VideoHelp Forum




+ Reply to Thread
Results 1 to 3 of 3
  1. Member
    Join Date
    May 2001
    Location
    God's Country
    Search Comp PM
    Basically in the process of encoding, programs like TMPGEnc or Main Concept are resizing the video and throwing away or cropping frames and pieces of the original video to get it trimmed down to whatever spec. you want.

    My question is basically this, Do you lose more quailty during the process when encoding an original which has much more resultion than the end product will have. Or if your original source is much closer to the final source, is that better?

    For instance, is quailty lost when encoding something that is 1200x960 to 480x480 verus something that is 640x480 to 480x480, or is the difference nil? Something tells me if the encoder works harder throwing away more info, wouldnt that lead to more quality loss than if the original source was closer in the first place?

    tygrus
    Quote Quote  
  2. Member flaninacupboard's Avatar
    Join Date
    Aug 2001
    Location
    Northants, England
    Search Comp PM
    the quality should be better using the higher resolution source. as there are more samples (pixels) to choose from when creating the resized frame, you get a more accurate representation of the original, and a smoother image. this means going from film resolution to DVD resolution creates a nice even picture which doesn't change much from frame to frame, meaning low noise, which means less work for the encoder and less visible artifacts.
    Quote Quote  
  3. Member
    Join Date
    Mar 2003
    Location
    Uranus
    Search Comp PM
    MPEG2 encoding does not work by cropping frames, throwing away frames, resizing, or anything remotely like that.

    It performs a 2 dimensional frequency transform on the pictures and...
    Sometimes discards higher frequency components (detail)
    Sometimes quantizes components to numbers with less precision
    Arranges coefficients into same value groups so that multiple occurences
    of the same number aren't stored.
    Sometimes stores only the differences between one frame and the
    next to save space
    Sometimes notices when a part of the picture has moved but not
    changed and stores only the motion vector.

    In general, the more data you start with the better. If a picture can be
    represented at 720 x 480 without loss, starting with a larger picture
    cannot improve it but will not degrade it either except for the slight
    loss in resampling it to 720 x 480.

    A noisy source will encode better if oversampled.
    Quote Quote  



Similar Threads

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