My vote would be to scan B&W neg as a single channel on the scanner.
Especially if it's a prosumer scanner. My reasoning would be the
optics and ccd alignment on these scanners are compromised thus the
"fringing". Also, the blue channel is usually a garbage channel
contributing most of the noise that the scanner will add. So, if
channel mixing, you would really want to zero blue out and fiddle with
the red and green. Ideally a scanner with a visual (salmon) filter
would be best, but I think those days are long gone.
The argument for scanning in RGB would be that you can finesse the
channels to get what you want. So, I guess you would have to evaluate
your hardware by scrutinizing a bunch of scans to decide if the extra
size, image processing and time are worth it. I operate a Hell S3900
drum that is about as well engineered a scanner as you can get, and I
don't waste a bunch of space and time fiddling with RGB from B&W negs.
I pick either the Red or Green channel.
Just my 2 pence.
Clark
On Jan 5, 2009, at 5:39 AM, Bill Morse wrote:
> I don't believe that's true. You can test this with a Channel Mix
> layer,
> just compare the conversion with and without.
>
> To the OP; if you have signifcant fringing, then it likely will
> affect the
> gray conversion as well as (though less than) the color image. That
> said,
> you're still much better off converting in PS than the scanner
> software, as
> there you have full control over the conversion process. Learn to
> use the
> channel mixer.
>
> Bill
>
> On Sun, Jan 4, 2009 at 10:36 PM, outdoornm <outdoornm@...>
> wrote:
>
> > Dale,
> >
> > ...Photoshop converts to grayscale directly in the
> > red channel. ie, Image>Mode>grayscale (100% red, 0% green, 0% blue)
> >
> > There are other ways to grayscale, using various values of red,
> green or
> > blue, but I would
> > suspect that scanners use the same 100% red channel method.
> >
> > Lincoln
> >
> >
> > - In
> DigitalBlackandWhiteThePrint
> @yahoogroups.com<DigitalBlackandWhiteThePrint%40yahoogroups.com>,
> > "dale40203" <dhoff@...> wrote:
> > >
> > > How does an RGB scanner determine Greyscale equivalents? Is
> there a
> > formula it applies
> > to
> > > the three components? If so, would the color "fringing" I see on
> some RGB
> > scans
> > contribute to
> > > a softening of Greyscale scans?
> > >
> > > The only way I can see to avoid this is for only one of the RGB
> > components to be used,
> > and I
> > > know that would not be an accurate representation of RGB
> translated into
> > GS.
> > >
> > > Thanks,
> > > Dale Hoffman
> > >
> >
> >
> >
>
> --
> Regards,
>
> Bill Morse
> Wm. Morse Editions
>
> http://www.MorseEditions.com/
>
> [Non-text portions of this message have been removed]
>
>
>
[Non-text portions of this message have been removed]