MP3: CBR Vs. VBR

Discussion in 'Archived Threads 2001-2004' started by NickSo, Jan 19, 2002.

  1. NickSo

    NickSo Producer

    Joined:
    Jul 2, 2000
    Messages:
    4,260
    Likes Received:
    0
    Real Name:
    Nick So
    Okay, i know that CBR is Constant Bitrate, and VBR is Variable Bitrate, but the question i have is, why the difference? From my understanding CBR means that the bitrate is always the same (like always 128kbps, etc), and VBR's change according to the information... Which is better in terms of sound quality? Are VBR's usually smaller in size than CBRs?
     
  2. Clinton McClure

    Clinton McClure Casual Enthusiast

    Joined:
    Jun 28, 1999
    Messages:
    4,279
    Likes Received:
    353
    Location:
    Central Arkansas
    Real Name:
    Clint
    Supposedly, you can achieve high-quality MP3s using VBR which take less space than those encoded using CBR. (A VBR file will use less space than a 320k CBR file.)

    However, I have found that even using the LAME encoders, VBR files are sometimes prone to clicking & popping and some first generation car MP3 players have had trouble playing VBR files.

    Personally, I just use 320k CBR to encode, but others prefer VBR.
     
  3. Shayne Lebrun

    Shayne Lebrun Screenwriter

    Joined:
    Jun 17, 1999
    Messages:
    1,086
    Likes Received:
    0
    If space is a consideration, use VBR. If space isn't a consideration, use CBR. If space REALLY isn't a consideration, use something losless, like, say, raw PCM. :)

    CBR is the equivalent of using a 2 gallon bucket to more water; if you're moving a cup of water, oh well, if you're moving four gallons, you're going to lose some. VBR is the equivalent of having a set of buckets, the smallest and largest defined when you encode into MP3, and using whichever one is right.

    I'll also point out that DVDs use VBR MPEG-2 for the video.
     
  4. Kimmo Jaskari

    Kimmo Jaskari Screenwriter

    Joined:
    Feb 27, 2000
    Messages:
    1,528
    Likes Received:
    0
    That bucket analogy isn't the best I've ever read. [​IMG]
    I don't think it's that hard to get. Silent or very simple sounds can be reproduced very well with small amounts of data, and more complex sounds get up to 320 kbit to make sure the sound gets reproduced properly. Seems like the sensible approach to me.
     

Share This Page