Forum Replies Created

Page 65 of 317
  • Chris Wright

    August 20, 2017 at 9:02 pm in reply to: Issues with Panasonic GH5 footage

    just enable high quality playback. its a premiere bug as of now.

  • Chris Wright

    August 20, 2017 at 5:04 am in reply to: deinterlace originally film footage.

    Jody,

    that’s helpful instructions. virtualdub seems like it has a good chance of doing everything in one pass. the newer virtualdub has an adaptive pattern phase ivtc filter (set field assembly mode, reduce framerate. pretty amazing its all free.

  • Chris Wright

    August 20, 2017 at 3:35 am in reply to: deinterlace originally film footage.

    no, your original assumption is incorrect. your video should look like this in mediainfo if its really 23.976.

    Display aspect ratio : 16:9
    Frame rate : 23.976 fps <——-
    Standard : NTSC
    Color space : YUV
    Chroma subsampling : 4:2:0
    Bit depth : 8 bits
    Scan type : Progressive
    Scan order : 2:3 Pulldown <——-

    see the difference?

    that’s because of mpeg2 flags; you can have all sorts of mpeg2 DVD stuff. some is 23.976 with 3:2 pulldown flags, others are 29.97i regular alternating interlaced frames, others if you had a 29.97p, it would be converted into 29.97i with progressive interlaced frames each frame 2 fields complete image.

    you probably have a 29.97i with fields permanently burned in, no flags whatsoever!!!

    that’s why you need to analyze the fields and write down frame by frame to see fields and complete frames ie. WWSSW in after effects.
    what troubleshooting steps have you done in premiere, after effects? tried placing it in 23.976 sequence in premiere, interpret lower/upper? a 59.94 comp in AE? deleted duplicates frames with AE script or ffmpeg?

    if, for some reason, there is no pulldown burned in, you have a doubly messed up video stream and might just need to deinterlace with qtgmc.

  • Chris Wright

    August 19, 2017 at 10:41 pm in reply to: Mini DV Tapes to Premiere CC
  • Premiere actually does grade you-get-what-you-see in sRGB if your monitor is also calibrated sRGB. premiere will ignore any other color profiles. quicktime is not sRGB(0-255 gamma 2.2). quicktime(depending on the version), RGB values are 16-235 with various gamma styles like 1.8 or 2.4.

    Furthermore, if you colorgrade in Premiere and your monitor is not sRGB(instead P3 or adobe RGB), you will get incorrect grades because you are grading in the wrong native color profile. VLC can match premiere perfectly if you set its video output to OpenGL. Also, nvidia control panel needs to be set to 0-255 so that your gfx card doesn’t change your RGB values in media playback.
    Adobe Media encoder can render video range(16-235) or full range(0-255) as well. Resolve has FCP legacy feature.

    Also,
    external displays are still limited by adobe’s mercury transmit protocol that is hard coded to rec. 709 so you’ll need to match that too.

    If you’re still having problems with youtube, it could be youtube changing the gamma/pixel values, in which case
    you can try manually baking in a new look.
    I can’t promise this will work for you, but many people have had luck using this LUT I made for youtube/vimeo.

    64 cube iridas lut for burning in darker 16-235 from 0-255 for youtube upload. it darkens image, then youtube/vimeo re-lightens again.
    https://f1.creativecow.net/10598/fixmyyoutube

    note:
    it doesn’t work with adjustment layers directly
    you have to use it in the dropdown for the export in adobe media encoder. or you can NEST it first.
    its a premiere bug. also it needs to be copied in both premiere-lumetri-technical and adobe media encoder-lumetri-technical

  • Chris Wright

    August 18, 2017 at 12:54 am in reply to: deinterlace originally film footage.

    if you place the 29.97i into after effects and into a 59.94 comp, each field will become its own frame, then simply
    download the duplicate frame removal script and it should remove the duplicate frames, thus deleting the embedded duplicate fields. render that out as new 23.976 comp no fields. should not matter if the cadence changes.

    https://aescripts.com/duplicate-frame-remover/

  • Chris Wright

    August 17, 2017 at 6:52 am in reply to: deinterlace originally film footage.

    looks encrypted to me.

  • Chris Wright

    August 17, 2017 at 5:32 am in reply to: deinterlace originally film footage.

    did you try renaming it to .mpg?

    that’s really odd. Jim Simon’s helpful program XviD4PSP works with a lot of people. did you select uncompressed avi and wave?

    i used to use a program called vob to avi. its still free i see. if you choose some compressed codec, you’d had to get the same plugin support for your editor. that’s why uncompressed avi is recommended, plus there’s no quality loss.

    short of sending a vob example to us, I don’t know what to tell you except try another vob to avi converter with uncompressed or codec support. you can check the codec with mediainfo in tree view.

  • Chris Wright

    August 17, 2017 at 2:48 am in reply to: deinterlace originally film footage.

    aw, direct opening vob’s. gotcha.

    https://forums.adobe.com/message/6034981#6034981

  • Chris Wright

    August 17, 2017 at 12:40 am in reply to: deinterlace originally film footage.

    1. does it work if you interpret video fields lower or upper and place into 23.976 sequence? or do you still see fields?
    try both upper and lower first.

    2. sounds like an old school problem. first, put it in after effects, interpret fields up/low into a new composition 23.976. it should spread the frames correctly. if not, then click detect guess pulldown type.

    if that doesn’t work, then place video into a composition double its framerate 59.94. then hit PgDn and write down 10 frames for us in this format. Each field will then get its own frame.
    W=whole frame
    S=new frames

    If you get a pattern after 10 frames, it should look like WWSSW then another WWSSW or something. then go back, and change ae’s reverse pulldown cadence to this.

    3. if you have no repeating pattern whatsoever, then the pulldown cadence is variably changing, in which case use alchemist OD, virtualdub, fcp compressor to remove adaptive ivtc. which can include 2:3:3:2, 2:3:2:3 and the strange 2:2:2:4

Page 65 of 317

We use anonymous cookies to give you the best experience we can.
Our Privacy policy | GDPR Policy