> I was wondering if these cheap digital photo displays could
> render text, cost effectively, and I realised that I don't really
> know how text is rendered on an LCD display.
> I'm guessing, working from the display backwards:-
> - How often is a non-changing display refreshed,
> and this would be done by a chip that's dedicated to the LCD,
> updating all the pixels of the display?
> - So that's 4 * 48* 600 bits for a 16 colour VGA display ?
> - What are typical resolutions for these 4 to 6 inch photo-displays?
> - the line of text would be encoded char-by-char to H-lines of
> pixels, where H is the pixel-height of the char-font ?
> Obviously the whole line of chars must be built before it's sent.
> - and then shifted into the 'dedicated LCD driver' ?
> - is it fed in 8-bits wide or what ?
> - what knd of signal do video DVD players output: analogue
> composite, or what ?
> - So you get: [char-line] -> [pixel-block] -> [display-driver]
> Can someone point me to a good online tutor ?
> == TIA
I'm not sure a tutorial is going to help you.
Perhaps you could start by purchasing and disassembling a photo display.
The technology likely changes every year, and as time goes on, the
approach will involve more SOC (system on a chip) approach to save
money. I understand that WiFi was supposed to show up on some
of the more expensive picture frames, in the last couple years.
It should have a processor, RAM, code EEPROM, larger flash memory for the pictures.
The LCD needs to be refreshed at regular intervals, which requires mechanically
reading out of RAM, sending the scan pattern over an LVDS interface to the panel,
and the panel takes care of the rest. For want of a better term, that would be
a "frame buffer", which is used to keep the static image on the screen.
Some picture frames now, play movies, so they have the ability to update the
frame buffer in real time.
So rendering text would be pretty easy. Prepare an "off-screen" frame buffer.
Draw the text into the frame buffer. When finished, the pointer in the logic
that refreshes the display, is updated to point to the new frame buffer containing
an image you want to display. It just so happens that the image contains a
rendering of text.
There are some other LCD technologies, where the interface is different,
and the methods used might also end up being different. For example, if
an LCD display device has internal scan memory, to contain the thing to
be displayed, there might be no frame buffer. If there is a "font generator"
inside such a device, perhaps you enter ASCII codes, to get a character
displayed. Such a display probably only displays text, and is not intended
for images. And that is the reason, it makes sense to have the font generator
onboard the silicon controlling the display.
But on the digital picture frame, considering the capabilities are very
close to those of a personal computer, rendering to an off-screen area
of RAM is likely to be the mechanism to use. You change one pointer
in the hardware, and instantly (one frame time), a new image appears
on the LCD. Meanwhile, your picture frame processor is composing the
next frame. The reason for doing the preparation off-screen, is
so the updating process is not visible to the user.
A two line or four line, monochrome, text only display, is radically different
than a 640x480 small LCD image display. I doubt I could do justice to the
economics of the two approaches, and explain how much money can be
saved by building the simpler text only displays. You can get some idea how an
image display works, from one of the pictures in this article.
I guess the difference, is "passive matrix" versus "active matrix".
Plenty of articles to read.