Ad
Ad
Ad
Pages: « 1 ... 3 4 [5]   Bottom of Page
Print
Author Topic: Using ACR after scan stage  (Read 11570 times)
Mark D Segal
Contributor
Sr. Member
*
Offline Offline

Posts: 6973


WWW
« Reply #80 on: January 31, 2013, 05:28:05 PM »
ReplyReply

Iíd be interested in your finer point, because what youíre saying doesnít seem possible.  On a related subject, I still donít understand why you believe that itís practical to work with linear data directly in LR.  Rather than sidetrack this thread with this discussion, however, Iíve started a new thread.  Iím always open to learning something new, so I hope you can explain where Iím going wrong. http://www.luminous-landscape.com/forum/index.php?topic=74747.0


Dean, I saw the new thread and this message. I intend to address all this comprehensively, but I don't have time this week and probably not next week either. Please bear in mind that all this discussion is a voluntary use of time which is fine and enjoyable when time permits, but for me right now this is not the case. I'm up to my eyeballs and beyond in real priorities and what I want to do for addressing this will take time I don't have now. I shall revert to it in my own way when I can and no number of reminders will change the timetable. I know it's out there and it will get attended in due course.
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
Alan Klein
Sr. Member
****
Offline Offline

Posts: 782



WWW
« Reply #81 on: January 31, 2013, 09:37:52 PM »
ReplyReply

Here are 45 year old Kodachromes scanned with a flat bed Epson V600.  I believe I used ICE with Epson scan followed with some spotting in Elements or LR3.  http://www.flickr.com/photos/alanklein2000/sets/72157626911395064/

Question I have about scan vs. camera RAW is that they both have CCD's.  IF you scan flat, wouldn't you be geting the same "raw" data in both cases?  Aren't you really limited by the film after all you can't get blood from a turnip.  What's there is there.  Unlike the original shot which has a wider DR than the CCD can handle, the film cannot be lighter than white or darker than black.  The film itself limits the DR.  No?
Logged
Tim Lookingbill
Sr. Member
****
Offline Offline

Posts: 1206



WWW
« Reply #82 on: February 01, 2013, 10:53:29 AM »
ReplyReply

Here are 45 year old Kodachromes scanned with a flat bed Epson V600.  I believe I used ICE with Epson scan followed with some spotting in Elements or LR3.  http://www.flickr.com/photos/alanklein2000/sets/72157626911395064/

Question I have about scan vs. camera RAW is that they both have CCD's.  IF you scan flat, wouldn't you be geting the same "raw" data in both cases?  Aren't you really limited by the film after all you can't get blood from a turnip.  What's there is there.  Unlike the original shot which has a wider DR than the CCD can handle, the film cannot be lighter than white or darker than black.  The film itself limits the DR.  No?

Not sure about sensor quality improvements between the two capture devices but this thread:

http://photo.net/digital-darkroom-forum/00b7Fk

...explains why it's the glass the data has to go through on a flatbed scanner that can impede a lot of fine detail from making it to the sensor. And detail includes color just as well as density.

Nice captures of those Kodachrome slides. Wonder how they'ld look in sharpness and detail capturing with the method used in the linked thread.

Not sure why you saved and upload to flickr in AdobeRGB so I had to drag and drop and view in their embedded profile in Photoshop. Just FYI.
Logged
digitaldog
Sr. Member
****
Offline Offline

Posts: 9189



WWW
« Reply #83 on: February 01, 2013, 11:36:17 AM »
ReplyReply

Question I have about scan vs. camera RAW is that they both have CCD's.

The big difference in my mind is one produces try, trilinear color and the other doesn't.
Logged

Andrew Rodney
Author ďColor Management for PhotographersĒ
http://digitaldog.net/
John MacLean
Full Member
***
Offline Offline

Posts: 169



WWW
« Reply #84 on: February 01, 2013, 01:02:16 PM »
ReplyReply

Not sure why you saved and upload to flickr in AdobeRGB so I had to drag and drop and view in their embedded profile in Photoshop. Just FYI.

Where do you see that info? I looked at the EXIF in Flickr and it's not showing.
Logged

dmerger
Sr. Member
****
Offline Offline

Posts: 686


« Reply #85 on: February 01, 2013, 03:12:28 PM »
ReplyReply

Question I have about scan vs. camera RAW is that they both have CCD's.  IF you scan flat, wouldn't you be geting the same "raw" data in both cases?

Scanner software and LR donít let you do real RAW processing, but you can get close.  The less you do to your scannerís CCD output, the closer to RAW.  For example, if you output aRGB, that is a step away from your scannerís native color space and a step away from RAW.  (Iím not saying that there is anything wrong with outputting aRGB, itís just an example of a difference from RAW.)
 
Aren't you really limited by the film after all you can't get blood from a turnip.  What's there is there.  Unlike the original shot which has a wider DR than the CCD can handle, the film cannot be lighter than white or darker than black.  The film itself limits the DR.  No?

No.  Depending on your scanner and film, itís possible that you wonít be able to capture all the highlights and shadows without clipping.
Logged

Dean Erger
John MacLean
Full Member
***
Offline Offline

Posts: 169



WWW
« Reply #86 on: February 01, 2013, 03:20:35 PM »
ReplyReply

Scanner software and LR donít let you do real RAW processing, but you can get close.

LR doesn't let you do real RAW processing? Please explain?
Logged

dmerger
Sr. Member
****
Offline Offline

Posts: 686


« Reply #87 on: February 01, 2013, 03:29:03 PM »
ReplyReply

... with the tiff output from his scanner.
Logged

Dean Erger
John MacLean
Full Member
***
Offline Offline

Posts: 169



WWW
« Reply #88 on: February 01, 2013, 03:36:20 PM »
ReplyReply

... with the tiff output from his scanner.

Ah yes, but of course. Sorry I wasn't reading it closely enough. DOH!
Logged

Tim Lookingbill
Sr. Member
****
Offline Offline

Posts: 1206



WWW
« Reply #89 on: February 01, 2013, 04:37:39 PM »
ReplyReply

Where do you see that info? I looked at the EXIF in Flickr and it's not showing.

My mistake. I was too quick when I opened it and didn't take the time to fully read the Photoshop CM policies dialog box stating there's no embedded profile and noticed the selected radio button was already set to choose AdobeRGB to assign to it. Sorry about that.

Still, the image doesn't have an embedded profile. Don't know if those viewing the Kodachrome on wide gamut display are going to see something different than intended. I'm viewing on sRGB-ish display so there's not much of a shift in color assigning sRGB in Photoshop.
Logged
Alan Klein
Sr. Member
****
Offline Offline

Posts: 782



WWW
« Reply #90 on: February 01, 2013, 05:50:29 PM »
ReplyReply

I don't know why there's limited EXIF data.  My medium format shots show a lot of info on the other hand.  The only thing I can guess is that i used another program to resize and the data was dropped.

Regarding clipping when scanning, I don;t see how that can happen.You would think that the scanner mfr adjusted the CCD circuits so that black slide would bottom out and clear film would provide data at the top of the histogram,  It's not as if in a real world situation the shutter/aperture limits the actual real world light at the time so you can clip.  The scanner knows the absolute maximum dark and light points.  Just set the circuit so you don;t clip.  I'll have to check back, but when I scan flat, the range seems to be in the middle, on the low end but not reaching clip points.
(edit:corrected spelling)  
« Last Edit: February 01, 2013, 05:53:15 PM by Alan Klein » Logged
dmerger
Sr. Member
****
Offline Offline

Posts: 686


« Reply #91 on: February 02, 2013, 07:27:35 AM »
ReplyReply

Scanner software and LR donít let you do real RAW processing, but you can get close.  The less you do to your scannerís CCD output, the closer to RAW.  For example, if you output aRGB, that is a step away from your scannerís native color space and a step away from RAW.  (Iím not saying that there is anything wrong with outputting aRGB, itís just an example of a difference from RAW.)

Just to be clear, so long as your output color space isnít smaller than your scannerís native color space, it probably makes no difference in LR whether you work with aRGB output from your scanner or the 16 bit linear output from your scannerís CCD (as long as you properly color manage that linear output).   My scannerís color space is larger than aRGB, so I use ProPhoto.  Itís likely that your Epsonís native color space is also larger then aRGB, but I donít know if that difference would be significant.

With my scanner, I can take the same 16 bit linear file and make two versions of it.  With one I assign my scannerís linear profile, but otherwise leave the file untouched.  Itís still in linear, which can be seen clearly from the histogram, even though the photo now looks normal.  For the other version, I go one step further in that I convert to ProPhoto.  In other words, I first assign my scannerís linear profile and then convert to ProPhoto.  Now, the photo not only looks normal, so does the histogram. If I then process both versions in LR, using all the same settings, the results are identical. 

I havenít done such a test using aRGB, so itís possible that using aRGB could make a difference (over and above just a gamut difference between my scannerís native color space and aRGB).
Logged

Dean Erger
jaapb
Jr. Member
**
Offline Offline

Posts: 94


« Reply #92 on: February 03, 2013, 05:30:33 AM »
ReplyReply

Just to come back at the scanning vs copying discussion. I have scanned many hundreds if not thousands of frames and it has been my experience that the major drawback of copying film is the lack of batch capabilities.
When copying you have to operate the setup continuously while with my Nikons 5000 and 9000 I can do other work while scanning (editing, writing this post) especially with the slide feeder.
I also have converted a sa-21 feeder for the 5000 to scan the whole roll of 35mm film (40 frames instead of 6) according to this procedure.

Furthermore a large part of my scanned and yet to scan archive consists of color negatives which I find difficult to copy, and color negatives somehow carry many imperfections which are no issue when scanning with infrared cleaning but need additional clean up when copied.
I know that high end scanners are worth their weight in gold nowadays but you may want to hold on to them if you still have of lot of film to go through. Obviously YMV, just my 2 cents.

Jaap
Logged
dmerger
Sr. Member
****
Offline Offline

Posts: 686


« Reply #93 on: February 04, 2013, 03:23:03 AM »
ReplyReply

With my scanner, I can take the same 16 bit linear file and make two versions of it.  With one I assign my scannerís linear profile, but otherwise leave the file untouched.  Itís still in linear, which can be seen clearly from the histogram, even though the photo now looks normal.  For the other version, I go one step further in that I convert to ProPhoto.  In other words, I first assign my scannerís linear profile and then convert to ProPhoto.  Now, the photo not only looks normal, so does the histogram. If I then process both versions in LR, using all the same settings, the results are identical.

Perhaps my test also provides an interesting observation about how LR processes photos internally vs. PS.  As noted previously, in PS my photo with my scannerís linear profile and as converted to ProPhoto look identical, but have very different histograms.  In LR, the photos not only look identical but also have identical histograms.  Both histograms look normal. Anyone care to say (or speculate) why this is so?  

EDIT:  Hereís another clue as to whatís going on.  If I do the same test, but with aRGB and sRGB, all four histograms look noticeably different in PS (Linear, ProPhoto, aRGB and sRGB).  In LR, the Linear and ProPhoto histograms look identical, but the aRGB histogram looks slightly different, and the sRGB histogram looks even more different.  Some of this difference with aRGB and sRGB is likely due to some clipping upon conversion to those smaller color spaces, which shows, particularly with sRGB, on the right end of the histogram. There are also some very small differences near the middle of the histograms, however, so maybe not all the differences are due to clipping. (This latest test was done with just one photo.  Perhaps other photos would show an even bigger difference.)
« Last Edit: February 04, 2013, 09:56:30 AM by dmerger » Logged

Dean Erger
Pages: « 1 ... 3 4 [5]   Top of Page
Print
Jump to:  

Ad
Ad
Ad