MP3 Degradation, an experiment.

[VirtualMark]

The last two don't have the same sample rate, so they would lose the higher frequencies. And some codecs low pass the lower settings automatically, as most people can't hear that high anyway. Did you check this before you posted?

A good way to actually hear whats missing from your audio is to subtract one file from another, then you get a track just made of the artifacts. But i personally can't tell the difference between a 320mp3 and a wav, and i'd like to meet someone that can. All this 24bit 96khz stuff on blu ray seems a bit overkill, i can't really tell the difference there either.

[Trebek]

Wow this was a good read, pretty crazy to actually see some results graph wise. I'm not sure if this was mentioned but I find it interesting, MP3's are compressed as shit I believe its an 11:1 ratio. Figured I would get that out haha commit my 2cents to the post.

[Depone]

The last two don't have the same sample rate, so they would lose the higher frequencies. And some codecs low pass the lower settings automatically, as most people can't hear that high anyway. Did you check this before you posted?

A good way to actually hear whats missing from your audio is to subtract one file from another, then you get a track just made of the artifacts. But i personally can't tell the difference between a 320mp3 and a wav, and i'd like to meet someone that can. All this 24bit 96khz stuff on blu ray seems a bit overkill, i can't really tell the difference there either.

I have no idea why that is, i used the same export settings etc, just changing the mp3 rate...

[like spinning plates]

What does this mean with regard to VBR?

[Depone]

What does this mean with regard to VBR?

Variable Bit rate.

It means that the quality of the mp3 will change during the most complex and simplest parts of the mp3. increasing the kbits when theres lots of complex sounds, and it lowers when its simple. Its another way of making an mp3 smaller. for max quality, stick to cbr (constant bit rate)

[staticcast]

for max quality, stick to cbr (constant bit rate)

actually, for a given compression ratio (ie 'kb/s' value), all other factors being equal, VBR will offer better perceived quality.

[Depone]

for max quality, stick to cbr (constant bit rate)

actually, for a given compression ratio (ie 'kb/s' value), all other factors being equal, VBR will offer better perceived quality.

It wont as 320kb/s is the maximum for mp3's constant or vbr.

[Kamex]

Great Post!

But im curious on the comparison on different bit rate mp3 from the source and ones that are ripped from sites

to show people who get there "320's" off youtube and feel the need to loop them and play them out.

[antipode]

Cool thread. I'm having audio engineering class flashbacks

Check the snare on this tune, sounds like mp3 compression being used creatively.


Also, doesn't mp3 compression mess with the stereo field aswell?

[abZ]

Again, the point is that these artifacts sound "nice", whereas mp3 artifacts don't.

Anything to do with "nice" is always subjective. Just heard for the nth time that in a blind test, younger people prefer the sound of mp3 (and mp3 compression) over other media, ie vinyl, cassette, and cd.

Sounds crazy right? But whatever you get used to, that's what you like. I've been there.

There was a time early in mp3 when I was exclusively listening to 128 -192, and did develop a taste for the high freq crunch.
==

Side note, congrats on the vinyl release, 110% about time.

Yeah we have endless threads on bit crushing on this forum and people can't understand how someone could like a lo-fi mp3 over lossless? Think about it.

[Depone]

Forgot i made this thread and wanted to bump it because I was recently sent a 92kbits mp3 from a producer... he didn't realise the damage he was doing.

so BUMP

[sunny_b_uk]

Forgot i made this thread and wanted to bump it because I was recently sent a 92kbits mp3 from a producer... he didn't realise the damage he was doing.

so BUMP

was it a good beat
also i wouldn't mind seeing a comparison of a 350 kbps or a 450kbps mp3 since those two higher options are in FL Studio to export at that rate.

[VirtualMark]

Forgot i made this thread and wanted to bump it because I was recently sent a 92kbits mp3 from a producer... he didn't realise the damage he was doing.

so BUMP

was it a good beat
also i wouldn't mind seeing a comparison of a 350 kbps or a 450kbps mp3 since those two higher options are in FL Studio to export at that rate.

92kbps?? what was he thinking.


as for those higher settings, i'd leave them. there is a free format mp3 which goes past 600k but nobody uses it. best to stick to 320 as its a standard format. if you want higher quality just use wave files, not that many people can even tell the difference between 320mp3 and wave.

[Artie_Fufkin]

Sometimes when I upload tracks to soundcloud I can hear a sort of ringing in the sub bass, anyone else know what I'm talking about?

The low end in these examples similar, even down to the lowest bitrate.

I get that in audacity! I haven't figured out what it is but I'll put a sample of sub in audacity and it will play with this ringing harmonic, but when I just play it in windows media player or vlc it doesn't have it.

I just want to echo what someone said about encoders. They are not all equal. The encoder itunes uses is junk I've read.
Just wondering: what did you use to make the graphs and which encoder you used depone?

[Depone]

Sometimes when I upload tracks to soundcloud I can hear a sort of ringing in the sub bass, anyone else know what I'm talking about?

The low end in these examples similar, even down to the lowest bitrate.

I get that in audacity! I haven't figured out what it is but I'll put a sample of sub in audacity and it will play with this ringing harmonic, but when I just play it in windows media player or vlc it doesn't have it.

I just want to echo what someone said about encoders. They are not all equal. The encoder itunes uses is junk I've read.
Just wondering: what did you use to make the graphs and which encoder you used depone?

Very true!
I use LAME codec to code mp3s.

For this example, i wanted to use a pretty poor sounding mp3 encoder, so I went with adobe soundbooth (Thanks adobe for my month trial for this!)

Might try one with LAME and see how it stands up.
And then ill do an m4a one (the format itunes purchases are) because they have a better frequency range at smaller bitdepths. IE better encoding of compressed audio

[Artie_Fufkin]

The last two don't have the same sample rate, so they would lose the higher frequencies. And some codecs low pass the lower settings automatically, as most people can't hear that high anyway. Did you check this before you posted?

A good way to actually hear whats missing from your audio is to subtract one file from another, then you get a track just made of the artifacts. But i personally can't tell the difference between a 320mp3 and a wav, and i'd like to meet someone that can. All this 24bit 96khz stuff on blu ray seems a bit overkill, i can't really tell the difference there either.

I think I tried this before, but mp3s have a very small little bit of "silence" added at the beginning so they didn't line up. Could try moving the wav to be in time with the mp3 though...-goes and does that-

Edit: Alright, well when I had thought to do this before, I didn't really believe I'd be able to match the waveforms up, but I think I got it just now. After matching them up, inverting one of the tracks and then mixing them(doing this in Audacity), I was left with a track that was pretty quiet and just sounded like white noise. It was mainly the percussion sounds, so I'm guessing the artifacts are more present in drum/noisy sounds or is there dithering in mp3 encoding? It didn't even have that "glitchy"/watery sound, just noise. I encoded the mp3 at 320kbps. It was peaking around -27 dB for most of the song, except for at the very end; there was a peak at about -19 dB. Not sure why that was there because there wasn't a peak at the end of the wav or the mp3.... (It looked like I matched up the waveforms perfectly, sample by sample, so this shouldn't be phasing....)

I'm finding myself really curious about this.... I'm going to try varying some things, like the bit depth/sample rate of the original wav, the bitrate of the mp3, and also just trying it with a sine sweep.