Digital BW, The Print

> Your  "histo-wrecking" comment makes me wonder if it is not 
> better to back up and do things in the color channels before I 
> convert to greyscale. 


Lincoln,

if you start with rgb scans, and follow proper color management, you have the option 
of doing a lot of work in that mode before dropping to gray. My point was to make 
you aware that going from rgb to gray is essentially a profile-to-profile conversion, 
and choosing each profile carefully (in the color settings of PS)  can improve the histo 
of the final file.

You don't _have_ to work in rgb of course, and people have preference for other 
options such as 16 bit gray (dropping into 8 eventually), or picking one of the rgb 
channels etc.



  I recently tried to make "improvements" in 
> the Tone Curves of my scanner prescans, but only managed to 
> change my beautifully smooth r, g's and bees into  storm 
> wreckage. :-( 

Starting with a good scan is essential. Depending on the scanning software you use, 
you may be better off scanning in 16bit and taking any corrections into Photoshop. 
Often the problem is poor or not color-managed preview in the scanning s/w. If you 
can get a correct preview (i.e. a big enough image on screen that looks identical 
between the scanner s/w and Photoshop), you can do all the main adjustments there 
and get away with 8bit grayscale scans. As an example, Imacon or Linocolor software 
allows for that. If not, you are best to go to plan B and get a high-bit scan into 
Photoshop and work from there.


Those 
> wrecked histograms don't necessarily show in the monitor 
> image, but I can imagine that a large print  will bring any 
> weaknesses into view.


The monitor is a poor judge of final print quality. At best it gives you a preview of 
how your overall tones will reproduce (in brightness and contrast). All else is mostly 
left to your experienced judgment, eyedropper "reality check", histo display 
(preferably of selected sensitive areas), views at different magnification % etc. 
We only wish we could see on the monitor what something will look on paper. Even 
with the best of color management and the most expensive of Barco monitors, our 
eyes will simply never see a phosphorescent white as being the same as a piece of 
paper, even if they are technically matched. Hence the need for actual proofs - the 
more experienced the user the fewer needed, because the real "match" happens in 
our brains.

And, you are right, you've got to try things out for yourself and see what works best - 
just be methodical about it because it's easy to get lost in a whole bunch of variables.

Happy Holidays to you too!

Antonis