First, we should define what a bitrate is. A bitrate refers to the number of bits (data) processed over a given amount of time. You’ll come across bitrates in many places. Your Internet connection, for instance, is measured in bitrate. You might have a Verizon Fios package with a “100 Mbps Internet”. That 100 Mbps is the bitrate, 100 Megabits per second (1,000,000 bits per second).
In digital multimedia, bitrate represents the amount of data stored per unit of time of a recording. Typically, the higher the bitrate — sometimes inaccurately abbreviated as “bps”, not to be confused with “beats per second” — the higher the audio quality. Another way of thinking about it is there’s more detail in the data at each point of a recording, making the audio sound more rich.
The difference between Constant Bitrate (CBR) and Variable Bitrate (VBR) is pretty straight forward. However, there are considerations to be made when deciding to use one over the other.
What is Constant Bitrate (CBR)?
CBR encoding keeps the rate of bits constant throughout an audio file’s duration. Typical values used in MP3s, for instance, include 128 (CD quality), 256 and 320 kbps (highest possible per the MP3 standard). The advantage to using CBR encoding is that the data can be processed faster. A disadvantage is that CBR encoded audio files are poorly optimized when it comes to quality versus file size.
What is Variable Bitrate (VBR)?
VBR encoding, on the other hand, allocates more bits to complex segments of an audio file and less bits to simpler segments, meaning a VBR encoded audio file has a bitrate that varies throughout the audio file’s duration. A VBR encoded audio file is typically measured by its average bit rate. These files are highly optimized, having smaller file sizes with no noticeable quality loss compared to equivalent CBR encoded audio files. But there are some disadvantages, too. It typically takes longer to encode and decode variable bitrate audio files because the process is more complex. You may also find that some devices and software, typically older, do not support VBR encoded audio files as well as CBR encoded audio files.
Between the two types of encoding, Constant Bitrate (CBR) and Variable Bitrate (VBR), VBR is preferred in most cases because it encodes data more accurately (better quality) using fewer bits (smaller file size).
I once opened an audio file with iTunes, years ago, and noticed the duration wasn’t displaying correctly. The audio file was a VBR encoded file. And iTunes was, well, iTunes. But I soon realized it wasn’t just iTunes that had this issue. A bunch of media players had the same issue. Or worst yet, when I tried to seek through an audio file, the software would just crash or bug out. It still happens today. Sometimes I feel my life’s calling is to get to the bottom of this.
From what I can tell, many media players estimate durations based on bitrate. But if your bitrate is variable, this becomes difficult to do accurately. A poor algorithm may display an estimated duration of 4:00 minutes, when the actual duration of the audio file is 4:45 — a considerable difference, especially if the audio stops playing at the 4:00 minute mark. Likewise, seeking to the mid-point of a file by means of a UI control, like a slider or seekbar, may mislead — the user may think they’re listening to the middle of a song, but may really be listening to only the first minute of a song.
Madness! People can praise VBR all they want, but in my experience it’s caused nothing but confusion and trouble. VBR encoding may not be as hot as some people make them out to be. A word of warning to the wise.