rules of thumb to keep in mind when it comes to bit rates. Here's number
1. The higher
the bit rate, the better the sound quality (and size) of the resulting
file. Lower the bit rate, and your sound quality goes down. But your file
sizes go down too.
your particular needs, you can move off the benchmark of 128 kbps. You
can go higher than 128 kbps for less compression and even better sound
quality. Or you can lower the bit rate to further shrink the size of your
mp3 files. Of course, sound quality will suffer the lower you go, but
you can get some amazingly small files in the process.
The table below
can help you estimate the kind of mp3 quality/compression and bit rates
to use for your own applications. We also give you approximate file sizes
to help you compare.
||File size per
||If the wave wav
file was sampled at 16 bits/44.1kHz, you've got CD quality sound
||Very low compression,
superior sound quality
"CD quality" bit rate
|| 96 kbps
||About the same
quality as an FM radio station
Anything below 128 kbps is going to put the hurt on your audio quality,
but you might want to experiment and see how low your file sizes can go.
For instance, you'll be surprised at how well "voice only" audio
can sound at the lower bit rates.
Constant bit rate vs. Variable bit rate
To further complicate the bit rate game, many encoders allow you to customize
the encoding process by selecting either constant or variable bit rates.
Constant Bit Rate encoding (CBR) maintains a steady bit rate stream through
the entire mp3 file. This means a consistent level of compression and
predictable file sizes. CBR has also been around longer, and is supported
by all MP3 players. It's the safe, reliable, consistent route to take
when in doubt. Settings range from 16 kbps to 320 kbps depending on the
file size you're targeting. Remember that files encoded faster than 128
kbps may have little or no detectable improvement in sound quality.
CBR has its drawbacks, especially when encoding certain kinds of music.
Consider this rhetorical question: is your music always consistent in
terms of dynamic range, complexity, polyphony, and stereo separation?
Many times, a single piece of music will vary some or all of these qualities.
Some passages are complex, some are silent with narrow stereo separation;
other passages are dense, complex, and busy with wider stereo separation.
And yet, Constant Bit Rate encoding pays no mind to all that and keeps
its bit rate steady as she goes. You run the risk here of using more bits
than is necessary for the music's quiet parts, and not enough bits than
you'd optimally want for encoding for the busier, more dense musical sections.
What can happen (experiment with CBR encoding to see if you can hear it)
is that the complex/difficult parts don't sound as good as other parts
of the music. On the flip side is the impact of things you can't hear:
the wasted bits encoded during the simpler, quieter sections.
Which brings us to rule of thumb number two:
2. Constant bit rate encoding = consistent, limited file sizes but variable
To address the shortcomings with constant bit rate encoding, Variable
bit rate encoding (VBR) was developed. VBR is newer than CBR; most mp3
players don't support VBR
yet. The most recent upgrades to FreeAmp,
Winamp, and MusicMatch support it. And the Xing encoding program supports
If you seek consistently high audio quality, and file size isn't as much
of a concern, you should consider using it. VBR makes intelligent decisions
before allocating its bits to the encoding process. It adapts the bit
rate to the complexity of the audio, based on a scale you can set in the
preferences of your encoder. The low end of a scale from 1 to 100 would
result in the lowest quality/highest compression. The high end of the
scale would result in the highest quality/lowest compression. The results?
Overall, you'll get better and more consistent sound quality compared
to CBR encoding at the same bit rate.
to Page 3; Back to Page
invite you to discuss issues or post questions on the Pro Sound
World Wide Users Group