VideoHelp Forum




+ Reply to Thread
Page 5 of 13
FirstFirst ... 3 4 5 6 7 ... LastLast
Results 121 to 150 of 390
  1. @Selur
    Can i replace the x264 builds with the builds by taro_06 in "hybrid"?
    Quote Quote  
  2. sure, make sure to name them accordingly (naming: x264.exe for the 8bit version, x264-10bit.exe for the 10bit version), but other than that I know a bunch of users which use taros builds with Hybrid.
    Quote Quote  
  3. "Variance-based AQ (VAQ) attempts to improve quality in flatter areas of the image at the cost of more complex areas of the image."

    so if i increase the aq-strength will the complex areas look bad since x264 will take bits from complex areas and apply it on the flat areas?
    Quote Quote  
  4. They don't have to look bad. Yes, there will be a bit rate distribution, but depending on the distribution beforehand this might not change the visual appearance of complex areas at all.

    Also here's a bit more about aq-mode/-strength:
    Adaptive quantization mode:
    Adaptive quanization is an algorithm that tries to optimally choose a quantizer for each macroblock. Depending on the strength setting it tries to improve quality in flatter areas of the image at the cost of more complex areas of the image, or vice versa. The can help to minimize the overflattening and blurring out fine detail in an image.
    Changing the mode changes what scope adaptive quantization re-arranges bits in:
    0: Do not use AQ at all.
    1: Allow AQ to redistribute bits across the whole video and within frames.
    2: Auto-variance AQ (experimental) which attempts to adapt strength per-frame.[/list]
    (command line match: --aq-mode <integer>)

    Adaptive quantization strength:
    Specifies the adaptive quantization strength. Adaptive quantization strength is ment to reduces blocking and blurring in flat and textured areas. As a guide line a value of 0.5 does a weak adaptation and a value of 1.5 a strong. In general I would recommend to stay with the default value of 1.0.
    value < 1: -> less bitrate around edges, more bitrate for flat areas
    value > 1: -> more bit rate around edges, less for flat areas
    (command line match: --aq-strength <float>)
    Quote Quote  
  5. Originally Posted by x264 View Post
    so if i increase the aq-strength will the complex areas look bad since x264 will take bits from complex areas and apply it on the flat areas?
    With CRF encoding the quality of complex areas doesn't go down, rather the bitrate goes up.
    Quote Quote  
  6. So x264 will apply more bits to the flat areas?
    Quote Quote  
  7. Originally Posted by x264 View Post
    So x264 will apply more bits to the flat areas?
    Yes.
    Quote Quote  
  8. Which dehaloing filter do you guys use?
    Quote Quote  
  9. depends on the halos, most times: DeHalo_Alpha or BlindDeHalo3
    Quote Quote  
  10. How does a dehalo filter remove the haloing?
    Quote Quote  
  11. Originally Posted by x264 View Post
    How does a dehalo filter remove the haloing?

    They are supposed to, but how effective they remove the "halos" depends on the source

    Usually dehaloing filters are very damaging at the strengths and settings required to remove halos effectively
    Quote Quote  
  12. basically, dehalo filtering is normally done in three steps:
    1. look for hard edges
    2. create smoothed version of image
    3. apply smoothed version of image around the detected edges from original
    -> get smoothness around sharp edges
    So dehaloing is basically a 'smooth around edges' function
    Quote Quote  
  13. Originally Posted by x264 View Post
    How does a dehalo filter remove the haloing?
    By doing the reverse of what causes halos. Halos are caused by sharpening. Dehalo filters work by blurring. As usual, some techniques are better than others. And results will vary depending on the source.
    Quote Quote  
  14. You guys are so helpful, thanks for all the advices you've given me . I got warned two times in a row for posting similar questions in the newbies section on Doom9.
    Quote Quote  
  15. Does frame interpolation using InterFrame require very high bitrates to look good?
    Quote Quote  
  16. How do i link the OP and ED with the main episode using ordered chapters?
    Quote Quote  
  17. Originally Posted by x264 View Post
    Does frame interpolation using InterFrame require very high bitrates to look good?
    It will require more bitrate than not using InterFrame. Why don't you just perform a pair of CRF encodes and find out just how much.
    Quote Quote  
  18. I tried encoding a small part of a BD and it looked quite good, sometimes the space around people look distorted.
    Quote Quote  
  19. Originally Posted by x264 View Post
    sometimes the space around people look distorted.
    Yes, all the motion interpolation algorithms generate artifacts around moving objects and at the edges of the frame, or just all over with certain material. See this post and the one after:

    https://forum.videohelp.com/threads/352741-Frame-interpolation?p=2215501&viewfull=1#post2215501

    Or this post:
    https://forum.videohelp.com/threads/339017-Motion-Interpolation-%28VidFIRE%29-Software?...=1#post2107107
    Quote Quote  
  20. How do i calculate the S.A.R after cropping?
    Quote Quote  
  21. most people translate SAR as storage aspect ratio, which is the width divided by the height
    Quote Quote  
  22. Isn't that the same thing as the A.R?.
    Quote Quote  
  23. There are variations in terminology. Some documents use PAR (pixel aspect ratio), some use SAR (sample aspect ratio). The frame size aspect ratio is sometimes called FAR (frame aspect ratio) and sometimes called SAR (storage aspect ratio). At least DAR (display aspect ratio) is consistent.

    It's best to spell out exactly what "aspect ratio" you mean.

    When you crop a frame you do not change the pixel aspect ratio.
    Quote Quote  
  24. x264 --sar <----- How do i calculate this?
    Quote Quote  
  25. Originally Posted by x264 View Post
    x264 --sar <----- How do i calculate this?

    --sar x:y can be though as the w:h of pixels. Use whatever solves the equation

    Display Aspect Ratio = Frame Aspect Ratio x Sample Aspect Ratio
    (again, these can have different names, but the meaning is the same)

    DAR = (w:h of frame dimensions) x (w:h of pixels)

    16/9 = 720/480 x ?

    for example , the answer here would be 32:27 , or --sar 32:27
    Quote Quote  
  26. According go the DVD MPEG 2 spec the DAR of the DVD comprises the full 720x576 pixels. So:

    Code:
    720x576  PAL 16:9 SAR = 64:45
    720x576  PAL  4:3 SAR = 16:15
    In this case SAR is the sample (or pixel) aspect ratio. The value that is specified by --sar in x264.
    Quote Quote  
  27. Will cropping the borders change the SAR?
    Quote Quote  
  28. Originally Posted by x264 View Post
    Will cropping the borders change the SAR?
    Take a piece of graph paper. Crop of the edges. Has the shape of the remaining grid changed?
    Quote Quote  



Similar Threads

Visit our sponsor! Try DVDFab and backup Blu-rays!