Jump to content

berndimax

Members
  • Posts

    6
  • Joined

  • Last visited

Posts posted by berndimax

  1. 4 hours ago, no_connection said:

    I did check both the video samples for download above with ffdshow internal reporting to .csv file.

    And it did report an I frame every 15 frames. 448 P frames and 32 I frames. For the clip with water.

    The magic happens when I open the clip in fusion. I expected to see mush with huge blocks at best. But this?
    Not sure if fusion is responsible for grain or not but I can't get anywhere near as good of a result by converting it to another format.

    fusion_xt2.PNG

    When I went through the file frame by frame in avidemux, I counted 29 P-frames in a row, but mediainfo reported a GOP of 15 as you said.

    I don´t know if avidemux shows it right.

     

    6 hours ago, John Matthews said:

    Where can I go to teach myself about this?

    Finished downloading the file and looking at in FCPX, it looks highly detailed, but whenever there's movement of any kind, blocks start appearing around the detail- looks like crud. Let's hope this isn't representative of Fuji "fixing" issues with aliasing. Let's hope this isn't what the final product will be!

    Go to Wikipedia: https://en.wikipedia.org/wiki/Context-adaptive_binary_arithmetic_coding

     

  2. Hello,

    I`m new to this forum, but a long term reader, because it is the best forum for information about NX1 (which I own).

    The Fuji got my interest for its pleasing colours.

    I downloaded two original files from http://www.dkamera.de/news/preview-beispielaufnahmen-der-fujifilm-x-t2/

    and discovered, that the codec implemantation may not be very good.

    It uses a "High L5.1" variation of mp4 at 102 Mbps with no CABAC and no B-frames unlike for example Sony RX100IV etc.

    One I-frame followed by 30 P-frames according to mediainfo and avidemux.

    Just like smartphones, but at a higher bitrate.

×
×
  • Create New...