[Openicc] beyond 8-bit precission [was: new version of xcalib]
Graeme Gill
graeme at argyllcms.com
Wed Mar 16 00:05:43 EST 2005
Kai-Uwe Behrmann wrote:
> What is the contrast ratio of film? Are todays LCDs and TV displays
> really that far from film contrast?
Contrast is a tricky subject. It depends a lot on how you measure
it. Most of the rather large numbers you see today are in
some senses a trick. You simply can't get the claimed contrast
ratios simultaneously (ie. measured with a checkerboard pattern).
Projected film has a very great dynamic range, and if contrast ratio
is measured this way (ie. full white screen compared to fully
dark screen), I gather some of the LCD's and other displays are
coming a little closer.
>>>Todays inkjets have left behind 256 steps - steps in the sense of
>>>stepping. Tell me if you like, I can send you an print, which shows
>>>clearly each step of 8-bit shades of gray output.
>>
>>Once again, I was referring to a carefully constructed space, not
>>a device space. The whole point of looking for RAMDAC tables with
>>greater than 8 bit entries is to handle the requirements of device
>>spaces with sufficient precision to be able to control them.
>
>
> Sorry but I dont understand. Both, the mentioned print and the LCD setup ,
> showed nearby each step in an 8-bit gray ramp. There is no space left to
> move intensities to the one or other end without degrading the used
> setup.
An inkjet printer will use a screen, and I can easily construct
a screen with as many levels as I want. For instance, if
I create a 256x256 stochastic cell, I get 1 part in 65536
control. (What I've done for some printing systems is to
construct the screen with the linearisation curves built
into the screen.) The same can apply for an LCD, where
internally the electronics driving the LCD cells has 12-14
bits of control (this is the case with some of the EIZO LCDs).
With such systems there is plenty of space between each 8 bit
controlled step, to set the brightness values for each step.
Graeme Gill.
More information about the openicc
mailing list