Switch to English Language Passer en langue française Omschakelen naar Nederlandse Taal Wechseln Sie zu deutschen Sprache Passa alla lingua italiana
Members: 70,514   Posts: 1,543,622   Online: 1113
      
Page 9 of 11 FirstFirst ... 34567891011 LastLast
Results 81 to 90 of 101
  1. #81
    heterolysis's Avatar
    Join Date
    Sep 2012
    Location
    Hamilton
    Shooter
    Multi Format
    Posts
    162
    Quote Originally Posted by wogster View Post
    How do you know it's the film? It could be your scanner, it could be your monitor, it could be the software rendering the image, it could be your video card driver, it's not always the film that is at fault. The best you can do is make an optical print, if you can't remove a cast in an optical print with a reasonable filtration, then blame the film. The best thing you can do, if you need it on the computer, is use the white balance feature in software and pick something white to balance on. If it can't correct the cast without making something else look weird, then blame the film.
    I could also be colour blind. I realize that there are well-known issues regarding the scanning of Ektar and getting an accurate colour profile, and it was not my intent to dwell on such issues. Can I colour correct the scans? Yes. Could I have used a good warming filter so that I didn't need to? Maybe. Could I filter it for an optical print? Probably---I don't have the capacity to print colour right now, so I can't know for sure. Hopefully someday.

    I never said that the colour was beyond saving, I was simply stating that from my experiences, Ektar is better in low light. I have shot it in all sorts of lighting, including in studio, and that's how I feel.
    The scans always come out blue---c'est la vie---but first and foremost, my point was that I don't like using Ektar in daylight situations.

    That said, being new here I didn't know if it was kosher to fiddle with the curves of a scan before posting a sample. I thought I had done my due diligence by admitting I knew they were off beforehand. Because, going back to the "lily" photos, you can tell right off that the photographer upped the yellow in the shot of the train in order to get the railings so vibrant, and it becomes very apparent in the grass as well as, as others pointed out, the sky. As that was such a contentious issue, I didn't want to play that blame game and just uploaded 'raw' samples.

    Here is a sample with, as far as my eyes, monitor and video card are concerned, his fur being its true colour.

    Click image for larger version. 

Name:	469150_10100262965357837_1717715859_o (3).jpg 
Views:	45 
Size:	711.0 KB 
ID:	58171

    I never said Ektar was terrible, and that was not my intention. Aside from daylight use, I personally think it has a lot of potential---great contrast and no 'grain'---and just wanted to point out a few issues I've had with it. Apologies if this was a bit of a rant, and I mean no offense by it, but I just wanted to clarify my position.

  2. #82
    Andre Noble's Avatar
    Join Date
    Dec 2004
    Location
    Beverly Hills, CA
    Shooter
    Medium Format
    Posts
    220
    Are you sure he doesn't look more like this:Click image for larger version. 

Name:	ektar cyan correct_b.jpg 
Views:	35 
Size:	771.5 KB 
ID:	58175
    Andre Noble, Beverly Hills California http://andrenoble.com/

  3. #83

    Join Date
    Nov 2011
    Shooter
    8x10 Format
    Posts
    20
    Guys, because of your technique is so weak you should ask a very professional photolab to make scientific tests for you (I mean optical prints from different c-films on the same paper, after shooting the same subject with the same camera and lens using the films that you're going to test).

    Anything else is just an amateur blablablabla....

  4. #84

    Join Date
    Sep 2010
    Location
    Midlands, UK
    Shooter
    Medium Format
    Posts
    919
    Quote Originally Posted by Bernard_61 View Post
    Guys, because of your technique is so weak you should ask a very professional photolab to make scientific tests for you...
    Agree. Either that or actually try spending time with colour photographs in books. You HAVE to look at photographs to develop a sensitivity to photographic colour. Your memory of the colour in the scene isn't enough because our brains process natural colour (light) and rendered colour very differently. If you have any well reproduced books, try looking at them just before editing one of your own photographs. That part of your brain really does need to be engaged sometimes. Landscapes are good for this because the photographer tends to strive for naturalistic results working with a broader spectrum of colour than a portraitist or... pet photographer. If you're editing on a computer, try stepping outside, looking around, then come back to the monitor to see how easy it is to confuse colour. Working in colour is perceptually challenging. It's not that anybody is colour blind, but obvious that people haven't spent much time seeing photographically, in colour.

    There isn't any sense of control or neutrality in the images posted here. They feel like experiments without reference.
    Last edited by batwister; 10-07-2012 at 07:16 AM. Click to view previous post history.

  5. #85
    wogster's Avatar
    Join Date
    Nov 2008
    Location
    Bruce Peninsula, ON, Canada
    Shooter
    35mm
    Posts
    1,266
    Quote Originally Posted by heterolysis View Post
    I could also be colour blind. I realize that there are well-known issues regarding the scanning of Ektar and getting an accurate colour profile, and it was not my intent to dwell on such issues. Can I colour correct the scans? Yes. Could I have used a good warming filter so that I didn't need to? Maybe. Could I filter it for an optical print? Probably---I don't have the capacity to print colour right now, so I can't know for sure. Hopefully someday.

    I never said that the colour was beyond saving, I was simply stating that from my experiences, Ektar is better in low light. I have shot it in all sorts of lighting, including in studio, and that's how I feel.
    The scans always come out blue---c'est la vie---but first and foremost, my point was that I don't like using Ektar in daylight situations.

    That said, being new here I didn't know if it was kosher to fiddle with the curves of a scan before posting a sample. I thought I had done my due diligence by admitting I knew they were off beforehand. Because, going back to the "lily" photos, you can tell right off that the photographer upped the yellow in the shot of the train in order to get the railings so vibrant, and it becomes very apparent in the grass as well as, as others pointed out, the sky. As that was such a contentious issue, I didn't want to play that blame game and just uploaded 'raw' samples.

    Here is a sample with, as far as my eyes, monitor and video card are concerned, his fur being its true colour.

    Click image for larger version. 

Name:	469150_10100262965357837_1717715859_o (3).jpg 
Views:	45 
Size:	711.0 KB 
ID:	58171

    I never said Ektar was terrible, and that was not my intention. Aside from daylight use, I personally think it has a lot of potential---great contrast and no 'grain'---and just wanted to point out a few issues I've had with it. Apologies if this was a bit of a rant, and I mean no offense by it, but I just wanted to clarify my position.
    I'm just sayin' a lot of people do a raw scan and if the colour isn't perfect they blame the film, rather then realizing that like optical printing, you need to colour correct the image before digital viewing/printing. One of the best ways to do this, and your dog allows for it, we know the kerchief is white and red, so if we white balance on the white part, we should get the other colours to come along, close enough. Barring the fact that no two films see colour exactly the same way, in fact two batches of the same film will see colour slightly differently, no two papers see colour the same way, no two scanners see colour the same way, no two printers see colour the same way, no two monitors see colour the same way. The closest we can come is, is the colour on the print or screen close enough to seem a reasonable facsimile of the original scene, given the fact it will never be exact. Even though there are people who worship digital capture, it's not going to be 100% exact either, no two sensors, unless from the same wafer, will see colour exactly the same, and they will not be 100% accurate either. This is probably why I often shoot colour for B&W, in other words, I use a colour film or d*****l camera, then in post processing, convert to B&W using the computer.
    Paul Schmidt
    See my Blog at http://clickandspin.blogspot.com

    The greatest advance in photography in the last 100 years is not digital, it's odourless stop bath....

  6. #86
    Diapositivo's Avatar
    Join Date
    Nov 2009
    Location
    Rome, Italy
    Shooter
    35mm
    Posts
    1,844
    Quote Originally Posted by heterolysis View Post
    That said, being new here I didn't know if it was kosher to fiddle with the curves of a scan before posting a sample. I thought I had done my due diligence by admitting I knew they were off beforehand. Because, going back to the "lily" photos, you can tell right off that the photographer upped the yellow in the shot of the train in order to get the railings so vibrant, and it becomes very apparent in the grass as well as, as others pointed out, the sky. As that was such a contentious issue, I didn't want to play that blame game and just uploaded 'raw' samples.
    The problem is that a negative scan can never be output "raw" (that would be negative). The process of inverting the masked negative always inescapably involves some filtering somewhere in the process. When one posts a scan as it comes out "right out of the scanner" or "right out of photoshop" etc. there always is some sort of filtering applied. Software involved in the process tries to filter out the orange mask (but not all orange masks are equal) and tries to white balance the image by applying some strategy, typically assuming that there is some white in the image, make the most quasi-neutral patch properly neutral, and hope that the rest of the colours falls in the right place.

    This strategy most often fails because the quasi-neutral was not neutral at all, or it was white or grey but it was quasi-neutral and bringing it to neutral sets a cast on the image (asphalt is not neutral, and even white cars are not neutral etc), or because the neutral patch was in the wrong place (white car in shade, rest of the scene in sun light) or because the white patch was tinted by a coloured reflection (white card hit by green reflection from the grass) and probably many other reasons.

    You can see it with the dog scans. Although the images were probably taken at short time interval and in the same light conditions the "raw" filtering of the two images is evidently different. The first is probably skewed toward red-magenta and the second toward yellow or green or cyan. That means two different filtrations have been applied by some layer of software, as the film, no doubt, is the same.

    So at the end of the day there is no such a thing as a "raw" negative scan. The scan as it comes out of the scanner is normally quite wrong (if there is no proper colour management involving scanner, film and monitor profiling). People post all over Flicker these raw scans and scream that the film is not good. Or they say it's good and the scan screams it is not. And the same scan is seen differently by different people in any case.

    Add to that several other layers of imprecision (lack of scanner calibration, lack of monitor calibration, lack of indication of reference space such as sRGB or AdobeRGB) and the final result on monitors is a fair of random colour casts.

    And we cannot discuss what could be addressed because, you know, it is taboo on this site.

    So my word of wisdom is: any attempt to appreciate the colour rendition of any colour negative film with a non-colour-managed workflow is moot.
    And also my intervention in this thread is moot because we cannot talk hybrid technology here.

    APUG is destined to forever host threads of people complaining about the colour rendition of their scan (Ektar or whatever) or praising a colour negative instead of another by posting, again, negative scans with "random" filtration.

    I have to absolutely make a decision and observe it to never any more intervene in threads discussing colour rendition on APUG.
    Fabrizio Ruggeri fine art photography site: http://fabrizio-ruggeri.artistwebsites.com
    Stock images at Imagebroker: http://www.imagebroker.com/#/search/ib_fbr

  7. #87
    wogster's Avatar
    Join Date
    Nov 2008
    Location
    Bruce Peninsula, ON, Canada
    Shooter
    35mm
    Posts
    1,266
    Quote Originally Posted by Diapositivo View Post
    The problem is that a negative scan can never be output "raw" (that would be negative). The process of inverting the masked negative always inescapably involves some filtering somewhere in the process. When one posts a scan as it comes out "right out of the scanner" or "right out of photoshop" etc. there always is some sort of filtering applied. Software involved in the process tries to filter out the orange mask (but not all orange masks are equal) and tries to white balance the image by applying some strategy, typically assuming that there is some white in the image, make the most quasi-neutral patch properly neutral, and hope that the rest of the colours falls in the right place.

    This strategy most often fails because the quasi-neutral was not neutral at all, or it was white or grey but it was quasi-neutral and bringing it to neutral sets a cast on the image (asphalt is not neutral, and even white cars are not neutral etc), or because the neutral patch was in the wrong place (white car in shade, rest of the scene in sun light) or because the white patch was tinted by a coloured reflection (white card hit by green reflection from the grass) and probably many other reasons.

    You can see it with the dog scans. Although the images were probably taken at short time interval and in the same light conditions the "raw" filtering of the two images is evidently different. The first is probably skewed toward red-magenta and the second toward yellow or green or cyan. That means two different filtrations have been applied by some layer of software, as the film, no doubt, is the same.

    So at the end of the day there is no such a thing as a "raw" negative scan. The scan as it comes out of the scanner is normally quite wrong (if there is no proper colour management involving scanner, film and monitor profiling). People post all over Flicker these raw scans and scream that the film is not good. Or they say it's good and the scan screams it is not. And the same scan is seen differently by different people in any case.

    Add to that several other layers of imprecision (lack of scanner calibration, lack of monitor calibration, lack of indication of reference space such as sRGB or AdobeRGB) and the final result on monitors is a fair of random colour casts.

    And we cannot discuss what could be addressed because, you know, it is taboo on this site.

    So my word of wisdom is: any attempt to appreciate the colour rendition of any colour negative film with a non-colour-managed workflow is moot.
    And also my intervention in this thread is moot because we cannot talk hybrid technology here.

    APUG is destined to forever host threads of people complaining about the colour rendition of their scan (Ektar or whatever) or praising a colour negative instead of another by posting, again, negative scans with "random" filtration.

    I have to absolutely make a decision and observe it to never any more intervene in threads discussing colour rendition on APUG.
    I think you are right, what comes out of the scanner is most likely wrong, unless it has all been calibrated. There are three steps to this:

    1) Calibrate the scanner.
    2) Calibrate the monitor
    3) Calibrate the film.

    To calibrate the film, is easy, take a blank piece of film, say the piece before the first image, this is usually unexposed. This should appear in a scan as white, since there is no image. You colour correct this to get white. Now if you scan a negative using the same correction, it should come out right, if it does not, then it's the colour attributes of the film that are inaccurate.

    Without doing all of these steps, you can't say for certain whether incorrect colour is the film or the scanning process.
    Paul Schmidt
    See my Blog at http://clickandspin.blogspot.com

    The greatest advance in photography in the last 100 years is not digital, it's odourless stop bath....

  8. #88
    Diapositivo's Avatar
    Join Date
    Nov 2009
    Location
    Rome, Italy
    Shooter
    35mm
    Posts
    1,844
    Absolutely yes.

    Calibrate the film with the "lock film base" method is a first attempt but, as far as I know, it's not optimal, and that is if I get it right because the orange mask is not uniform (like an orange filter) and does not affect all image zones in the same way.

    The method I know - but that I still have to put into practice - is actually to take a picture of a IT 8.7 (ISO 12641) calibration target and then pass this image, together with the reference file given with the target, to an application that creates a profile.

    Actually the standard method would call for the use of a reflective calibration target, but I plan to use a transparent calibration target as I think it should give the same results (and it is easier to set up).

    So my plan is to buy a slide duplicator, put a slide IT8 2.7 target on it, take a picture of it with a negative film, let's say Rollei CN200, develop the film, scan the image, than use this image for the "film calibration process" and obtaining a film profile, then carefully inspect my workflow so that the film profile is honoured (not stripped and not ignored) in every stage of the workflow.

    An explanation of one such method is given here:

    http://www.hamrick.com/vuescan/html/vuesc18.htm#topic15

    Another important requisite is that all the workflow (scanner, post-processing software, monitor) use and honour ICC profiles or the entire pain will have been moot*.

    That means that APUG readers are supposed to have calibrated monitors and to have browsers which (are configured to) honour ICC profiles!

    http://www.gballard.net/psd/go_live_...profiles.html#

    * As the gballard site explains, the de facto standard colour space being sRGB, an image without ICC profile can be seen by all "exactly" if the browser considers any web image as being in the sRGB colour space AND also converts it to your monitor profile. So ICC profiles can be absent from web images provided that the images are sRGB and that monitors are calibrated at both ends.
    Last edited by Diapositivo; 10-07-2012 at 02:27 PM. Click to view previous post history.
    Fabrizio Ruggeri fine art photography site: http://fabrizio-ruggeri.artistwebsites.com
    Stock images at Imagebroker: http://www.imagebroker.com/#/search/ib_fbr

  9. #89

    Join Date
    Jul 2011
    Shooter
    8x10 Format
    Posts
    2,601
    A lot of garbage-in/garbage-out shoot-from-the-hip testing mentality here. Exactly the kind of thing which tells one next to nothing relevant about the film itself.

  10. #90
    Andy K's Avatar
    Join Date
    Jul 2004
    Location
    Sunny Southend, England.
    Shooter
    Multi Format
    Posts
    9,422
    Images
    81
    Just out of curiosity, why is scanning being discussed on APUG?


    -----------My Flickr-----------
    Anáil nathrach, ortha bháis is beatha, do chéal déanaimh.

Page 9 of 11 FirstFirst ... 34567891011 LastLast


 

APUG PARTNERS EQUALLY FUNDING OUR COMMUNITY:



Contact Us  |  Support Us!  |  Advertise  |  Site Terms  |  Archive  —   Search  |  Mobile Device Access  |  RSS  |  Facebook  |  Linkedin