Ad
Ad
Ad
Pages: [1]   Bottom of Page
Print
Author Topic: New to scanning - need advice.  (Read 13867 times)
DonaldSmkn
Newbie
*
Offline Offline

Posts: 15


« on: September 30, 2010, 09:08:08 PM »
ReplyReply

Hi all,

I recently purchased a plustek 7400 35mm film scanner and use vuescan pro.

I have calibrated the scanner with a Wolf Faust it8 target (my film of choice is Provia 400x).

The main issue is not colour - colour is actually fine. The issue is inconsistency in image brightness and contrast.

I believe calibrating and then locking "RGB Exposure " (which I assume is gramma) at the exposure at which you have calibrated.

The problem is White and black points consistently change and I can not match the contrast of the original image.

I have had some luck by setting white and black point using the it8 target (using the defined white and black points on the chart) and then locking colours. I am still finding I am having to compress the far left and far right levels by about 20 "points" on each side in photoshop though.

Any advice would be appreciated.

FYI, I tend to like the colour casts that appear on film, so my only real issue is matching brightness and contrast!

Best Regards,
Don
Logged
Jonathan Ratzlaff
Full Member
***
Offline Offline

Posts: 194


« Reply #1 on: November 02, 2010, 12:56:34 AM »
ReplyReply

I have had the best results by setting the black point in the scanning software to approx .1% and the white to zero so there is no adjustment for highlight and contrast in the scan. A scan should look flat to ensure that all possible detail has been picked up by the scanner.

With scanning slide film in vuescan you set the colour to neutral and ensure that the exposure does not clip the highlights.  With transparency film I don't ever recall setting the actual transparancy type as my results were always worse than selecting just image. 

I take it that your monitor is calibrate; scanning transparencies this is more important than calibrating the scanner at least for my taste.

You need to determine whether the contrast you see in the film actually represents the scene.  Remember slide film has an exposure latitude of about 5 stops.  When you talk about white points, what white point are you setting to?  Same with black.  If you want to have detail in both the highlights and shadows you need to be judicious in what you set your black and white points at. 

Jonathan
Logged
Mark D Segal
Contributor
Sr. Member
*
Offline Offline

Posts: 6945


WWW
« Reply #2 on: November 04, 2010, 06:59:00 PM »
ReplyReply

I'm not sure I completely understand what you are doing, but I don't think "locking exposure" should be part of your process, because different images require different exposure. The profile simply helps to assure that whatever your exposure, what you see on your colour-managed display should be reproduced in the file generated from your colour-managed scanner, provided the pre-scan shows the impact of the exposure adjustment you make in pre-scan mode. Profiling scanners for the shadow areas can be difficult. I agree with Jonathan's point to avoid clipping highlights and shadows at the scan stage. It is very difficult to match the appearance of a slide and a scan of the slide because there is no luminosity management of the conditions in which you are viewing the slide outside of the scanner, unless you do a lot of work on the profile to get the result to match a very consistent slide viewing condition, such as from a light table whose lighting remains pretty constant. You may wish to consider downloading a demo version of SilverFast Ai6 for that scanner, obtaining one of their scanner profiling targets for the film of most interest to you and use it to generate a custom scanner profile using their Auto-IT8 feature. See whether that helps. It may be useful for you to read a couple of my recent articles on this website which describe this profiling process and its outcomes in some depth.
Logged

Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....." http://www.luminous-landscape.com/reviews/film/scanning_workflows_with_silverfast_8.shtml
langier
Sr. Member
****
Offline Offline

Posts: 639



WWW
« Reply #3 on: November 05, 2010, 02:05:03 PM »
ReplyReply

Unless all you are shooting are IT8 targets under controlled conditions, you'll probably need to set your white and black points image-by-image. Each image is usually a little different than the next unless you are working in a studio shooting the same subject and the same lighting.

Try running a pre-scan and set your black and white values and give it a try.
Logged

Larry Angier
ASMP, NAPP, ACT, and many more!

Webmaster, RANGE magazine
Editor emeritus, NorCal Quarterly

web--http://www.angier-fox.photoshelter.com
facebook--larry.angier
twitter--#larryangier
google+LarryAngier
Chris_Brown
Sr. Member
****
Offline Offline

Posts: 796



WWW
« Reply #4 on: December 12, 2010, 07:46:01 PM »
ReplyReply

There are two approaches to scanning film: set input parameters for each and every image, or, when using a profiled scanner and scanning chromes, set the scanner to the input parameters used when making the scanner profile & scan in 16-bit color (see info here), then making the appropriate adjustments in Photoshop.

When scanning film negatives, the DMax & DMin varies too much to use a single input parameter setup.
Logged

~ CB
ced
Sr. Member
****
Offline Offline

Posts: 267


« Reply #5 on: December 16, 2010, 07:15:35 AM »
ReplyReply

What you might try to verify is if the end points are being analysed with or without the image being cropped as this also plays a huge role in the result.
It would be best to crop the image then get the end points set.  This may also give you inconsistent results depending on the density of the darkest point in the image.
Then a greyscale alongside may help you.
Logged
Jonathan Ratzlaff
Full Member
***
Offline Offline

Posts: 194


« Reply #6 on: January 06, 2011, 11:14:15 PM »
ReplyReply

The main thing to worry about in a scan is to enure that as much data from the image is captured in the scan.  A scan from a negative should look really flat, as the Dmax of negative film is considerably lower at 2.5 than for example veliva which is somewhere about 3.8- 4.
For scanning slides, for my results, I have found that setting the software (vuescan) so that the clipping levels for highlights are at .01 and the clipping levels for shadows is 0.  That seems to give me the broadest dynamic range in my scanner. 
All final adjustments are done in the editing software.

The output may not be what you expect and with negatives it will be flat.  Images should be scanned at 16 bit; this is especially critical with negatives as that 2.5 dynamic range has to be expanded to give you the desired contrast.  There are two images, the basic raw scan and the second with rough adjustments.  The scan was from kodak portra 160nc so the negative contrast and saturation are quite low. 

Negatives are way harder to scan than transparencies as the slide gives you a reference point to work from.  A good scan from a slide should be slightly lower in contrast than the original scan, at least from my experience and depending on the original exposure.
Logged
Pages: [1]   Top of Page
Print
Jump to:  

Ad
Ad
Ad