Olympus-OM
[Top] [All Lists]

[OM] Re: RAW file by scanning

Subject: [OM] Re: RAW file by scanning
From: Moose <olymoose@xxxxxxxxx>
Date: Wed, 02 Jan 2008 18:14:30 -0800
Richard Ociepka wrote:
> Michael Wong wrote:
>   
>> I have found there may be saved as RAW file by Nikon Scan & Vuescan. Is
>> there any benefit for these RAW files?
>>     
First of all, I don't know that opening them in anything other than 
VueScan is useful, as any other converter won't know what to do with 
them. What they are is the direct output of the scanner, without any 
processing by VueScan at all other than to put them in TIFF format for 
storage. He calls them RAW, as they are in some ways like a camera RAW 
file, being the unprocessed scanner output. But they are quite different 
too, as they don't know anything about the color and exposure parameters 
of the film that was scanned.

As far as I know, their major advantage is in workflow. When scanning a 
roll of film, I scan to RAW first. That goes quite quickly, as VS 
outputs to file as it scans, so there is no processing overhead time, 
just one mechanical scan after another. Once I have the film all 
scanned, I then "re-scan" from the RAW files, previewing as a batch, 
making any adjustments needed to the previews, then scanning to ordinary 
TIFF and/or JPEG as a batch. The scanning, processing, writing takes 
time, but happily goes on in the background while I am doing other 
things, on the computer or elsewhere.

I find this much more efficient than adjusting each preview from the 
film directly, then waiting while VS does the full resolution scan, the 
second scan for IR channel, the image and dust removal processing and 
writing the files before I can go to the next frame.

The other advantage is that I can go back and "rescan" any frame or 
frames with different settings if I want to change something. So if most 
of the shots are under similar lighting and exposure, I may just set 
parameters for the first frame, which sets the default settings for all 
later frames, and do the batch scan without looking at each frame. If 
one or another frame needs different settings for best results, I can 
easily rescan from the file. I keep the RAW files until I'm completely 
finished working with that roll, then discard them. If I'm not entirely 
satisfied with any frames, and think changing the scan may help in the 
future, I keep those RAW files.
> The ability to adjust the color temperature, saturation and contrast.
>   
Weeeelll... Yes and no. No more ability to do so than in processing a 
TIFF file in ACR, PS, etc. It's not like a RAW file from a camera. The 
RAW converter can read from the camera file the color temperature the 
camera used in taking the shot, tell you what it is and allow you to 
adjust it directly to a known different temp.

With a film scan, the scanner has no way to know what the color temp of 
the film was. So if you open a scan output file in ACR, for example, it 
can let you adjust color temp relative to whatever is in the scan, but 
not to a specific temp. You still need a reference of some sort. 
Shooting a WhiBal or other black-white-gray reference will allow 
adjustment pretty close to true color, but is only good for that.

Creating film icc profiles(s) allows you to ignore color temp. Scan 
using a daylight profile and you get the color effect of other light. 
Magic hour light, for example, will come out correctly warm. If you want 
to correct light to look like daylight, make a profile for that light. 
So you may shoot at sunset, under mixed artificial light, etc. and it 
will look like daylight. With multiple profiles for different light with 
the same film, you can very easily adjust white balance when scanning 
from RAW files.

I don't know of any way to take film and accurately end up with a file 
that you can simply slide the color temp slider on to hit a specific 
temp. BUT - using icc profiles is actually more accurate. Camera sensors 
are highly color linear, so simply adjusting a slider works well. Film 
is not as color linear. The curves of the different color layers are not 
exactly the same, so correct balance at mid tones may not, and usually 
doesn't, guarantee correct balance in shadows or highlights. Building a 
table of corrections at many points of color, an icc profile, is the 
best way to color correct film.

Film and digital are very different technologies and simply require 
different techniques for optimal color management.

I've posted this many times, but I think it's such a good illustration 
of the value of color profiles for film. The first or top image in each 
pair is the output frm VueScan using Neutral WB setting. Perfectly 
decent scans that may be fairly easily adjusted in an editor. The second 
image of each pair is exactly the same except that I used an icc profile 
in the scan process. There has been absolutely no editor processing 
other than to downdample for web display. As you can see, not only is 
color corrected for WB, but the tonal curve is corrected.

This is the biggest time saver I know of in film scanning. Many, often 
most, images are ready to print/display without further work. AND it 
corrects the color better than anything else.

Moose

==============================================
List usage info:     http://www.zuikoholic.com
List nannies:        olympusadmin@xxxxxxxxxx
==============================================

<Prev in Thread] Current Thread [Next in Thread>
Sponsored by Tako
Impressum | Datenschutz