View Single Post
Old 07-28-2012, 10:41 PM   #44
geekmaster
Carpe diem, c'est la vie.
geekmaster ought to be getting tired of karma fortunes by now.geekmaster ought to be getting tired of karma fortunes by now.geekmaster ought to be getting tired of karma fortunes by now.geekmaster ought to be getting tired of karma fortunes by now.geekmaster ought to be getting tired of karma fortunes by now.geekmaster ought to be getting tired of karma fortunes by now.geekmaster ought to be getting tired of karma fortunes by now.geekmaster ought to be getting tired of karma fortunes by now.geekmaster ought to be getting tired of karma fortunes by now.geekmaster ought to be getting tired of karma fortunes by now.geekmaster ought to be getting tired of karma fortunes by now.
 
geekmaster's Avatar
 
Posts: 6,433
Karma: 10773668
Join Date: Nov 2011
Location: Multiverse 6627A
Device: K1 to PW3
Quote:
Originally Posted by NuPogodi View Post
...
(UPDATE) It seems that 8bpp-improvement is some kind of marketing mythos. Hex-codes in the attached raw file are always with the same high and low 4bit-nibbles, like 0x00, 0x11, ... 0xEE, 0xFF. It means that there are actually 16 colors. It might be a bit improved by dithering, but nevertheless... only 16 grades of gray, like in old-fashioned models (k2, k3 & kdx), just a bit remapped to increase the buffersize
See my newtrix demo to see how to mirror the top 4-bits into the bottom 4-bits on an 8-bit framebuffer. The GPL device driver code actually says that the bottom 4 bits MUST be equal to the top 4 bits, or the eink hardware can behave in an "unpredictable" manner (depending on specific hardware). My newtrix demo also does dithering to get 256 shades of gray into each of those 4-bit pixel copies in the second demo in that program. Check it out.
geekmaster is offline   Reply With Quote