I'm sort of non-plussed by this whole thread. Doesn't anybody read the
manual and learn what they are doing with new equipment? The exerpts
from various posts in the threads are in no particular order and
unattributed, but handily allow me to rant:
jking@xxxxxxxxxxxxxx wrote:
>Hi sorry yet more questions from me. I decided to buy a relativly cheap
>scanner. Minolta Dimage IV (3200dpi) on the principle I could not justify
>the cost of a 5400dpi film scanner. the aim was to scan some hundreds of
>print film negatives with as accurate and repeatable results as possible
>with minimum fuss.
>
A perfectly reasonable goal. And likely achievable with your scanner and
Viewscan. I can't speak for the Minolta software, as I've not used it.
The Canon software with my FS2710 is consistent but very limited.
>And at least in terms of the post-scanning software the color correction to
>compensate for the negative base is always going to present a challenge to
>color accuracy, whereas a slide is inherently neutral.
>
The scanner software should compensate for the film base color in the
scanning process. Viewscan has a great many film base profiles built in
and adds more all the time. A slide is not inherently neutral. The base
may be neutral (although Viewscan has base adjustment for slide filme
too.), but all slide films have color biases.
>I would agree with observations on this list that colours are less saturated
>and that the result is less sharp when scanned. (I have not tried manual
>focusing)
>
That is simply not my experience. For an extreme example I have posted
before, see this comparison. The appearance of the print is rather
accurately reproduced on screen (although larger than 4x6") on my
monitor and the other image is from a scan.
<http://www.geocities.com/dreammoose/Portra160NC/>
>the dynamic range of a slide is much greater than that of a negative, and it
>must therefore be more demanding of the scanner to reproduce that dynamic
>range.
>
To clarify, the range of densities on a slide is indeed slightly greater
than the range of densities in a neg. However, the neg records a wider
range of brightness from the subject, i.e. scene brightness is
'compressed' relative to slide film. Somehow in this thread people
started referring to the scanner in question as 'cheap' and acting like
it is a poor product. Is it not true that the Dimage IV was Minolta's
top model before the 5400? As a relatively recent design, it should have
sufficient dynamic range to deal with any film. My Canoscan 2720 dpi
scanner is old compared to the Dimage IV and I have no trouble scanning
slides and negatives. I just select the correct Media Type in the Device
menu and the correct film Vendor, Brand and Type in the Color menu and
both scan fine.
>My real hate of negative film is that without an accurate profile, it
>becomes subjective as to what is 'right'
>
Some leap of faith there. There is no technology currently available
which accurately records the whole light spectrum from a scene. All
films, negative or reversal, depend on imprecise color filtering into 3
or 4 layers of panchromatic emulsion. This process results in some
lovely images that may seem to accurately reproduce the original scene,
but that is simply not true. Take a bunch of shots at the same time with
identical (hah!) cameras and lenses and several different slide films.
Every slide will look somewhat different. Which is right?? Working with
color neg film simply addresses this directly, relying on the person who
creates the final image to create something that matches the memory of
the original. Sure, slide films are rather repeatable and consistent,
but 'right', I don't agree.
>Perhaps comming from a record playing hifi background I wanted a pure
>approach with little or no automatic correction and tinkering. I guess
>this is probably not practical.
>
Interesting. That record you play with no adjustment was adjusted like
crazy in the process of recording and mixing. Likewise, the prints you
have been getting are highly adjusted in the automated printing process.
Viewscan will let you do it either way or somewhere in between. But you
have to make a choice. If you want consistent results without learning
how to do a manual scan, but also without automated exposure and color
balance, you will be disappointed.
>With negatives on Vuescan, just to get you going, go to the "Colors"
>menu and select "White balance". That will work for many many images
>and get you something at least acceptable for almost all of them.
>
Yup, a good starting point. It will, however, give weird results with
highly monochromatic images. A pic of a green meadow with many
different shades in the trees scattered through it will come out pretty
far off. The Neutral setting is more reliable for many images.
>I'm sorry my version of Vuescan is not up to date. I hesitate to say more
>about it because it changes like the wind.
>
Why not? Updates are free.
>Ultimately, if you can get software that presents you with a histogram
>and allows you to set the dark and light points on it, you will have the
>control you desire.
>
So update your Viewscan, it provides histograms, allows setting black
and white points, different brightness curves and more.
>At the moment I only take colour prints and have never use slide film. I
>tend to use Fuji 100, 200 and occasionally 400 ASA film.
>
I use almost only color print film, although I lean to the Kodak side.
>I tried the software and can get consistent results from the **same**
>frame but **even** **more** **widly** **inconsistent** results between
>frames on the same roll. (I should mention the prints I got back from the
>film are fine).
>
Those prints that are fine are the result of automated exposure and
color balance adjustments in the printer.
>These scans where made with the *lock* *exposure* and *colour* *ballance*
>settings. so the software does not adjust the exposure and colour ballance for
>each frame individually.
>
So why are you locking them? I have never done that. If you are going
to do so, for whatever reason, you should take the first setting from a
very average image. If you get it right for an image which is not
representative of the others on the roll and lock the settings, the rest
will be wrong.
>So the quesition in my mind is, bearing in mind that the developed photos
>look fine. what is going on?????????????????????
>
>1. is scanner *correctly* seeing large differences between exposures on
>the same film which is being compensated for by the developer when they
>make the prints. This would imply either
> 1. My om 3's light meter is not consistent
>
Easily testable. Probably fine. Take a roll of slides to test both this
and the following question.
> 2. I am rubbish at controlling the exposure!!!!!!!
>
It is really easy to get a bit sloppy in exposure technique when using
print film and getting automated prints back. That printing equipment is
really good at its job of adjusting brightness and color balance. Some
of the folks who adjust the machines have 'interesting' ideas of how to
set them, though.
> 3. Its the nature of print film and not to worry
>
It is indeed the nature of print film to record a much wider range of
brightness than can be reproduced on print paper. Thus someone or
something has to decide how much of that range and what part of it to
put in the print.
>2. there are not large differences in exposure between shoots on the same
>film and that the scanner is not consistent. (however scanning the same
>shoot on different occasions with scan view give consistent results)
>
Sound like the scanner is ok.
>3. I should just let the ascanning software automatically set the exposure
>for each shoot and not worry about this at all.
>
For a mass project like you are undertaking, I would recommend some
experimentation with the software options to find an automation setting
that works well for several sample images. Then use that for the
project. Those few great shots (OK, if all your shots are great, you may
have a problem, but you have one with the prints too. :-) ) that don't
look their best can then be rescanned with custom adjustment. The
crucial thing is to get the overall histogram about right. Anything else
can be adjusted in the photo editor. If you scan in 64 bit, even
histogram errors aren't much trouble, as there is enough dynamic range
in the digital file to allow later adjustment.
>4. I should start taking slides and forget about print film because it is
>too inconsistent.
>
Obviously up to you, but consistant quality scans from print film are
quite possible and the greater range of subject brightness recorded on
color neg film is a definite plus. As a simple example, I've found that
the rather blank white skys common to prints of snapshots of people
outdoors actually have quite a bit of detail and color available on the
neg which can add a lot to the image. On slide film, those blown out
highlights are simply gone forever. I won't say that the images I've
posted in the various subdirectories of
<http://www.geocities.com/dreammoose/> are the greatest photography in
the world, but they are all scans of color negative film and I would
claim that they have good sharpness, contrast and color balance.
A wordy Moose
==============================================
List usage info: http://www.zuikoholic.com
List nannies: olympusadmin@xxxxxxxxxx
==============================================
|