Hi all!
I have a question that I need a sure answer to, and would need reference to if possible.
I'm having a debate on the subject going from wav to mp3, where a person says that "going from 44.1kHz - 16bit wav to a 320kbit/s mp3 will still provide 44100 samples at 16 bit, you will only lose some accuracy and size, not quality".
This gets me really confused.
That person also makes an equivalent to bmp vs jpg, which to me IS a clear downstep in quality even if the eye can't see it.
PLEASE, give me the real and un-questionable facts on this! :)
Topic Tags
Comments
Yes, this is my view also. Quality equals accuracy to me, and is
Yes, this is my view also. Quality equals accuracy to me, and is corresponding to the correct values of the recorded samples.
If you compress the size, you'd have to compress the original information by taking away som info.
You take away this or that part of the original data depending on how hard you compress, and replace it with an algorithm to decode the remaining material, and add new calculated "should-be-data" based on what's left instead.
You may be able to spit out 44100 samples a second at 16 bit depth, but it just CAN'T be 44.1kHz at 16 bit, of the original recorded material, can it?
How much of the original data is left at 320kbits/s mp3? Is it 20% original data and 80% calculated that we hear?
Take a look at an mp3 through a spectrum analyzer.......the high
Take a look at an mp3 through a spectrum analyzer.......the high end nosedives above 10k. To me, it's the digital equivalent of a cassette. Can the average listener hear it or do they care? ipod sales are brisk, so apparently not. I'd rather listen to a .wav, but then we engineers are paid to hear subtle differences.
MP3 Hi, new here. I am wondering if the MP3 playback is equed
MP3
Hi, new here.
I am wondering if the MP3 playback is equed a bit to cover the high end roll off. It can't replace everything but it can boost the highs and pull the lows. Plus the earbuds most people use have bad low end reproduction anyway. All of that might be balancing it out a bit. I know when I play back an MP3 on my studio monitors the lows are HUGE. On and Ipod--they aren't.
I have a few thoughts about MP3's. First off, they are Lossy (m
I have a few thoughts about MP3's.
First off, they are Lossy (meaning not as good)
They use a few tricks to persuade you into thinking they are *CD* Quality.
First off, they sum the high frequencies into mono. The frequency varies depending on your bit rate, and algorithm, but if you listen side by side with a 16/44.1 you can hear it. Second they use perceptual coding which analyzes which frequencies are used the most through the song and gives them a smaller identifier to save space.
There are more tricks, I'll dig out my resources and look them up again, that's all I remember off the top of my head.
-Ryan-
mp3 for cd back up when you convert back from mp3 to wav is the
mp3 for cd back up
when you convert back from mp3 to wav is the file restored to its origanal form or is 80% lost for ever? in other words is mp3 a good way for me to back up my cd collection? i want to back yhem up because the (cds) keep getting scrached but i dont want to use all my hard drive space( i have over 300 cds)
thanks D.J.
Try Lossless encoding. You don't get the extreme file size redu
Try Lossless encoding. You don't get the extreme file size reduction of mp3, but you can save a bit of space (not too much :( ).
I personally archive my CD collection a few ways. First, I rip WAVE files as an exact copy of the CD (I add CD-Text in Nero). Then, I burn 2-3 CD-R copies of this identical WAVE copy (1 for the 400 disc changer, one for the two 10-disc changers in my ride, and 1 for the jam-room PA). Then, I put away the original CD, and don't mess with it (True archive). Then, I encode these remaining wave files into 320kbps mp3's, and add these to my MP3 jukebox/Music folder. Then, I delete the Wave files off my HD.
This way, I have 3 WAVE CD-R copies (w/ CD Text), the original CD which is not used again (unless I need to re-rip), AND I have smaller "descent quality" 320kbps MP3's for my PC's jukebox. I also have my 45Gig music Folder on 3 different HD's, and on 2 sets of DVD-R's for another layer of security. Granted, I could ALWAYS go back and re-rip from the original CD's, but that would take FOR EVER!!! Backing up the MP3's (in addition to having a non-compressed archive) will save you a TON of time if your MP3 HD dies.
With hard drives as cheap as they are, I'd probably favor WAVE files over lossless compression. OTOH - MP3 is not an ideal archive format, but I have burned MANY CD's from my 320kbps mp3's, and they sound just fine to me :) ...
:cool:
Interesting question. But I guess it might boil down to how you
Interesting question. But I guess it might boil down to how you define quality and accuracy. Size I´m sort of clear about.
This is my impression of things.
First, there are two ways to compress things : lossless or lossy (I guess you could have more cathegories of course). Zip-files are lossless, compress a file and decompress and you should get exactly the same thing. mp3 is lossy, compress a sound file and decompress it and you are not guaranteed to get out exactly the same thing. This is one reason why you would not distribute programs as mp3-s, there is no such thing as "almost the right set of instructions".
For sound though, mp3 is one of several possible ways to reduce the "size" of a sound file without disturbing the brain hearing the music too much. The last part is the important thing, the brain is not disturbed too much. So any audio algorithm is tied to how the brain actually interprets and "hears" music.
From a pure technical point of view, any mp3 will probably distort the sound, and from a technical point of view this is a degradation of quality. But if the brain does not hear the difference, from a phsycological point of view it might have the same quality. So yes, both could be correct, you do not lose quality and you do lose quality. Back to the question what definition you have.
44.100 kHz x 16 bits is 705.600 bits per second, or 705.6 kbps. 320 kbps means that about half of the information has been removed (you keep 45%). From a technical point of view this is a loss of quality. But your brain might even hear this as an IMPROVEMENT in quality.
I can´t really figure it out either.
<< Edit: I forgot that it is actually two channels we generally talk about. Stereo means 1.411.200 bits per second, that means you remove about 80% of the size >>
Gunnar.