メインコンテンツへスキップ

⚠️ Please note that this topic or post has been archived.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

Auto-mask question

Michael Sonshine
I have been spending quite a bit of time testing (and learning) the Color Editor masking functionality in version 9 and, as part of that, removing the selected area from parts of the image that I don't want included. In testing the functionality I have used raw (Canon CR2), tiff and dng formats and found, much to my surprise, that C1 will not allow me to use the auto-mask functionality with dng or tiff files. Can anyone tell my why?

I have looked at the online help file and it explains use of the auto-masking functionality, but I already knew that. It says nothing about restrictions in using the auto-mask with specific image types. The error message I get is:

The Auto Mask brush cannot be used with this image format.

Why is this?
0

コメント

6件のコメント

日付 投票
  • HansB
    Would be nice if someone from PO's development group could give some more information here. 😊

    For as far as I know,
    - automask relies on raw subpixel information. That information is for sure not available in jpeg and tif images.
    - dng translates the camera's proprietary raw format into some kind of Adobe standard file format. I don't know if that file still contains and allows access to the original subpixel information, but I don't think so.


    Regards,
    Hans
    0
  • Michael Sonshine
    Well, that would certainly explain it. Thank you for the help.

    What continues to surprise me is that each new C1 release that I play with ends up teaching me more about the version I already have since I spend so much time trying to learn what is in the new release and coming up with all of these questions which apply to the previous release as well. The new Color Editor masking functionality caused me to spend hours working with it and made me realize that I did not know the reasons for some basic CE limitations (like auto-masking).
    0
  • SFA
    Hans,

    I think you are right.

    There was a reply in one of the "Why are the Fuji sensors not fully implemented ...." threads that referred to the way that edges are identified based on the analysis of a standard Bayer array (or something close to that wording) which suggested that any file that does not have such information to start with would not be able to use the current method of edge detection.

    So that means either one comes up with a different approach (possible issues about the perceived quality of the results?) or several approaches, perhaps accepting compromises for each (and then trying to explain them to users who are not very interested in listening).

    It's some amount of work. I would guess that a cost benefit analysis might be difficult to produce let alone position in a list of benefit analysed development projects.

    Considering the proposal from a purely selfish personal current point if view ... the time and cost expended would add nothing to C1 that I would benefit from (so far as I can tell) and I would rather the effort was expended elsewhere.

    Obviously I fully understand that people with Fuji cameras and TIFF/DNG workflows are likely to have exactly the opposite opinions! If that was me I would see it as my "problem" to resolve - most likely by using multiple applications or maybe changing hardware next time that need/opportunity/desire came around in the equipment cycle.


    Grant
    0
  • HansB
    [quote="MikeFromMesa" wrote:

    Well, that would certainly explain it. Thank you for the help.
    ...

    No warranty for that explanation, because I'm not a Phase One developer and I don't have the Capture One source code. But I'd love to peek into it. 😁

    [quote="SFA" wrote:
    Hans,

    I think you are right.

    There was a reply in one of the "Why are the Fuji sensors not fully implemented ...." threads that referred to the way that edges are identified based on the analysis of a standard Bayer array (or something close to that wording) which suggested that any file that does not have such information to start with would not be able to use the current method of edge detection.
    ...

    I missed that one, Grant. I should read the Fuji threads more carefully. 😊


    Regards,
    Hans
    0
  • SFA
    [quote="HansB" wrote:
    [quote="MikeFromMesa" wrote:


    I missed that one, Grant. I should read the Fuji threads more carefully. 😊


    Regards,
    Hans


    It was just a few days ago Hans as I recall. Certainly within the week.


    Grant
    0
  • SFA
    Hmm.

    Just on a week ago.

    viewtopic.php?f=61&t=21332&p=101171&hilit=fuji+auto+mask&sid=c3cc1788f94943b7c6e6c61759d60a40#p101171


    I saw it first just 2 or 3 days ago.

    I think this is a something I read some months back as well.

    One of the challenges with technical developments is deciding how to approach the requirement. Do you go all out for what you believe to be 100% even though is might be difficult and limit future options (maybe)?

    Of do you go for something less technically challenging on the basis it will be "good enough"?

    Or do you take a completely different view that avoids being tied in to any deep underlying requirements leaving you flexible opportunities well into the foreseeable future - but at maybe only 80% of the possible quality/performance/Usability levels?

    Obviously I am just speculating in very general terms here based on my experiences in software development over a number of years. There is no simple single answer - or none that I have found so far. In recent years even choosing a development platform has not been an easy task nor with a guaranteed potential outcome. Multiple total application rewrites have become quite normal in the "agile" development world.


    Grant
    0

投稿コメントは受け付けていません。