Applied Science
4 HN recommenders · recommended in 8 months, 2018–2021
27:55 · 582K views · Oct 29, 2017
Ben Krasnow of Applied Science explains how electrophoretic e-paper displays work and how he modified the driver firmware and waveform lookup tables of a 4.2-inch panel to reach roughly 3Hz partial refresh. It covers the tweakable constants in the refresh process and touches on colored e-ink microcapsules.
Whenever e-ink refresh speed or driving waveforms come up, HN points to this as the clearest hands-on explanation of how these displays work and how to hack them.
12 comments from Hacker News
Ben Krasnow (Applied Science on YouTube) did some research into driving e-Ink displays, and he made some similar observations: > All of the lookup table information has been removed from this datasheet, because the industry thrives on secrecy or something Highly recommended, as usual for him it's a great video/write-up: * this video * [LINK]
Turns out that e-ink displays typically don't mix colors this way. In order to get multiple colors in a single display, they use sort of colored-balls of various densities suspended in a fluid. Then, using the known viscosities of the fluid in the capsules, they can vary the charge of the electrodes very quickly to sort of _drag_ a color up to the top. But when you do this, you're dragging a _single_ color. _All_ the colors are in the capsule, but you can only truly pull a single one at a time, a true pigment-like mixing isn't possible. If you pull the Cyan up, you're necessarily losing the Magenta, for example. If you had a high enough resolution addressable grid, you might be able to put the pixels close enough that you couldn't tell, but you'd still be basically doing sub-pixel color at that point, not truly mixing the various colors at the base level. Maybe there are other e-ink displays doing something else, but all the ones I've seen are doing some form of this approach. Here's a fascinating video on how these displays work, and how you can modify the firmware of the driver to get faster refresh rates (with an increased risk of burn-in). this video
yeah I believe that was the only option for older / cheaper displays. I recently got a kindle for a present and notice that when it can, it will try to refresh the screen as little as possible. For example, when you press the top of the screen to bring up the settings, only the top cycles, and then the same when you tap below to get rid of the settings. To that end, I would be interested in watching some high-speed camera footage of the displays just to see the gears turn. I’ll have a look round and edit this comment if I find any. e: found one. [LINK] also a video from an engineering channel that normally gives good insight into these things, I have watched it yet so it may not be great, but Applied Engineering normally brings some interesting material. this video
It's possible to refresh e-ink without a global flash if you are careful and have a clever controller. I don't think this video directly addresses this but it gives a good overview on the problem space this video
Great video on these sorts of displays, what they're capable of, and perhaps some inspiration for any hobby projects: this video