Best to scan in 48 Bit HDR? Or use 48 Bit + modify during scan?

Hello. I am new to scanning and trying to understand the best workflow
to scan and edit photos. I've read a *lot* on the web and have become
more and more confused.

My question is:
I already have SilverFast Ai, but do not intend to use SilverFast HDR.
Is it best to scan images in 48 Bit HDR Color and then do the
processing with Aperture (I already have a copy) or Photoshop (don't
have a copy yet) later? Or should I really be using SilverFast HDR?

I am having a really really tough time figuring out what workflow I
should be using which will give me the most flexibility - darn
confusing stuff to a new scanner. I'm sure I'll get it eventually!!

New to scanning, I am guessing that scanning in a HDR format will allow
me to retain as much information from the original scan as possible,
and then have the flexibility to modify the file as I wish later on (or
leave the editing to someone else). For some reason HDR seems like a
better idea than scanning in at 48 Bit Color, with corrections made
during the scanning process, which seems would result in the loss of
some information that later I may wish I had. I don't know if that is
true or not, however. Additionally, I would prefer to scan all my
photos in now (without spending a lot of time manipulating the images),
though I may not use all the files in the future.

Thank you.

P.S. Oh - I should add that the resulting scans (using the 48 Bit HDR
Color setting) look so dark that it still blows my mind that I will be
able to make as many edits to the image as I would if simply using 48
Bit Color and adjusting the image as I go...

 

I personally use Silverfast during scanning to get the best possible scan
and save it to disk. I then use Photoshop CS2 to finally tweak the image.

 

To learn about scanning and how to's go to:
http://www.scantips.com/

 

Set the box gamma for HDR to bring up the mid tones. Otherwise, use
curves in whatever application to do the same thing.

output for HDR is a "raw scan", but you still need to check the exposure
and possibly set the white and black point to use the most of your
scanners dynamic range and capture as many tones as possible.

 

As you mention - is it required that I bring up the mid-tones (I
presume this is why it looks so dark when I scan) from within
SilverFast? Or if I just wish to scan quickly, I can always leave that
for later work.

It sounds like checking the exposure from within SilverFast is
mandatory to get the full range, and that I couldn't just assume the
scanner gets it right and I can correct the exposure from within
another application. Is that right?

 

i have not checked to see if when i check "gamma for HDR " and set a
value other than 1, that PS or other apps actually can use it. i use a
HDR workflow, so i have never tried it that way. Certainly, there is no
issue applying a curve later after the raw scan.

Depends on how picky you are - there is no "correct" answer, only
different approaches. i have done it both ways - sometimes rescanning
when i was not happy with the auto results.

i have always been happier with images i manually scanned for final
quality, it just takes longer. On the other hand, for most images i can
batch scan and accept the auto setting and move on. Depends on what you
are happy with.

 

If you do not have Silverfast HDR (or DCPro) you do not want to scan in
48 Bit HDR Color mode.
This will just give you a dark scan that you'll have a heckuva time
trying to get right in Photoshop.
The setting you want to use is "48 Bit Colour."

The Silverfast HDR program is great if you want to do batch scanning
and correct the pics later.
Of course, you can still do simple batch scanning with Silverfast Ai,
but you'll be stuck with whatever settings you have chosen for that
batch.

I can get great results with Silverfast Ai and my Minolta Scan Elite
5400, but it is slower than molasses and very tedious work.

 

I forgot to ask. Do you (when doing 48 bit color, not 48 bit HDR) use
the histogram and then select the shadow point and highlight point (the
triangles on the far left and right of the histogram), or do you use
the Highlight-Shadow-Tool (the 2nd icon from the right in the main
SIlverFast box, where you select white and black points.

I've heard both mentioned on the internet and am wondering your
opinion.

 

Just this week I was faced with a very difficult scanning assignment.
This one particular slide was almost totally red. I figured I'd just
fix it in Photoshop, but I could not get an acceptable result that way.
Only the Red channel was usable.

So I rescanned it using Silverfast's "48 Bit HDR Colour" mode and then
used DCPro to get a usable pic. I only have the Demo version of DCPro
which puts little watermarks all over the pic, so I had to make
multiple scans and mask out the watermarks.

Here is the scan in question.
http://www.pbase.com/image/73093698

The image on the left is the original scan without adjustments.
The image in the center was the version I settled on.
The image on the right was my "artsy-fartsy" version.

 

One more question to tack on to my last -
if scanning as TIFF using 48 bit HDR, am I going to "lose" any
information that is not recoverable from another application later on
(eg Photoshop) if I only select the white and black points (or select
the left and right "end" points of the histogram, I assume it is an
either/or choice).