Looks like my post got a second chance. I'm not affiliated with the project, but I am an interested e-ink monitor enthusiast.
You may be interested to know that this project is a fully open project, all source code, verilog, algorithm documentation and hardware design files are on github:
That turned into a community project and eventually into building our own FPGA-based e-ink display controller to improve refresh speed and reduce ghosting.
The result is a dev kit and monitor: open hardware, supports 6–13" monochrome and color e-ink panels over HDMI or USB-C, up to 75 Hz with sub-100 ms latency, and includes a C API.
I'm probably getting one for desktop use, but I'm also curious for other use cases, what's the power usage like? while idle or doing high frame rate video?
I talked to my wife constantly about how badly I want an e-ink monitor, they’re just still so expensive for the specs I want haha. I’d take a solid 24”, maybe 21”, b&w 30hz at this point if the price was right (sub-$500).
Yeah, to get a 24" e-ink monitor anywhere near that price, we’d need much higher production volumes. Right now, it just isn’t mass-produced enough to bring costs down to the sub-$500 range.
Since you are really into that stuff, could you please share your opinion about the widely shared view that e-ink remains artificially expensive due to patents ?
Or point to reading material you find truthful. Thanks !
Being Dutch I am proud to see NLNet and the EU financially supporting this project.
We deny ourselves so much progress by forcing smart individuals with a passion into conglomerates that are merely busy destroying competition. Small to medium sized organizations have the biggest potential for innovation, and look what two people even can do.
It's an interesting project, but I'm confused why they got funded with EU money. Their company is based in Boston and it looks as both of them are living in them US, which suggests neither of them is European citizen or pays taxes in the EU, which ultimately is where the money for these grants comes from. The funding requirements of NLNet state "it is a knock-out criterion for each project to have a "European dimension [...] where a significant contribution towards the vision of the Next Generation Internet initiative also qualifies".
I'm not trying to pick on this particular project, I just found out non-EU projects can apply at all. I hope the majority of NLNet money goes to projects that actually are executed in Europe and build expertise there.
Thank you for your kind words. I agree that small teams can achieve a lot.
We’re grateful to NLnet and the EU, whose support made this project possible and especially in overcoming complex challenges. We’ve made a point of attending conferences and sharing what we’ve learned. Now that the project is ready, we can focus on expanding that knowledge and collaborating with people everywhere.
There was a really neat post here from November of last year about a person making an e-ink display for his mother with amnesia. It may be of interest if you like this stuff.
The devboard shipped with this kit has USB Type-C DisplayPort and DVI via microHDMI, both provided by off-the-shelf interface ICs. I would love to see a version that can take LVDS from a typical laptop motherboard to allow for modding a laptop (maybe a Framework 13)? The design appears to use a low-end FPGA with fairly modest resource usage, so I suspect it is technically feasible to interface video LVDS direct to the FPGA. I'm not sure about the power requirements however.
The Type-C DP input appears to be using a DP to LVDS converter, so the FPGA is already accepting LVDS, it just doesn't expose a connector for it. But for interfacing with laptops, these days it's eDP that matters, not LVDS. So the issue is mostly the mismatch between an internal eDP connector vs the board's type-C connector.
This would be absolutely amazing for a productivity device. I've rooted my Kindle Paperwhite and set it up with a terminal and used it to SSH into my laptop just to try it, but the latency makes it a bit irritating just to keep up with typing. To be able to use a fully graphical environment in e-paper, even in grayscale, would be amazing.
I built a cyberdeck that primarily uses a pair of XReal AR glasses as its display, but to have the option to use either those or this would be so awesome.
Even without the advantages of e-ink, a display which one can view _anywhere_ is a compelling thing, which opens up a lot of possibilities --- my favourite Windows device was a Fujitsu Stylistic ST-4110 w/ a transflective LCD, which I used as my main computer, e-book reader, notepad (writing/drawing/annotating with a stylus), and map display when traveling.
The transflective display eliminated issues regarding reflection or the display getting washed out by even full, bright, direct sunlight (I would use it for reference material for building sand castles at the beach).
Unfortunately, transflective LCDs do _not_ showroom well (dim) and no one seemed willing to make the investment to show their capabilities (build a daylight-equivalent light booth on the store floor).
I keep eyeing a Daylight Computer, but these days, I just use a Kindle Scribe for reading and note-taking/sketching/reference, and I limit my activities when in full sun to those things which it can do well, changing location/finding shade when I need to do other things.
HannsNote2 looks amazing in the sunshine. But very slow processor and very low capacity battery. It’s a niche use case but great for a couple of hours of KOReader in the sun.
Have you written about this anywhere? If not, I’d love to see a ShowHN post about your project and I bet I’m not the only one.
What’s your background? I ask because I really want something like a TRS-80 Model 100. Is interfacing a low end (power efficient) board to an LCD or eInk display something an electronics neophyte could reasonably take on?
A lot of homemade slab computers seem to rely on Raspberry Pi boards but those draw quite a lot of power and tend to run Linux. It’s not a combination that I’m going to be able to run for weeks at a time on 4 AA batteries (one of my goals).
When building your cyberdeck, were there any resources in particular that you found helpful?
There's a lot better being done by Boox, and even with colour e-ink in terms of ghosting. I think a lot of that is software end, though.
The one unfortunate thing is that this monitor seems to have a glossy screen, not matte, but maybe that's an additional layer over a dev kit?
If this truly is 'open', then it should be trivial to write special X11/Wayland drivers for it, to handle a lot of the ghosting issues at that end. I think Boox actually refreshes portions of screens, and a double or triple video buffer in X/Wayland could do the same.
(One problem with Boox is their relentless phone-home to servers in China, which cannot be disable by normal means.)
Another problem with Boox is the disregard of the requirements of the GPL family of licences. I've been interested in some of their devices but won't touch them due to that (and now due to the issue you stated - though I was unaware there was un-disablable “telemetry”, I'd have to look into that if they ever did something about the lack of GPL compliance).
The screens included in these kits are glossy, however the board can support pretty much any screen up to 13”, including matte and flexible. On the ghosting - we’ve been thinking a lot about Wayland drivers and other software solutions. We’re working on an SDK that allows a compositor or other software to not just refresh portions of the screen but also change the display mode for certain portions as well.
Thanks! Yes, it does update quickly, and ghosting is still an issue, though I think there’s room for improvement on the OS/software side. I gave a talk at the Linux App Summit earlier this year where I shared some ideas for how this could work.
I think, if you look at Boox and how they have display options and config for each app, you could do this for X11 quite easily. Especially a custom backend video driver which flattens to greyscale, and distinguishes change on a per app basis.
An incredibly hacky thing, is that (as an example) Sawfish is an immensely extensible window manager. You can literally drop in lisp code linking actions by tagging on window names, application names, and so on.
So you could actually extend sawfish quite easily, and then do what Boox is doing. Change per-window methods of refresh, by interfacing with the backend video driver in X11, and so on.
I think a lot of people who don't use Linux, would "put up" with Linux if there was a way to use a framework with a display. So I think, strongly think, that the display of course and your work of course must be done, but that it relies very strongly on slapping together a more holistic environment.
Thank you for your thoughts and support. I agree, doing this in X11 is definitely doable. I’ve seen others achieve similar results by running a daemon that listens for damage regions. I hadn’t considered using a custom backend video driver; that’s something I’ll be looking into.
My focus has mostly been on Wayland, but possibly prototyping in X11 could be a starting point. Some people I've met are interested in using computer vision techniques at the hardware level, or making a camera rig.
What excites me most is getting this into people’s hands, seeing what they create with it, and working together to improve it and explore new use cases. One use case that I learned about recently:
E-Ink News: The mystery of cuttlefish disguise, scientists use E Ink technology to explore its secrets
It looks amazing for e-ink. You'd probably don't want this for video anyway, besides it's not even in color. I'm impressed that it works at all. For a lot of work though it could be amazing, depending on your environment.
Say you'd want to pop outside in the nice weather to do some programming. You quickly find why that wasn't as glamorous as you expected. But if you had a laptop with such a screen I would expect it to work great.
I built a colorscheme for Vim [0] that is close to monochrome when I was worried about blue light (I've since had kids and now this kind of worry doesn't rate at all), and I basically use bold and highlights (I hate underlining, italics, and squiggles, but I think you can do all that too in the GUI version). I found it surprisingly usable.
I want to try some monochrome UI, but that amber is definitely not my taste. Let me see if there is a "paper-like" theme, or maybe I can make one myself.
You have the option of a few greyscale levels, bold, pehaps italic, and depending on font maybe extra-bold and light. That should be enough for the essentials, though it will feel like a downgrade if you have got used to a richly colourful environment and rely on it for reasons other than liking it being pretty.
I don't require it myself but that's a concern, it's nice to have. Maybe someone can build an editor that uses different fonts within one file instead of different colors. Could be something out there for color blind folks already, though seeing no colors at all is unusual. But e-ink has grayscales so you could at least make comments a bit lighter, I think I'd be happy with that.
There's plenty of options available. A coworker of mine used to print out code for reviews. You can use italic, bold and underline as alternative to colors. Grayscale might work nicely for eInk too - for laser printers just thin/regular/bold probably works better.
Other fonts... I could see myself being distracted by changing fonts in a document, except maybe for comment blocks. But for those italic/thin seems to work well already.
Tried to find the tool... it's GNU enscript. Syntax highlighting for several languages, outputs to postscript.
I'd do 3 greys, from lightest to darkest: comments, syntactical cruft like braces, semicolons, certain keywords, etc., and then 'actual' code, variable names and so on.
Much more variation than that with 256 colours is mostly just making it pretty rather than offering helpful distinction imo.
That was my very first thought as well. I'm wanting to build a display that can show the status of various things - garage door, temperature, air quality, etc. Ideally it could run on battery too so I could mount it anywhere. Just need to find the time and motivation to do it, among all the other priorities & interests.
Their big claim, at least it seems on quick glance, is in the controller, not the display. You can use any e-ink display with it. The display with glare that you're complaining about I think is just the cheap display they're including in the dev kit.
Oh really? I completely missed that. I thought what they were offering was an e-ink display with a latency that's low enough to use as computer screen.
Soldered Inkplate is a solid option for anyone wanting a device with Wi-Fi, backlight, touch screen, battery, and runs arduino or micro python. Recently set one up as a display that updates if I'm in a meeting or free that I hang on my office door.
Also shout out to https://usetrmnl.com/ which is much cheaper than OP if all you need is a display that updates every 5+ minutes. The founder is a really cool guy and cares a lot about this project
I'm disappointed by how expensive these things still are. Smartphone size eink panels are so cheap that every supermarket has thousands of them as price display panels. but try and find epaper bigger than 10" and you'll be paying $350+ just for a panel that does nothing by itself.
You may be interested to know that this project is a fully open project, all source code, verilog, algorithm documentation and hardware design files are on github:
https://github.com/Modos-Labs/Glider https://gitlab.com/zephray/caster