• Life Advice & Visual Arts Welcome Guest
    Posting Rules Bluelight Rules
    🌌 Share Your Dreams! 🌌
    📸 15 Years of Photo Contests 📸
  • LAVA Moderator: Mysterier

Question about sound compression (VBR)

StarOceanHouse

StarOceanHouse

Bluelight Crew
Joined
Jul 16, 2003
Messages
7,522
If I were to burn an mp3 with VBR onto a CD and then rip it again with the same VBR would I lose sound quality?
 
does ripping create an exact copy of the file? i imagine that's what you're asking, so i don't know. can't you compare file sizes? if they're the same size down to the byte, you have a perfect duplicate. if it's not an exact copy of the file, you're working with lossy so you're technically losing quality.
 
I'm asking if I were to "convert" an mp3 with VBR 0 to WAV and burn onto a CD then rip it again using the same VBR (0) would there be any loss. I already know VBR is lossy but what I'm asking is if I lost any sound quality during the 2 conversions.

I had an mp3 that I burned onto a CD a while back. I can't find the mp3 anymore so I'm ripping the CD.
 
oh, you are losing sound quality to the best of my understanding. the data in that wav creates the same sound quality as the original mp3, and now you are going to compress that wav data using lossy compression. "to the best of my understanding" means i don't think that wav file contains unnecessary data that can perfectly be lost; it's not like reverting back to the original mp3. i don't know how perceptible the loss would be.
 
Last edited:
Going from mp3 (lossy) to wav (lossless), known as transcoding, is looked down in the p2p file sharing community. On some private sites, it will earn you a ban because, yes, there is some loss.

Practically speaking, I couldn't tell ya how percetible it would be since it depends on your own auditory processing and equipment. Anything sounds crappy playing from a potato. ;)
 
afaik the software that does mp3->wav keeps the encoding and just adds a wav header to the file so if you convert that back to mp3 with the same algorithm it was originally ripped using then you should end up with a carbon copy of the original. an easy way to confirm that is to just do it and see if the final file's size is identical to the original's size.
 
thujone said:
afaik the software that does mp3->wav keeps the encoding and just adds a wav header to the file so if you convert that back to mp3 with the same algorithm it was originally ripped using then you should end up with a carbon copy of the original. an easy way to confirm that is to just do it and see if the final file's size is identical to the original's size.
Click to expand...

I don't think this is correct. Source?
 
the encoder source code is well documented and explains what it's doing. the quantize.c file is where the magic happens in regards to the audio data, the other files deal with other information contained in the .mp3 package that isn't related to the audio itself. the quantization math itself is a headscratcher but obviously the methods don't magically change after compilation so the output should be what is expected. there might be some slight fluctuation in before and after file sizes depending on how the converter software implements the encoder so tag data and padding might be removed or added automatically but that won't affect the actual audio data quantization.

Assuming the file does not get re-encoded at a higher VB mode (goes from 0-9, lower = better) than what the original source was encoded at, the worst thing that will happen is the algorithm will create extra data that will add to the file size but change nothing about the audio quality.

FWIW i'm just going by what the code says and what I know about the data structure of an mp3 file, quantization maths is beyond me so you would have to go to some sort of engineer for that but good luck understanding them, lol
 
I've always been told lossy to lossless is always bad, but perhaps that is a generalization and not applicable to mp3->wav.

At the end of the day, I'm not a true audiophile so perhaps I would never be able to tell the difference in quality anyhow.
 
in this case it doesn't affect the sound quality assuming the final file is re-encoded at an equal to or higher than sample and bit rates than the source files but in general yes upsampling is awful because you end up with a file that says "hi, I'm high quality audio" but the actual audio source inside it is of a lower quality. for archiving purposes, it's a nightmare.
 
You must log in or register to reply here.
Top