Digital BW, The Print

> * try what I may I cant get rid of the lack of smoothness between 
the 
> * tones in the cloudy sky section of a landscape shot. There's no 
point 
> * in posting a jpeg for you to see 'cos it looks good on screen 
even at 
> * 100% view but prints out like - i'm not even sure how to 
> * describe it - 'mottled' isn't exactly it but its close. This 
> * problem is most 
> * evident in the area where a darkish cloud texture meets the 
lighter 
> * sky texture.
> * 
> * I did my scan and most of the initial corrections in 16bit and 
then 
> * some in 8. I'm working with Win98SE and PS7 and an 1160 printer 
with 
> * MIS sepia>neutral inkset on photorag.
> * 
> * One symptom which may provide a clue for solving the problem, for 
> * someone, is that I had Layers for darkening/burning in  the top 
of 
> * the image and the corners. The problem was excessive. Then I 
> * eliminated each of those layers (lighteninig the image segments) 
and 
> * the problem reduced or just became less evident
> * 
> * I also have a High Pass sharpening layer - could that be it ?
> * 
> * I just looked at Image>Histogram and find it pretty "comb" like - 
is 
> * this it ? Is it 'posterization' ? Should I go back to the 16 bit 
scan 
> * and try to do more there before converting to 8 and printing ?
> * 
> Frank,
=============================================== 
> You can get away with some or a lot of "combing" in the histogram 
depending
> upon the nature of the image. In your current situation I would 
suggest
> applying your adjustments to a 16-bit version of the image and see 
what
> happens
> 
> If you still have the problem then you may not be getting a 
perfectly linear
> response with the RGB separation curves in the tonal range that is 
crucial
> to your sky. One way around this is to try a different separation 
curve or
> use two or more separation curve layers with different curves and 
different
> opacity (adding to 100%) so that the effect of the curve is 
blended. 
> 
> Beyond that you might want to take a look at the scan or scanning 
process to
> see it anything was done that might have damaged the data. Try 
printing the
> unadjusted scan file and see what you get.
> 
=============================================

I need a little more explanation on this "you may not be getting a 
perfectly linear response with the RGB separation curves in the tonal 
range that is crucial to your sky. One way around this is to try a 
different separation curve oruse two or more separation curve layers 
with different curves and differentopacity (adding to 100%) so that 
the effect of the curve is blended."

Are the adjustment curves I'm using to affect the contrast in the 
image what you've called "RGB separation" curves ? Or, Paul's curves 
that i apply at the end? 
============================================