Open sergarcia opened 3 years ago
This also looks like it could be a promising addition: https://www.crowdsupply.com/e-radionica/inkplate-10
Apologies, I missed this: I'm notification bankrupt on github...
Overall, this project was meant to be a prototype. Ease of replication was not a goal here.
Coming with a mass-production friendly design is still on my infinite todo-list. It'll mostly depends on life and whether I'll manage to get a massive amount of spare time to sit and do that in the future :)
Will it work on the rpi 4?
As long as you manage to enable the SPI bus, there's no reason it won't. You may need to modify a bit the SPI init routine https://github.com/NinjaTrappeur/ultimate-writer/blob/master/screen.c#L15 though.
How does it relate to PapperTTY and other eink projects?
I'm not involved in these projects. This project's scope is obviously more shallow than papertty.
For what it worth, in hindsight, I think both this project and papertty took a wrong approach: we should rather implement a linux framebuffer driver instead of trying to hack around a e-ink specific terminal emulator.
That's definitely the way I'll push if I come back to this at some point.
What work is need it to make this usable (fast enough refresh) in larger screen sizes (programming background needed, difficulty, feasibility, etc)?
We need to reverse engineer the kobo spare screens. They're cheap and crazy fast.
As always, you don't need any particular background aside from a healthy amount stubbornness and some spare time :)
Now Waveshare has multiple displays of up to 10.3 inches that according to them have a full refresh time <1s and take ~.3s for partial refresh, including a HD 6 inch one. Could these displays make the project usable?
Sounds cool, but 250€ is definitely out of the price range I'm targetting.
This also looks like it could be a promising addition: crowdsupply.com/e-radionica/inkplate-10
Nice! Using spare screens is definitely the way to go! The precious waveforms are published in their source code, that's exactly what I was looking for.
Thanks for sharing this link!
Nice! Using spare screens is definitely the way to go! The precious waveforms are published in their source code, that's exactly what I was looking for.
Thanks for sharing this link!
No problem 😊
You might've noticed this already but if you don't want to wait they also have a 6-inch screen that's already available for sale (and is slightly cheaper). I linked to the larger screen first as I thought it might be more aesthetically pleasing with a full-size keyboard.
Indeed.
You did managed to nerd-snipe me: I spent a couple of hours doing a 2021 state of the art about driving spare kindle screens this morning.
Good news: the situation greatly improved since last time I checked (~2018). The waveforms are now widely available on the internet and on top of that, there are multiple open hardware driving boards around.
The most promising one to me being this one: https://github.com/vroland/epdiy. They already support a wide range of screens through a esp32-based driver.
Count ~20€ for a spare 6" screen, ~30€ for a 9.5" one. You'll have to handle the PCB manufacturing by yourself though.
This is veering away from the issue ticket, but I'd be curious to read more if you wrote that up. I don't have the skills to engineer this myself but I'd be very interested in following your instructions for assembly if epdiy addresses the refresh issue.
Hello all, I've started yet another "poor man's freewrite" eink typewriter project based on the Raspberry Pi Pico called picowriter also here on reddit. I am completely up in the air as to what to do about the screen. Right now I'm using a tiny 2.9" screen just because it was cheap to test with.
FWIW I got this to work on the Pi Zero W with no modifications as well. It has about 1/3 the battery usage, but also takes a few times longer to boot.
I find the project quite interesting, but I am not sure if people have been successful implementing it and what are its main limitations.