> I rather mean the firmware of the e-ink driver board, which is a trade secret. I don't know, maybe it is not even firmware in an MCU, but they have a dedicated driver chip and it just has the look-up-tables. Anyway, the secret sauce that tells you how to drive the display cells.
"Secret sauce that tells you how to drive the display cells"? You mean like a voltage table that is also present inside every LCD or OLED? The difference would be that the electrophoretic display would need a much bigger table so it would have to be kept on the SoC since it can't possibly fit into a single voltage driver circuit. That's not software, that's just a big table of voltages that's hardcoded for each unique panel. Is that what you think is "secret sauce"? Do you also want to extract that table inside each LCD drive circuit as well? I guess you could use it to make your LCD or OLED panel show brighter colors but at the risk of burning and damaging the crystals. I imagine the same risk would also be true for electrophoretic panels or anything where you can change the physical voltage that is being applied to the material.
It's actually simpler than a table of voltages. It's a series of trinary values that indicate whether to use positive, negative or zero voltage. The actual voltage used is static (even the specialized EPDC PMIC (which _is_ a separate chip) doesn't allow changing it on the devices I've seen). The waveform (as they call the lookup table) is sometimes actually stored on a separate flash chip soldered on to the display's built-in cable. Years ago I wrote a tool to decode and convert the proprietary formats used by the E Ink corporation for these: https://github.com/fread-ink/inkwave
Just curious, I previously saw claims on HN that E-Ink is a very brutal cruel company that is evil and attacks everybody. That didn't line up with how their staff, at least the materials science guys, seemed to be when I encountered them at SID. I've been asking for evidence for this on HN. Did they try to take down your tool or anything like that?
Yes, but at the same time I think it is more complicated than a table of voltages. I don't know about the OG kindle, but more recent e-ink displays have a multi-parameter look up table. The applied voltage and duration also depends on this history of the cell (the current color you think it is showing) and the temperature. Depending on how many color and temperature steps you use, the tables can become very big.
In hobbyist displays, the tables are very simple, and I'm pretty sure this is one source of quality differences.
Huh? What e-ink driver board? There's no such thing inside a Kindle. It is a straight NXP SoC that drives the e-ink panel directly. There is no MCU. The driver is open source. https://github.com/canselcik/libremarkable/blob/master/refer...
"Secret sauce that tells you how to drive the display cells"? You mean like a voltage table that is also present inside every LCD or OLED? The difference would be that the electrophoretic display would need a much bigger table so it would have to be kept on the SoC since it can't possibly fit into a single voltage driver circuit. That's not software, that's just a big table of voltages that's hardcoded for each unique panel. Is that what you think is "secret sauce"? Do you also want to extract that table inside each LCD drive circuit as well? I guess you could use it to make your LCD or OLED panel show brighter colors but at the risk of burning and damaging the crystals. I imagine the same risk would also be true for electrophoretic panels or anything where you can change the physical voltage that is being applied to the material.