dng photos

Answered

Comments

11 comments

  • Avatar
    ton14

    Don't use any RAW or DNG! Reality capture uses embedded JPG in DNG file.
    Use JPG or TIFF-8.
    Use JPG direct from camera. Or use a software from the camera manufacturer for RAW converting.
    Or use new 16-import for TIFF-16 and other 16-bit formats.

    0
    Comment actions Permalink
  • Avatar
    Erik Kubiňan CR

    Dear suntchan,

    As ton14 mentioned it is not recommended to use DNG format as it only uses it's thumbnail you can put RAW photos into RC and it will process them successfully but to save up space and have access to tweaking the raw file. I'd also suggest you to first convert and postprocess the images and then use JPGs.

    0
    Comment actions Permalink
  • Avatar
    suntchan

    Thank you.

    What kind of postprocesses do you recommend ?

    0
    Comment actions Permalink
  • Avatar
    Erik Kubiňan CR

    Well decreasing highlight values, increasing shadow values, lowering noise, tweaking white balance. Just don't do the lens distortion outside RC it is better to let RC do it.

    0
    Comment actions Permalink
  • Avatar
    ton14

    I tested RAW(Sony .ARW and Cannon .CR2), TIFF, JPG
    RAW have worse alignment.

    0
    Comment actions Permalink
  • Avatar
    suntchan

    I tested NEF (Nikon) on one model and the alignment was as good as with JPG (no highlights, boost shadows, noise reduction).

    So I wonder if it is worth the time to postprocess the images ?

    RC seems to do very well with NIKON RAW

    0
    Comment actions Permalink
  • Avatar
    Erik Kubiňan CR

    It will do well with RAW formats and it is completely up to you, but if you convert your RAWs into JPG and delete the RAW images afterwards you will save a lot of disk space.

    0
    Comment actions Permalink
  • Avatar
    suntchan

    Actually NEF files are almost the same size as JPG, so no saving.

    0
    Comment actions Permalink
  • Avatar
    Erik Kubiňan CR

    Alright as I said it's up to you, this is just my suggestion from an experience working with NEFs.

    0
    Comment actions Permalink
  • Avatar
    andrew breckill

    I use a two fold approach, take images as raw and do a two step process, I increase contrast and sharpen slightly (without introducing more noise), if shadows are too black i lighten them and reduce any blown out areas, (modern dslr;s have fantastic ability to capture detail which can be recovered if you shoot raw). I then save those out as 95% quality jpegs for doing the solve. I then flatten the raw images to as neutral as possible using adobe raw in photoshop (aiming for as linear as possible) these 'flat' images are then use to texture the geometry got from the jpegs, the reason for that is to have a texture output file that the render package used can then apply its own tone curves to to keep as much data as possible.

    0
    Comment actions Permalink
  • Avatar
    Erik Kubiňan CR

    Absolutely understand and I use the same workflow almost. What I am saying is that after you flatten the RAW file you can just export it as a jpeg and that's it, you can delete the RAWs. You wont get much more using the RAWs instead.

    0
    Comment actions Permalink

Please sign in to leave a comment.