Strategy for scanning complete 35mm negative collection

  • Thread starter Thread starter redtag
  • Start date Start date
redtag said:
Guys,
I have a Nikon Coolscan IV and have played succesfully with scanning my
35mm negatives. Works a treat. However I have several years worth of
negatives to get through!
Can someone share practical advise on how to go about this? I was
thinking of using the scanner as a catalogue device and just save a
thumbnail of all negatives, and correlate them to the negatives. I
would scan a negative, save the thumbnail and file the negative.
Then I can go through the thumbnails and select the pictures I want to
scan. More importthatal pictures as well.
Any thoughts on that? Can I save just the thumbnails on the Nikon?
Should I just scan all the negatives first time round - will take a
looonnngggg time :-)
Cheers

I'll throw my own thoughts onto the thorough advice you've already
received. As Preston Earle wrote, the main factor in scanning
thumbnails vs. originals is the expected ratio of keepers to discards.
If you'll only keep two or three per film roll, you'll probably save
some time by building your thumbnails before reloading for the final
scan. Otherwise, your scanner is supposed to be quick enough that
you'll probably lose more time just by reloading the film a second
time. :)

I don't know NikonScan. If it re-scans even when thumbnailing, you
might save a bit more time by hitting PrintScreen on your keyboard with
all of the NikonScan thumbnails visible, then Edit|Pasting the
screenshot into your image editor, where you can crop and save in short
order. :) That's out of the question if NikonScan can produce speedy
auto-thumbnails or doesn't show all images at once in its preview.

It also depends a bit on how much disk space you have. You might start
to run out if you scan every single image at full size. Whether that
means pruning via thumbnails, scanning fewer/smaller files, or buying
more storage space is up to you.

The rest of this post might be things you're already well familiar
with. I apologize in advance if so. :)

Preston also recommends scanning below max resolution. Pay extra
attention to the *detail* present at the higher resolution and see if
it's any better than that of the lower res. Try resizing a lower-res
scan to the pixel dimensions of the max res scan and comparing them.
I've seen plenty of P&S scans where the original image is blurry enough
that anything above 2700ppi (or much less!) is just grain. On the
other hand, if you have the disk space and enough processing power for
the largest scans, you aren't losing anything by scanning large. One
possible benefit of big images is that noise removal programs such as
Neat Image might give you better and easier results (once they finish
processing the huge file...) when all the grain is smaller than the
smallest image details.

As for scanning into sRGB...doing so will compress the original colors
of the image into a range suitable for unadjusted display on your
monitor. If you're happy with the result, as well as your printed
output, it's a safe choice. However, scanning into AdobeRGB or a wider
color space (I don't know how wide for negative film, though I've seen
EktaSpace and ProPhotoRGB floated around) will preserve more of your
film's overall color range, which is good if you want to take advantage
of the added color range of a high-quality printer or hypothetical
future display technologies.

If you scan into anything other than sRGB, you need an image editor
capable of color managment (such as Photoshop, Photoshop Elements 3&4,
Picture Window Pro, or Paint Shop Pro X). The editor needs to be
configured to use the same color space (AdobeRGB, sRGB, etc.) as the
image, or to convert the image into the current space. If at all
possible, set NikonScan to embed your chosen color space/color profile
into the image when you scan. If you want your image to look correct
on a web page, in a non-color-managed word processor, in a Windows
viewer, or in any other non-color-managed environment, you'll need to
convert a copy back to sRGB. Otherwise the image will look flat and
desaturated. Bear in mind that color management support is improving,
so that future web browsers and Windows programs might display an
AdobeRGB image properly without adjustment.

Your choice of color space is important because you can't go back and
change it later (well, you can, but it won't restore any lost color
range). Color management is a complicated topic, so if you're
perfectly happy with what you're doing now, don't worry about it. :)
Although...if you're going to be tweaking your negatives to get the
colors just right, consider investing $200-300 in a hardware
calibrator/profiler for your monitor (GretagMacbeth EyeOne Display 2,
Monaco Optix XR, ColorVision Spyder2) or at least running the free
software calibrator Adobe Gamma (or possibly Paint Shop Pro X's
built-in Monitor Calibration Wizard, but I can't vouch for its
accuracy). If you calibrate and profile your monitor properly, then
set up your photo editor to use the proper monitor profile, you can be
more confident that your screen is showing you your image and your
edits the way they're actually stored in the file, so that you won't
have to re-correct all your files at some point down the line. :)

Whatever you end up doing, back up your initial scan files (TIFF is a
fine choice for max compatibility and lossless storage) and make any
edits or corrections to copies. Then you can always go back to the
original if necessary.

Have fun reliving your photos. :)
false_dmitrii
 
Roger said:
On 5 Oct 2005 22:24:46 -0700, "cubilcle281"


I don't think you will find scans to be RAW in the sense of RAW from a
camera. Second, it's a very good idea to do any dust and scratch
removal at scan time using ICE which shouldn't degrade the image
quality. I had some problems with early versions, but every thing in
the last year and a half has worked very well. ICE works far better
than any post processing I've seen.

ICE _is_ a post processing step that happens not in the scanner but
in the PC. The scanner just delivers four channels of data, red,
green, blue and infra-red. And cubicle suggest to save exactly this
data, including the IR channel, before applying ICE-algorithms.
I suppose he things on Vuescan, when suggesting this.

At least thats what I am doing for my favorite photos: Saving the
raw files generated by vuescan. This allows me in a few years to
take the future algorithms to interpret the IR data even when the
original slide should be damaged in the meantime.
Tis true it's *relatively* cheap, but at max res those files are
*big*.

Yes. I am sorry, that's right. Therefore, I keep the raw scans only
for my favorite slides. For all other, I only keep "thumbnails"
with half the resolution and downsized to 8 bit per channel...

Best regards,
Kurt.
 
Back
Top