> I don't quite understand that.... What exactly do you mean? Not saying
> you're at all wrong, just want to understand it (the bane of my
> existence!).
I meant that PS seems to do less damage to the data, ie rounding
errors and consequent drop-outs, combing etc, than OE scanner software
does. Or than some lesser imaging programmes do.
I don't know why this should be the case, or even whether it definitely is, but
that is how it looks to me.
> A histogram is nothing but counting the values, and displaying them...not
> really an algorithm...
Quite so. But evidence of this presents via histograms (is what I meant)
> If you mean PS somehow does gamma correction
> 'different' than I'd like to explore that.
Yes, that's how it looks.
> Gamma correction (ie, curves)
> is only 'mapping' the input values to new output values...that's all there
> is to the algorithm....
Well, I am speculating, but it is possible to imagine that further processing
is possible - eg look for anomalous isolated spikes caused by rounding errors,
and try and eliminate them somehow.
All I am saying, and this is tentative, is that PS seems not to quite make
such a mangled mess of 8bit operations as OE s/w often does.
Regards
Tony Sleep
http://www.halftone.co.uk - Online portfolio & exhibit; + film scanner info &
comparisons
