MP3 bitrate question

[Johnbear007]

is there anyway to take a lower quality bitrate and convert itup to higher? like 128 to 320?

 

[Damascus]

You won't be gaining anything.

 

[Johnbear007]

why is that?

and is it possible?

 

[PrincessGuard]

MP3 is a lossy compression algorithm, which means that data is thrown away in order to make the file small. The more data you throw away, the lower the bitrate, the smaller the file.

You can't magically recreate the data that was thrown away (unless you have to original WAV file) so you gain nothing from re-encoding a 128kbps file to a 320kbps. In fact, the re-encoding process will lower the quality even more.

 

[GigaCluster]

If it's still not clear, I'll offer another explanation.
Let's say the original WAV contains 100% of the information, a 256 Kbps MP3 contains 80% of the original, while 128 Kbps MP3 contains 60% of the original.

If you create a 128 Kbps MP3 from the original, you have 60%. Now you say you want to raise the bitrate to 256 Kbps. If you don't have the original, your existing (128 Kbps) MP3 will have to function as the original. So, you create a 256 Kbps from the 128 Kbps; you now have 80% from the original... BUT the original is already 60% of the WAV!

Now the quality of the result = 0.80 * 0.60 = 0.48... so if you do what you're wanting to do without the original WAV, you will end up with a MP3 song that is a mere ~48% of the original. Of course, that number holds true only if my original percentages are accurate, which they're not. Suffice it to say, though, that your result will be of lower quality than what you started with.

 

[Workin']

Let's say the original WAV contains 100% of the information, a 256 Kbps MP3 contains 80% of the original, while 128 Kbps MP3 contains 60% of the original.

Here are the "real" numbers in case anyone cares:

original wave file = 100% of the data
128 kbps mp3 = 1/12 or 8.3% of the data from the original file, but it's not really "apples to apples" comparison because the mp3 file has more efficient encoding, we're really just comparing file size not actual "data" if that makes any sense.

But what GigaCluster posted is pretty much the way it works. With a lossy compression scheme there's no going back to the original file, and transcoding to different bitrates introduces greater losses than encoding directly from source to the desired bitrate.