Re: DPI
Printing needs high DPI because it can't do a greyscale without dithering. Some Computer printers can compensate somewhat by using variable size dots.
Simplest paper printing is one shade of black. All shading is by dither patterns or variable dot size (halftone).
Simplest colour was C Y M. The blacks are muddy and fringe, so black is also used, CYMK. High Quality print can use maybe twenty colours or more, or six plus black on an inkjet.
Earlier eInk only did about two shades of grey as well as milky liquid (greyish white) or black (all the little black balls at the front). Current eink alleges 14 shades as well as "white" and black. Likely that's not completely accurate.
Anyway a 150 dpi LCD can look better than 600 dpi inkjet colour page and even in monochrome might look as good as 1200 dpi laser. High quality lithographic printing can indeed be a much higher dpi.
How is this possible?
It's the idea that you use intermediate shades to fill in the jagged appearance of a low resolution image. So with 150 dpi you can easily simulate 300 dpi or even 600 dpi pure black and white by using intermediate shades on the edges. Thus a 14 grey + black + white eink at 300 dpi can look as good as a 600 dpi laser print.
See
https://en.wikipedia.org/wiki/Spatial_anti-aliasing
It turns out that for Roman text we are more sensitive to horizontal resolution that vertical. It's not true for Arabic, Chinese etc. Think how most serifs work.
So we can colour the edges to take advantage that most colour LCD used to be RGB sub pixels for each pixel. Using that and shading can make a 150 dpi LCD black text look nearly like 600 dpi to 1200 dpi laser print.
and mainly for text
https://en.wikipedia.org/wiki/Subpixel_rendering
Unfortunately many OLED and newer LCD screens are not 3 times as many subpixels horizontally, but twice vertically and horizontally:
R G
G B
Thus each full colour pixel uses two columns and two rows rather than the three columns per pixel. Some displays even add yellow subpixels.
If it's not a pure RGB or GRB or BGR or BRG set of three pixels per pixel, or a CRT, then the subpixel text enhancement needs to be disabled. Called Cleartype on Windows.
Regular eInk has no subpixels in the OLED / LCD sense, so it's only trick is to fill the jaggies with grey.
Also though print companies may want 300 dpi or better images, the real print resolution in colour or greyscale may be like a 150 dpi to 200 dpi true colour LCD image because they are using the 2000 dpi to 8000 dpi effective print resolution to dither and halftone.
See
https://en.wikipedia.org/wiki/Dither#Applications
and
https://en.wikipedia.org/wiki/Halftone
So we likely don't need 600 dpi or 1200 dpi ever on eink for text, just better rendering of the 14 inbetween shades and text display that uses them to anti-alias. Maybe 400 dpi and better font rendering.
Actually looking with a magnifying glass on the 300 dpi Kobo, I see jaggies on diagonals and no evidence of greys.