Re: Display hurtful on LCD screen with Wheezy
Top posting because I don't want anyone to feel attacked. We are all
users here.
Hmm.
Look, even the engineers have to operate from incomplete models. No
one in this world knows everything. And that's true of the tech we
use, including computers. Shoot, even back in the days when some of us
thought a 6809 ought to be all the processor anyone would ever need,
the guys who thought they knew every thing about the micro-computer
devices they designed were often operating more on superstition than
fact in some areas.
(Yeah, hubris is an important quality in engineering. Being lucky is, too.)
(If you aren't familiar with the 6809, s/6809/Z80/ -- it's close
enough for the analogy.)
That said, I'm working from memory here, and I'm liable to use the
wrong jargon, but the last time I looked closely at the construction
of LCD monitors, most were using internal refresh.
In theory, the polarization bias of the display elements is static.
But in practice, well, plug "LCD bias" into a search engine.
I'm probably spouting engineering legend here, but my memory is that,
in actual implementation, engineers have generally found it convenient
to, rather than use a static bias, depend on decay times with LCDs in
a way similar to how CRTs did (and still do). The "legacy" of scanning
in CRTs probably lent a bit of push this way, and the necessity of
multiplexing pixel addresses probably lent a bit more.
Even without the electron gun, putting a running count on a
multiplexor input (with a memory read to get the pixel value to
refresh) provides a convenient scan, which can refresh the pixels, and
then you don't have to include circuitry to keep each pixel biased to
the last state written.
The pixels decay relatively slowly. Fortunately, they can be switched
well before they completely decay, but, especially with early LCDs,
the response was slower than a screen refresh. (My memory tells me
some of the really early LCD monitor displays took full seconds to
respond.)
The slow decay/response time is the reason LCDs don't do the strobe
thing when you wave your fingers in front of them.
What that has to do with Mark's issues, I'm not sure. Generally, the
engineers that design LCD monitor panels design optimal internal
refresh into the panels. I'm not familiar with whether some
manufacturers might provide some API to play with it.
It is not inconceivable that something plain Debian is doing in the
drivers (or not doing?) conflicts with the internal refresh
assumptions, but I'm hard pressed to guess what that might be. Unless
there is an API for Mark's monitor.
Well, maybe there is something to the frame rate. Not the video
refresh rate, but the effective frame rate for the system's update to
the screen. We talk about frame rates for games and video feed, but
the system is also going to have a something like a frame rate,
whether that would be a real frame rate or an effective frame rate.
And there could be some unfortunate interplay between that and Mark's
eyeballs. (Or, rather, the nervous system between the conscious brain
and the eyeballs.)
Perhaps, the effective frame rate causes a beat effect with the LCD's
internal refresh, but I would not expect that to be an issue with
anything but games.
Well, I have heard that some people have picked up bit-twiddling
between frames, trying to get a little better effect than simple color
adjustments, but I would be surprised if either Debian or Ubuntu is
doing that.
But I'm going to tend to lean towards some static variation in the
lower bits of the colors being put out, perhaps because of the
anti-aliasing settings or something.
(Mark, did you set the screen grab to give you raw images or compressed?)
Or it could be something as simple as the amount of line-spacing his
favorite text editor gives with the fonts he's choosing. I find myself
squinting a bit too much when I use Japanese fonts that have strange
leading.
On 9/21/12, Mark Allums <mark@allums.com> wrote:
> On 9/20/2012 3:37 PM, Lisi wrote:
>> On Thursday 20 September 2012 20:50:40 Mark Allums wrote:
>>> You were obsessed with the refresh rate.
>>
>> This is absurd. I mentioned it twice, amid a lot of other things. It is
>> you
>> who are obsessed.
>>
>> If you want to continue this absurd attack on me, may I suggest that you
>> go
>> off list.
>>
>> Lisi
>>
>>
>
>
> I'm not attacking you. On the contrary, I am the one who feels attacked.
>
>
>
> --
> To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org
> with a subject of "unsubscribe". Trouble? Contact
> listmaster@lists.debian.org
> Archive: http://lists.debian.org/505B820E.2040507@allums.com
>
>
--
--
Joel Rees
Reply to: