Digital BW, The Print

> From: Tyler Boley <tyler@...>
> Reply-To: <DigitalBlackandWhiteThePrint@yahoogroups.com>
> Date: Tue, 07 Dec 2004 18:06:52 -0000
> To: <DigitalBlackandWhiteThePrint@yahoogroups.com>
> Subject: [Digital BW] Re: Tonal range and linearization
> 
> 
> 
> --- In DigitalBlackandWhiteThePrint@yahoogroups.com, Steve Kale
> <stevekale@b...> wrote:
>> Hi Tyler and others following along...
> 
> Well, at least I understand what you are getting at now, even though I
> quite adamantly disagree. But lack of consensus is better than lack of
> understanding.
> 
> Your classroom example has some problems... You assume if all three
> tones are properly exposed their relationships will be correct.
> Really? You've found a perfect straight line film/processing combo?
> The scanner you used was dead linear, or perfectly profiled profiled?
> Or perhaps you captured the scene. What were the characteristics of
> that device? How was the file processed? Anyway, different conversation...
> 

For these parts of the workflow I have colorsync to help me.  For the print
end I unfortunately don't.

> 
>>> The resulting tonal progression is something porportionally or somehow
>>> mathematically related to the reference, but not exactly the same rate
>>> of change. 
>>  
>> The existing relationship to reference is variable and changes for every
>> ink/paper combination.
> 
> Yes! Thanks goodness! They all have different characteristics.
> ...
> 
>>> ...The whole point is to maintain the tonal
>>> relationships, unless of course you want come up with your own unique
>>> definition of "linearization".
>> 
>> I am not defining linearization just using it...
> ...and...
>> I am saying that linearization to LAB should actually do that, not some
>> other function whose relationship to LAB changes for each and every
>> ink/paper combination.
> 
> OK, strictly yes. But the standard use of the term linearization for
> an output device these days is not that, and that's what I'm talking
> about. Even though the "linearization" results in a compression, the
> tonal relationships within that range bear a relationship to LAB from
> dmin to dmax, you must admit. Had one done a similar "linearization"
> to gamma 2.2 (or whatever) that same relationship would hold between
> the linearized output and 2.2, even though compressed. Strictly, the
> rate of change, is no longer exactly that of the target, granted.
> We are after two completely different things.

As I said in the note Roy, the way linearization is done at the moment
(straight-lining the LAB values) is analogous to creating a whole new colour
space. You have lost any consistent relationship to LAB.  A straight line
over 0 to 1 of L=16 to L=96 is completely different to 0 to 1 with L=5 to
L=96.  What's more the density curve generated by this is nothing like that
defined by LAB, the way the eye sees and all that good stuff.

>You want middle gray to
> be middle gray whether you are printing on a brown paper bag or
> Photorag, utilizing some (in this extreme example) rather massive
> change in the other tonal relationships, and those changes will likely
> be different between middle gray and dmin, and middle gray and dmax.
> I need all those relationships to be proportionally maintained,
> absolutely, and that is what I think is generally accepted as
> linearization, even though that moniker may be incorrect by strict
> definition.
> 

Well a brown paper bag is an extreme example and I wouldn't generally
recommend using it as a paper base for B&W printing. dMin is a pretty high
density and the available tonal range is actually quite small (likely much
smaller than EEM).  But yes it is capable of some tonal range - remember I
don't actually care that it is brown just that it has a low maximum
reflectance.  Whether mid grey is in-gamut or not would depend on how dark
the brown of the bag is.  If it is out of gamut I don't get it in the print
- period - and I would have to remap middle grey in my image to an in-gamut
value.  I probably wouldn't like the necessary compression and choose
another paper.

Cheers

Steve