VideoHelp Forum




+ Reply to Thread
Page 2 of 2
FirstFirst 1 2
Results 31 to 34 of 34
  1. Originally Posted by manono View Post
    Originally Posted by swetekman View Post
    The statement that files size do not relates to frame frequency (fps) is still wrong.
    Nope, dead wrong. It's entirely a function of the bitrate and the length of the video. Framerate has nothing at all to do with it.

    A video at, for example, 30fps, at a bitrate of 1000 and a minute long will be the same size as another video at 24fps with the same bitrate and length. Roughly 7325KB.
    Sorry, I can not not respond.

    Yes you are of course right in your statement and all I am trying to tell you is that FPS is a factor in bitrate. This is easist shown by taking your example all the way:

    A video at, for example, 30fps, at a bitrate of 1000 and a minute long will be the same size as another video at 1fps with the same bitrate and length. Roughly 7325KB.

    So what do we have: Two videos that are the same size. One video that runs smootly and one video that stutters at 1 frame per second but they still take up the same space. Something is wrong in fixed bitrate land.

    The error you are making is that you do not understand what the bitrate of a video file is built up from and assume that you always should set a fixed target bitrate on everything you encode.

    So reencode the 1 fps file and change from a target bitrate to a quality setting where the encoder chooses the appropiate bitrate. This will probably drop the bitrate and file size of the 1fps file to something closer to 1/30 of the 30fps file . And it will still look very close to the 1 fps video that were 30 times bigger.

    From wikipedia
    In digital multimedia, bitrate represents the amount of information, or detail, that is stored per unit of time of a recording. The bitrate depends on several factors:
    The original material may be sampled at different frequencies
    The samples may use different numbers of bits
    The data may be encoded by different schemes
    The information may be digitally compressed by different algorithms or to different degrees
    In video, FPS is related to sample frequency.
    Different number of bits per sample is relates to frame size.

    I think it is the compression part that confuses you. But that is just my opinion. You are entitled to your own. Now I will not explain anything else.-)

    All moments like these will be lost in time, like tears in rain ... Time to die.
    Quote Quote  
  2. In real world we can theorize about targeting quality or targeting size (bitrate). If we start to mix those two goals it gets perplexing and one can lose what the priority is.
    Quote Quote  
  3. Originally Posted by swetekman View Post
    Can you agree on this assumption:

    Given a uncompressed video file, dropping half of the frames and retaining the same running time produces an uncompressed video file that has half the size, half the bitrate and half the FPS.
    This is not an assumption, this is a fact.

    In your uncompressed video example, this is pure CBR, and every frame is the exact same coded frame size. Thus a reduction of frames by 1/2 will exactly equal a 1/2 reduction in filesize. ALWAYS.



    If we agree on that, we should agree on the assumption that filesize has a relationship to FPS otherwise the size of the file in the assumption above would not have change.

    That is my point. Nothing else.
    But did you just ignore the rest of my post ? Do you see the "nothing else" part of your post is the problem to your understanding. You're making other assumptions about rate control and quality

    If you don't explicitly state those assumptions then you are wrong: then FPS has nothing to do with bitrate or filesize. BUT, If you state those assumptions then you are almost saying the same thing that was stated earlier

    Please Re-read post #11 and #29



    He dropped the frame rate and ended up with a file that has half the size of before and still has a quality he is happy with.
    That is what this thread should end with, a simple way to reduce file size and bitrate.
    And I'm 100% certain he's made a mistake in this reported observation. Show me 1 example where encoding at the same CRF value, using typical encoding settings, that a 60p file will result in 2x the filesize compared to a 30p version of that file . I will bet money on this. Unlike your uncompressed video example, long gop encoding only stores the differences between frames as the residual in P and B frames. Because there is similarity in adjacent frames, you will never get 1:1 reduction of frames to bitrate



    I will use his handbrake settings as inspiration to drop the bitrate on a couple of 720p60fps videos which has to high bitrate for my computers. By reencode them in 720p30fps I will get half the bitrate and probably be able to watch them on any of my computers.
    But this won't happen. The bitrate will be lower, yes (again because you are using CRF rate control ), you will probably achieve around 10-20% reduction in bitrate only, not 50% reduction in bitrate

    I just encoded some 720p60 7D videos and the average was about 1.1-1.2x the filesize for the 60p version vs. 30p version . (of course it will depend on content complexity and compressiblity, but the point is, you won't get near the savings you think you will get)

    Is it worth it? To some people yes, to others no. But Temporal resolution is part of quality - the choppiness just isn't worth it to many people . For something like sports videos (football/soccer etc..) it might not be worth it
    Quote Quote  
  4. Originally Posted by swetekman View Post
    I will now stop explaining this any more and let the rest of you who still disagree stay in your bubbles of bitrates that are independent of FPS.
    Good, so you don't make yourself look even more the fool.

    Did it even occur to you that the 2 pictures which you seem to think are proof positive of your misguided opinions are showing the bottom video at twice the length as the first? That is, half the frames aren't removed, but it's been slowed to half the speed (30fps from the original 60fps) and of course the bitrate has been halved for the same final filesize.

    Edit: Damn, you removed the bitrate calculator pictures. Did you suddenly realize they proved our points, and not yours?
    Quote Quote  



Similar Threads

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