Skip to main content

⚠️ Please note that this topic or post has been archived. The information contained here may no longer be accurate or up-to-date. ⚠️

Preview saturation

Comments

9 comments

  • NNN636858814109688210
    OK so I did a little test.

    I downloaded RAW files of other camera companies that are available online.
    I tried Sony, Nikon, CR3 of Canon EOS R and just to be sure some review raw files of my camera also.

    I imported all the files to a new session (import format: eip)

    Nikon (Z7) and Sony (A7iii) also on higher ISO (6400) didn't show my problem.

    All files from Canon CR2 and ALSO CR3 had the same problem.

    The difference is easiest to see on skin tones but generally on any warm color. Apparently C1 does something strange with Canon RAW format.

    Are here some other Canon users who could post about their experience?
    To perform the test, set your image cache so that you can toggle between preview proxy file and the 'live rendering' by zooming in and out.

    My camera is only few weeks old and I'm already thinking about getting a new one 😊 Does it ever change as one gets more professional? 😉
    0
  • SFA
    Some people seem to dislike the standard "warmness" of default Canon previews or even the entire look is thought to be too 'warm' by some.

    Yes there can be some interesting results comparing rendition at preview resolution (matching the native resolution of you screen presumably) where many pixels will have been binned to cut things down to the preview pixel size, and a 100% view where all available pixels (after lens correction if required) will be in play.

    Running a 1920x1080 screen I find any notable colour change from zooming comes in at around 50% zoom for my files - but my kit is likely older and so small image dimensions than yours if you bought new and recently. For detailed colour work it is always best to work zoomed in in my opinion.

    Also you can do your own thing.

    As loaded the Base Characteristics will choose a default ICC profile and pick a "Curve" that is intended to look good for the majority of users and sort of emulate something a bit more like a "film" response than a flat digital file of data.

    If you don't especially like that you can change it.

    Change the Curve to "Linear" and you will see what is basically the data, interpreted into an image that is somewhat 'flat'.

    You can make your own changes to the file using all of the usual tools to make it into something you are happy with and then you can save that for future use and, perhaps, set is is your preferred default for the images from that camera. (I'm assuming you are working with RAW files - if not you probably should be.)

    You might, if you are particularly keen to fine tune things, have multiple interpretations available for deployment according to, for example, subject matter. However that would be a much larger discussion to cover and in any case such things are usually covered in one or more of the on-line tutorials and recorded webinars and I would recommend that you take in some of those and use the context sensitive links to on-line help to get a good backgrounder knowledge in place. You may well find that most of your questions are answered to at least a working level of understanding onto which the forum population can add specifics and experiences.

    FInally let me say that Canon files, out of cameras jpgs and all, at default settings have always tended to look a little "warmer" than, say, Nikon (perhaps more towards green and 'flatter') or Sony (always seem to have very mid blue mid blues, notably skies. However those are generic observations and individual images will not always conform to such generalisations.

    HTH.


    Grant
    0
  • NNN636858814109688210
    Hallo again!

    Thank you so much for your response.

    I am not sure if I was understood properly. The following paragraph seems to describe a bit what I mean:

    Yes there can be some interesting results comparing rendition at preview resolution (matching the native resolution of you screen presumably) where many pixels will have been binned to cut things down to the preview pixel size, and a 100% view where all available pixels (after lens correction if required) will be in play.


    But then you follow with recommendation to use features of C1 like ICC Profiles and so on.
    The thing is I don't dispute the warmness of C1 RAW files rendition or differences to other software. I like how C1 sees my photos out of the box.

    The thing is I am not able to properly judge what all those features do, when I'm looking at the whole image.
    I agree, one cannot judge noise reduction at 25% zoom. But a color shift is pretty noticeable.

    I'd like you to run a quick test to see what I mean. I tried it on few computers and the results are same, so I suppose you'd be able to get what I mean and judge for yourself if that amount of color shift between proxy preview and 100% picture is acceptable.

    So that you can see same image I found some review with raw samples. Here they are:


    Scroll down to the list of RAW files (not JPEGs) and download the 'books' at ISO 100 and ISO 12800 (file names: canon_eos_200d_01.cr2 and canon_eos_200d_08.cr2) and import them to C1.

    Then zoom slowly both photos. On ISO 100 there is no shift in color tones.
    On ISO 12800 when you zoom in past the proxy preview point (depending on your cache settings) there is a significant change in warmness.

    I checked in this way also files of Nikon and Sony format with different ISO values. It happens only on Canon data (also cr3). So I guess it is the way C1 is dealing with noise of higher ISO photos from Canon.

    You can also see, that when I'm working on a higher ISO pictures I cannot judge properly what ICC profiles or any C1 feature actually does to my pic until I zoom in (then I miss context of the whole image and can check only one object in frame) or render to any output recipe, which makes working on that data extremely frustrating.

    Can someone from Phase ONE address the issue and confirm if that's the way it is supposed to be. If yeah, poor Canon shooters... Maybe I'm gonna get new camera 😊

    Greets and thanks in advance for running 'my test' 😊
    0
  • Ian Leslie
    NNN636858814109688210 wrote:

    Can someone from Phase ONE address the issue and confirm if that's the way it is supposed to be. If yeah, poor Canon shooters... Maybe I'm gonna get new camera 😊


    This is a user to user chat forum. If you want to report something to Phase one or get their advice on a topic you have to open a support case: https://www.phaseone.com/en/SupportMain ... pport.aspx
    0
  • SFA
    For someone new to digital photography I think you are making a lot of assumptions that may or may not stand up to deeper investigation.

    Either way I would strongly suggest that using high ISO is best avoided if at all possible even though some more recent camera developments have greatly improved the perceived quality of the images that can be shot at high ISO through a combination of electronics and software. Just because a manufacturer offers it does not mean that it should be considered to be a regular part of your photographic work.

    This is especially true for consumer level bodies. They can and do produce excellent result in the right situations and with conditions that suit them BUT in the main they will always have weaknesses of some sort.

    I have long thought that if one needs to shoot at high ISO it would probably be best to make use of whatever facilities the camera offers for processing since most likely the designers and engineers have done as much as they can reasonably achieve with internal processing. So using the in-camera settings and shooting jpegs might make more sense that shooting RAW with compromised inputs data and then taking a lot of time seeking the best possible improvements with a RAW file when, for most of the time, the resulting output will likely have very limited circulation or ultimate usage.

    That said there are some things that can be done to better understand the way RAW data from a sensor, as provided by the camera manufacturer, is changed into something that resembles a visible image with recognisable content by RAW conversion software. And after that we change it again to make it look like something we wish to look at more than once.

    I can make some suggestions but first I will take a look at some similar images from equivalent cameras from one or two other manufacturers.


    Grant
    0
  • Keith Reeder
    SFA wrote:
    Some people seem to dislike the standard "warmness" of default Canon previews or even the entire look is thought to be too 'warm' by some.

    No "seem" or "thought to be" about it, Grant - clearly there is an ongoing and protracted issue here for many users.

    The regularity with which the subject comes up, time and time and time again, attests very eloquently to that.
    0
  • SFA
    Keith Reeder wrote:
    SFA wrote:
    Some people seem to dislike the standard "warmness" of default Canon previews or even the entire look is thought to be too 'warm' by some.

    No "seem" or "thought to be" about it, Grant - clearly there is an ongoing and protracted issue here for many users.

    The regularity with which the subject comes up, time and time and time again, attests very eloquently to that.


    I know this is a very central concern of yours Keith but I just don't see the same problem that you see or at least not in the same way or, perhaps, to the same extent.

    I have been looking at some sample Canon images as suggested and some sample Nikon images from what seems to be an equivalent camera. Sadly the source doe not have any RAW images form an equivalent Sony camera.

    If one strips things back to the Linear curve and observes the Exposure Evaluation histogram there are some interesting parallels and some interesting differences. But in general the same levels of latitude for exposure accuracy and colour shifting, image by image, can be observed.

    THe base Nikon colours match more smoothly than the Canon colours across the histogram but what that might mean in real terms to the observer is not obvious - at least to me.

    With the suggested target files both systems seem to produce very similar results in terms of colour tones. Noise is different. Colour pins again suggest that Nikon is a little smoother in matching colour levels (reducing the potential for colour cast?) but not entirely consistent.

    In both cases white balance readings from the camera are not the same shot to shot.

    The net result at a 4 image display of previews on my screen is that both the Canon and the Nikon files show very similar shot to shot results for a given ISO setting as far as Colour is concerned.

    Others may find otherwise.

    Assessing sharpness, etc, is not so easy since, inter alia, different lenses are in use.

    It would be interesting to make the comparison with different camera bodies across the price range.


    Grant
    0
  • Joshua Valleau

    OP: Were you ever able to resolve this issue? I'm test-driving C1 this week, and dealing with seemingly the same issue (but using a Sony camera!). So far I've scoured for information and tried everything, but so far no luck.

    0
  • BeO
    Top Commenter

    Also high ISO? Which camera exactly?

    I believe the noise reduction in C1 kicks in only in higher zoom levels, 50% I believe. This might possibly explain it for higher ISO images (but of course it doesn't help much).

    To verify, you could set all NR sliders to 0 and see if you also observe a color shift?

    Btw, is the shift so pronounced that it could be observed here in a screen shot posting?

    0

Post is closed for comments.