Re: Quality screen for colorimetry

m wrote:

There is much more scope for
adding differentiation there, than in the LCD panels themselves,

Of course. That's something that various companies do quite well.

So what's your point in claiming definitively that noone makes
LCD's with better than 8 bit D/A's ?

That's my whole point. The LCD panels are only good for a fixed set
of 256 levels. You can screw up these levels with bad
electronics...but there's nothing you can do to add levels or shift
where the transitions happen. They are fixed in hardware. Period.

I don't know what sort of LCD you're talking about. All the literature
I've come across indicates typical LCD's have a native analog characteristic,
and there no is such thing as an inherently 256 level LCD (unlike, say plasma
displays). What would be the mechanism ? Now I'm sure there are practical
issues with driving the LCD at higher gradations, but for instance the
article "1-Billion Color TFT-LCD TV with Full HD Format"
in IEEE Transactions on Consumer Electronics, Vol 51, No. 4, November 2005
only talks about the following issues with getting 10 bit output:
Drive system signalling speed, chip area of 10 bit D/A converters,
ensuring uniformity of image, and in the panel itself the problems are
listed as parasitic coupling of gate and data lines. There is
no mention of any mechanism inherent in LCD's that somehow
limits the grey levels possible.

Look. It's easy. Take one apart. Get the part number from the LCD.
Then visit either Samsung or LG Philips LCD page. Get the specs.
They are, with almost absolute certainty, stock 8 bit panels.

If that were the case, yes. But the claim is that they have created their
own electronics, so it won't in fact be a standard Samsung or LG Philips LCD,
even though it may be manufactured for them by such a company.

Not trying to put down a good company. I just think that it is
important for professionals to work from a foundation of facts rather
than the alternative. When I see people on S.E.C talk about a 4096
level LUT somehow magically making a stock LCD panel behave
better...well, it is not likely to be based in fact.

BTW, you can also characterize the monitors with a good
spectroradiometer and get the truth that way. I prefer Photo
Research, either the PR-650 or the PR-705.

Love to. Unfortunately I don't currently have access to a display
that claims to support such resolution, and it would take a little
reverse engineering of their PCI interface to figure out how to set
the Lut values inside the display.

Graeme Gill.
.

Relevant Pages

  • Re: How to find HorizSync / VertRefresh rates?
    ... completely meaningless with LCD panels. ... Well, the rates are both related to the video card, not the display. ... depend on the enforced horizontal sync and vertical refresh rates. ...
    (freebsd-questions)
  • Re: About 3D Imager
    ... Back when Kev was working the bugs out if the LCD glasses adaptor, ... of transmissive LCD panels. ... One thing we discovered during all of this was that the spinning disk ...
    (rec.games.vectrex)
  • Shortwave Radio RFI Noise ? : What About LCD and Plasma TV Displays . . .
    ... Which type of TV Display produces the least RFI Noise ... Horror Stories a -wrt- LCD and Plasma TV Displays? ... LG makes nearly all LCD panels (not the complete ...
    (rec.radio.shortwave)
  • Re: Quality screen for colorimetry
    ... Don't confuse the analog nature of the LCD pixel with the digital ... set in stone by the interface offered by the module OEM. ... with color LCD panels, the best you can hope for is 8 bits. ... to approach by a monitor manufacturer (it has to come form the LCD ...
    (sci.engr.color)