Can someone explain to me how the bitrate comes into play since theoretically 44.1Khz is “CD quality”. My guess is it’s something like compression on a JPEG where the size/quality ratio occurs. Yet that does not satiate my curiousity. I still wonder, what is “overdoing it” as far as bitrate is concerned. I want to make “CD quality” Mp3’s yet I don’t the parameters to use. I see Mp3’s as high as 256bps and some as low as 92/112bps. 128 is the standard, but I always figure that’s for the casual listener and I enjoy my detail in music. Would each song have “calculated”
to find the optimum bitrate? I read something about nyquist thereoms and such… I’m awfully confused. If you know what you’re talking about, explain this to me like I’m five, lol. Thanks in Advance. =)
Google better. I typed “MP3 FAQ” into Google, and got this site as the first link.
To answer the question, you’re talking about two different things by mixing “44.1 kHz” and “128kbps” in the same post. The former refers to the sampling rate used to record the original CD, in most cases, 44,100 two-byte samples per second. (Some CDs are recorded using 48,000 samples/second, but the difference should be negligible.)
The latter refers to the compression used to make MP3 files. The answer to “what rate should I use” is “whatever you want.” 128 kbps (that’s kilobytes per second, as opposed to kilobits) gives roughly 10:1 file compression, and a good quality sound. You’re free to use a much higher ratio, but you’ll get poor sound quality. If you want better quality files, use 256kbps, you’ll just be able to fit half as many files onto your disk.
Basically, it will boil down to the quality of your equipment and ears. If you’re using tiny little computer speakers or a car stereo, you might not notice the difference between 128 and 256. You’ll notice the difference between 128 and 56 over a telephone…
It depends on the complexity of the sound that you are encoding. I find 128kbps works for me for encoding pop tunes. I prefer 160-192 or more for classical music. If you are encoding voice-only (like stand-up comedy, interviews, etc), you can drop the encoding bitrate WAY down without losing much quality.
If you tell your encoder to use a variable bitrate while encoding (VBR), it will automatically encode a song at a higher bitrate during the “complicated” parts, and drop the sample rate way down during a quiet or silent section (no need to encode silence at 320kbps).
WinAmp can handles these files just fine but a lot of portable MP3 players can’t handle variable-bitrate MP3s.
Another format you might want to check out is Ogg-Vorbis. It’s already supported by WinAmp. I find I can encode music at 96 or even 64kbps and get a smaller file that sounds just as good as a 128kbs Mp3. Ogg-Vorbis seems to capture the high-frequencies much better and sounds much less “mushy” at lower bitrates.
I recommend dbPowerAmp. You can right-click on a sound file in Windows Explorer and choose between several encoding formats and bitrates. The converter is freeware. You can experiment with encoding the same song at different bitrates and hear what they sound like.
Getting optimal bitrates for mp3’s is a tough problem spanning across the fields of psychoacoustics and information theory. Thankfully for you, a lot of smart people have spent a lot of time researching and implementing the solution.
All you need to do is get a copy of LAME, the best mp3 encoder available. Then encode using one of the following settings:
lame --alt-preset standard
(VBR bitrates 180-220 kbit/s – usually averages around 192 kbps)
lame --alt-preset extreme
(VBR bitrates 220-270 kbit/s – usually averages around 256kbps)
lame --alt-preset insane
(CBR bitrate: 320 kbit)
VBR is part of the official MPEG-1 spec, so any mp3 player that does not support VBR mp3’s is not fully MPEG-1 Layer 3 compliant. You should be able to return it as a defective or falsely advertised product.
The Nyquist Theorem states that you can digitally reconstruct a waveform by sampling at two times the highest frequency in the original analog signal. Since human hearing is (nominally) up to 20 kHz, 44.1 kHz covers that with a little room to spare. CD’s are recorded at a sample rate of 44,100 samples per second. Each sample is 16 bits and there’s a sample for each channel. This is uncompressed audio.
MP3s are compressed audio, so you can’t compare MP3 bitrates to the 44.1 kHz of CDs. Apples and oranges. Or maybe like comparing BMPs to JPGs.
The bitrate in MP3s refers to the number of bits per second that is used to encode the audio. Generally, more bits per second = less compression = better quality. A WAV file ripped from a CD would have a ‘bitrate’ of 44100 X 2 (channels) X 16 (bits) = 1411 kbps. So your high quality MP3s recorded at 256 kbps are compressed by a factor of 5.
No bitrate is going to truly be CD quality, because there will always be artifacts due to compression. But 256 kbps gives very good quality. (Attempt to pre-empt flaming: “good quality” is a very subjective term.)
As for a recommendation, I’d say 256 kbps is worth it, if you’ve got decent speakers and can hear the difference. Hard drives are getting cheaper all the time.
As far as calculating the best bitrate for each song. Not exactly, but there is something called Variable Bitrate Recording (VBR) which adapts the bitrate to the complexity of the music moment by moment, resulting in better quality and smaller filesizes.
Audio CDs (CD-DA) are not compressed. Neither are .wav files. That’s why MP3 and other compression algorithms are used, so one can store a larger number of tracks in the space available on your music player or hard drive.
Sound on an audio CD, whether it is a pressed CD or a CD-R, is encoded in PCM. The tracks on an audio CD aren’t files, and it doesn’t have an organized volume structure like a CD-ROM.
CD audio is uncompressed 44100samples/sec x 2 bytes/sample x 2 channels = 173 KB/sec. which is considered 1X for CD readers.
>> The tracks on an audio CD aren’t files, and it doesn’t have an organized volume structure like a CD-ROM.
They have a structure, just a different structure. In fact, CD Audio is the earliest standard and CD-ROM was developed later using CD audio as basis so it is not developed from scratch and there are slightly different standards developed.
I know they have a structure (how could a CD player find the tracks if they didn’t?), just not one comparable to the hierarchical system used for data. A CD-DA has no notion of files or folders or volumes. The organization of music into tracks is a very simple, low-level system, mostly handled by hardware.
