?
Their brightness can also be conveniently, even subconsciously controlled, by how they are held.
LCDs can be dimmed and brightened, but matching the E-Inks “response” in both brightness and contrast over a high range of ambient lighting would be difficult. Probably impossible without an LCD specifically designed to do that.
I frequently stumbled upon the assumption that a LCD screen is pulsed and flickers and that makes all the difference as E-Ink is more steady. (Artificial lightsources can also flicker, but with reflection it evens out)
In (old?) theory too fast for the eyes to notice, but I surely notice a difference.
You're a software developer. How do you function? Do you spend pretty much your entire waking life with a headache?
For text updates, where there's literally a cursor which moves at typing speeds, update frequency is quite low. Where you're updating or paging through documents, paginated navigation (where the whole screen refreshes at once, then remains unchanged for several seconds to minutes or longer) is quite efficient.
It seems much more practical (if a little less power-efficient) to implement the no diff -> no refresh logic for screen regions in the display hardware. The RAM and logic for a display-side framebuffer can't be expensive today, a couple of Euros for the extra board space and chip(s). If that stuff takes off, just additional transistors in the all-in-one ASIC.
For the whole screen, that more or less already exists in laptop hardware: "panel self-refresh". HDMI and DiplayPort might need a new extension or something? Is there anything?
LCDs use articulation of liquid crystal chemicals that change shape thus polarization upon application of voltages. They tend to slowly deform back to "the other" state when voltages are removed, and also tend to chemically break down if not moved back to the neutral state. LCDs are driven in pseudo-alternating current for this reason, and never held at either extremes for long time, for this reason.
So you can drive E Ink at 75Hz or whatever, it'll just take more power than it takes LCD to do so, and the last pixel states will persist. Or you can leave LCDs at extremes and disconnect the power, but it will lead to degradation if intentionally used that way.
What you can't do is 1) "watt per frame" figures of LCD, with 2) persistence, and 3) long life. (1, 3) is LCD, (2, 3) is E Ink, (1, 2) is LCD abused as if it's E Ink at expense of rapid degradation, and (1, 2, 3) is the holy grail.
If so, you're probably still right when it comes to watching a video or something, but e-ink could be more efficient for drawing, writing, or reading.
The bare panel, for both LCD and EPD, would consist of a pair of glasses vapor coated with transparent indium tin oxide, chemically etched as bunch of horizontal lines on the first one and vertical in the other one, with corresponding choice of fluids suspended inbetween. It would be possible to wire a custom fabricated controller onto those row and column electrodes to drive individual pixels. I guess that is what is being done here.
Mysterious future Tech#3 will break the rules. OLED for example uses far less power on black pixels. It's just different.
Color was lower contrast of course.
https://www.youtube.com/live/okjJURIejIY?feature=shared&t=24...
There are refresh modes now which are very good at partial updates.
Basically, it's intentional and relatively rare (unless you have really really short chapters).
https://www.amazon.com/dp/B0CDBYFH81
and a colour Boox. The refresh on the Boox is so fast, you can watch video.
I don't think the Meebook has refresh issues, it's definitely very fast for back/forward. Of course I don't have page turn animations on. I use the kindle app on it, and KOreader. If you get this, know that there is a thread to drop in a script, so the volume buttons page turn correctly when upside down.
If you live in Amazon turf, you can always buy and return if not suitable?
There are some reviews on Amazon.
In any case, it is not clear that, even if EInk Displays are somewhat worse in power consumption, they aren't a superior display technology.
If so, won't high refresh rates degrade eink rapidly.
https://github.com/esphome/feature-requests/issues/1109#issu...
This could be actual burn in, or it could be a failure in how they are refreshing (with some potential fix if refreshed properly). I’m not familiar enough to be certain myself, but I personally suspect they are likely being driven too hard and are truly damaged.
In normal e-reader use I’ve never seen this as a practical issue.
https://www.youtube.com/live/okjJURIejIY?feature=shared&t=24...
I also have another display that was exposed to full sunlight through a window for about 8 years. It's now a bit faded as a result.
All in all, I consider it pretty good.
https://files.waveshare.com/upload/4/4e/2.13inch_e-Paper_V4_...
https://files.waveshare.com/upload/6/60/7.5inch_e-Paper_V2_S...
Would it be possible to re-use the power that is stored in them?
I dunno if LCD screens are multiplexed, but embedded LCDs could need as many as 8x on/off cycles per pixel (because you save 87% fewer wires if you chain 8x pixels on the same line and then put them on different biases and have weird COMmon pins and rows and crap).
Sharp claims that a bit of storage is on each LCD/capacitor and this saves power somehow with smarter decision making. I assume it's minimizing the wasted power somehow (or even recycling the previous power shoved into the capacitor, which typically just goes to waste).
It's some proprietary formula in any case, so it's all guesswork. Only the Sharp Engineer/Inventor would know for sure.
The net effect is that the app can vary refresh rate and also update a small part of the screen while leaving the rest static. In Garmin smartwatches, this is used to switch between 1 Hz, 1/60 Hz, and a hybrid mode where the overall screen is 1/60 Hz and small part is updated at 1 Hz. E.g. the seconds digits may remain active while the rest of the screen is static.
It also seems like they use more interesting sub-pixel arrangements. The limited color depth may involve different combinations of 1-bit sub-pixels rather than any cells driven into partially twisted states. I've seen diagrams claiming some white (unfiltered) and RGB filtered sub-pixels of different sizes.
I am absolutely not surprised to see his name behind this startup. I've been following his work for years at this point; his YouTube channel has always deeply impressed me, and he's done wonderful open source work in the realm of E-paper for quite some time now.
Kudos to him, and I wish him all the best.
[0] https://www.crowdsupply.com/modos-tech/modos-paper-monitor/u...
I went through that and then bought a Carta 1200 display BOOX 13.9 and it's amazing. Black and white only, but the contrast makes the device usable.
If you know you won't return the device, get it on their website because they'll give you extra pen tips and a case. I got mine on Amazon, so I missed out on the extra stuff because of my return experience.
And pixel-level addressing isn't innovative either. If you've written on an e-ink tablet and observed that the screen doesn't refresh with every pixel change under the stylus, that is surely because pixels are being toggled individually instead of doing a full screen refresh.
So perhaps the only difference is that it's an open source controller that's competitive with commercial e-ink display controllers? That's no small achievement and worth celebrating in and of itself. But it's not at all made clear by the article.
- Making the project open allows people to reuse displays they already own.
- Others can contribute and build on what’s been created.
- Open source firmware, documentation, and the driver board make development more accessible and help remove barriers that previously slowed community projects.
- It’s designed to work with a variety of electrophoretic panels, not only those from E Ink.
In the long run, this openness will strengthen the ecosystem, making it easier for new ideas to take shape and spread.
Regardless of manufacturer (remarkable, boox, supernote…), all e-paper tablets have one major performance problem: quickly scrolling through multiple pages of notes. No idea if the display is the limiting factor, or the cpu, but I’ve hit this issue on all tablets I’ve used. If you like riffling through pages in you paper notebook, you will hit the limit too. I know at least 2 people who stopped using their tablets over time because of this issue.
If this tech helps solve that problem, it’s more important to me than an eink monitor.
Edit: this is mainly important for notes, because sketches, scribbled diagrams and quick notes half-taken in meetings are not really searchable. PDFs and ebooks don’t have this problem.
Is there anything mitigating that issue?
Call me crazy, but I'd rather see these guys get a couple million than yet another chatgpt wrapper.
16 levels of grayscale support.
> How bad is the ghosting?
Ghosting depends on the mode you're using and the content.
> How white is the background?
Varies, depends on the panel you're using.
> Is it clear enough to be used white-on-black?
Yes
> How often does it need a full screen refresh?
That's up to you; you can manually clear with a button press, use auto-clear mode, or programmatically control it.
https://www.youtube.com/watch?v=AoDYEZE7gDA&ab_channel=Modos
> Ghosting depends on the mode you're using and the content.
>> Is it clear enough to be used white-on-black?
> Yes
I agree that the dark mode on the boox is terrible (I find for example obsidian absolutely unusable). But this could be easily fixed with bolder fonts.
I'd love something similar to the Lightphone but a bit smarter.
And a laptop or device I can code with outdoors at a picnic table in broad daylight.
Besides the obvious benefit of having a single tablet with both types of screen, in "laptop mode", the e ink screen could be used as a soft keyboard and touch pad.
Something like this in phone sizes could be really great too.
In this parallel universe, I can afford the flagship laptop with the latest and greatest CPU and ample RAM because I have saved big on getting the 4x3 1600x1200 screen instead of the 4K AMOLED colour screen. I also get to eek out extra battery life big-time since the laptop goes into suspend e-reader style so I don't even know.
On the train the laptop works fine with nobody wanting to steal my PC. If I really need colour, I just pull out my mobile phone.
But there will never be the market for that. Monochrome LCDs went at the start of the century since everyone just wants colour. It is the same on desktop where, like yourself, I would really want one and be prepared to buy one even though I would have never tried f.lux.
For some of us it is not the greyscale that we find easier to read. e-Ink reflects light, while LCDs emit light. My eyes like first a lot more.
To get past that, we would need a different panel technology, a type of reflective ("e-paper") panel that is not based on electrophoresis.
Years ago there were many such display types in development. One option is electrowetting displays. Liquavista was a company that had a screen where tiny oil droplets were switched between being either round and small or flat and large, using high voltage. The flat droplets would cover the background of a pixel and make it dark, while the small ones would "hide" in the corner of the pixel to make most of the background visible. This is pretty fast because the oil droplets are surrounded by air, which doesn't resist the movement of the oil, in contrast to moving solid pigment through a liquid.
Another option was to to have microscopic mechanical (MEMS) plates inside a pixel, which produce color by creating light interference. Qualcomm's Mirasol tried to do that. The wavelength of the reflected light depends on the gap between the plates.
The cool thing with interference e-paper is that you can theoretically make a color display which doesn't need RGB subpixels. Colors could be created by continuously adjusting the gap rather than doing binary switching between black (UV or IR) and either red, green or blue. Not having RGB subpixels greatly increases contrast on colored screens because it can reflect much more light. An issue is that shades of white and magenta can't be straightforwardly created with interference, because those are not monochromatic colors with a single wavelength. Anyway, Qualcomm closed Mirasol just as they tried to make these subpixel-free screens viable.
[1] https://www.good-display.com/product/452.html
[2] https://news.samsung.com/global/samsung-showcases-color-e-pa...
https://www.crowdsupply.com/modos-tech/modos-paper-monitor/u...
Curious.
Compared to LCD, oled or what have you, my understanding is that it uses significantly less.
See the microwatts of power that Sharps MemoryLCD displays have. They often beat comparable EInk screens in power draw.
The parts of the screen that doesn't update, courtesy of being e-ink, don't use any power at all. LCD will use power if you're looking at a static image, eink won't. And a lot of the time, 95% of the screen is a static image and only 5 percent actually updates. One of Modos' biggest innovations is successfully taking advantage of that.
That's unfortunate.
I'm imagining a fast scrolling game with complex backgrounds where most of the pixels are changing values every frame, I assume it completely breaks down in that case.
Or that's how I understand it anyway.
I saw that Alex Soto himself is in this comment thread, he'll know a lot more than me, I'm just spreading what little knowledge I've gathered from his blog posts and some of the discussions in the modos mastodon server.
I've probably misunderstood a lot of that too, I'm not a hardware engineer, just a lowly java dev with a strong but hobby level interest in eink.
Modos is my dream laptop, but it's currently unclear when that'll become reality.
Again, Alex Soto will know more.
It would be cool to see a Linux distribution with a gui and windowing system specifically designed for e-ink displays.
Not sure what optimizations would even be needed…
This would be amazing
I enjoyed that quote.
Not really knowledgeable enough about the tech, to comment further, but I like EInk, and look forward to seeing it be more useful.
Thanks!
[1]: https://www.crowdsupply.com/modos-tech/modos-paper-monitor
I basically want to build a custom e-reader with a RasPi Zero for learning/home use, 8-10inches would be great.
Don't care much about it being touchscreen.
Though smaller than your target size, I did[1] something similar, and it was great fun. I believe there are still Waveshare hats with larger display sizes.
Yeah even I could only find waveshare displays at the size I wanted that were cheap enough not to burn a significant hole in my pocket.
It's high resolution, snappy, and the whole package is light as a feather and with batteries that last for ages.
I know some people prefer paper, but I love modern e-readers. They're amazingly tuned.
Monochrome e-ink has a better resolution and contrast ratio than old-school LCD devices (I'm comparing my experiences with a Palm Pilot in the 1990s and an Onyx BOOX in the 2020s). LCD can refresh far faster, in the 100+ / 100s Hz range, where typical e-ink refresh rates in my experience have been in the single-digit to low-double-digit Hz range (video is doable but far from ideal).
E-ink also displays quite nicely with a "frontlight", which brightens the background (whiter whites) without washing out the foreground (print/ink). Illuminated LCD displays tend to wash out the dark fields, though I've not viewed e-paper directly and cannot speak to that.
TFA is describing a far higher e-ink refresh rate than I've experienced directly.
The thing about MIP is that the viewing angles are just not that amazing. I have had a Kindle and a Kobo, and they look like paper no matter how I hold them. My Playdate however needs to be positioned at a pretty specific angle with respec to the light to get the best contrast.
So, it is comparable to LCD sure, but to an oldd LCD of CSTN tech or such.
For color e-ink displays, instead of competing with LCDs, target a niche market: 8-color terminals for programmers.
We always aimed at fixing the lag, be it terminal rendering performance, network jitter (mosh anyone?), proper tab completion (including the ones that require network responses to complete), TUIs...
I'd still use eink for terminal, if it was cheaper. Just saying refresh rate is important for terminals.
Is refresh rate necessarily tied to ghosting? Like higher refresh rate also means higher ghosting?
The thought of high response high resolution passive lit screens appeals.
Sick and tired of seeing really neat announcements with pricing out-of-bounds for hobbyists.
(At least those who aren’t prepared to spend thousands just to experiment with a new toy screen)