Apparently E Ink managed to move four different types of colored particles (white, cyan, magenta, yellow) through the fluid inside the pixels: https://www.inel

Apparently E Ink managed to move four different types of colored particles (white, cyan, magenta, yellow) through the fluid inside the pixels: https://www.ineltek.com/wp-content/uploads/2017/08/ACeP-How-It-Works.jpg https://www.ineltek.com/wp-content/uploads/2017/08/ACeP-How-... The cyan, magenta and yellow pigments are transparent like in inkjet printers. They are used to subtractively mix red, green, blue and black. The white particles are not transparent, so they block the rest when they are on top. So E Ink is able to create the eight primary colors: https://technewsspace.com/wp-content/uploads/2023/04/1681215995_840_E-Ink-introduced-Spectra-6-color-displays-with-electronic-ink.jpg https://technewsspace.com/wp-content/uploads/2023/04/1681215... I'm not sure whether they are able to create other colors inside a pixel, but these eight are enough to create full color displays with dithering. But the problem seems to be the time required for separating the pigments. 12 seconds is absurdly long for a full refresh. I think the electrophoresis screen technology is fundamentally a dead end here. Moving solid particles in a fluid introduces drag, which will always be slow. I thought the second generation "SMI" Mirasol screen technology by Qualcomm looked a lot more promising. It doesn't require subpixels and probably would have been fast like the first generation. Here is a video from 2014: https://youtube.com/watch?v=VMDaJB2y4vc https://youtube.com/watch?v=VMDaJB2y4vc However, Qualcomm stopped investing in developing the technology shortly after it was shown. It would have worked by moving solids (MEMS) through the air instead of through a fluid, which doesn't introduce the drag problem present in E Ink's electrophoresis displays.