Any audiophiles here?

Yes, but the bitrate still controls how much information in these frames is retained. That's why further on in this chapter a simple sine wave is shown displaying how a 64kbs retains less information than a 128kbs sample. I guess mine is a very basic view of the process- I deal with this with my Information Processes and Technology students (final year of high school) and they don't have to know the process in great detail. I basically just tell them that noises we can't hear are removed and show how the sampling rate works- using the sine wave example.

cobb2 wrote: Yes, but the bitrate still controls how much information in these frames is retained. That's why further on in this chapter a simple sine wave is shown displaying how a 64kbs retains less information than a 128kbs sample. I guess mine is a very basic view of the process- I deal with this with my Information Processes and Technology students (final year of high school) and they don't have to know the process in great detail. I basically just tell them that noises we can't hear are removed and show how the sampling rate works- using the sine wave example.

You mean image 2-3 on this page:

http://oreilly.com/catalog/mp3/chapter/ch02.html

That image is actually not correct. It shows the effects of quantisation at different sample rates, which - as Dean also said - are not to be confused with bit rates of MP3 streams.

The internet ... sometimes even O'Reilly can be wrong.

EDIT: They're actually clarifying the difference later on ("Bitrates vs. samplerates") so they're not actually wrong, but merely inconsistent.


Edited by Mr ProgFreak - July 04 2009 at 04:57

You guys know far more about this than I do 

For a typical 16bit/44.1khz WAV file the bitrate is 1411kbps, and that's what Windows will display to you. The important thing is that an MP3 file at for example 256kbps is not just a downsampled version of the original 1411kbps WAV file. The reduction in size is achieved primarily by looking at the frequency distribution of the signal and leaving out those parts that the human ear/brain can't perceive because they're masked out by other parts of the signal. Of course there are limits to this method of reducing size ... whether it's successful depends on many factors, most importantly the complexity of the signal and the resulting bitrate. If you encode a file at 128kbps, for most signals too many parts will have to be removed in order to accommodate the small file size. 128kbps is less than 10% of the original bitrate. If you encode at 256kbps (about 18%), you usually won't be able to hear a difference, or in other words: The mp3 encoder operates in a range that allows it to only remove parts of the signal that we can't hear, while still achieving the desired bitrate/file size.

I think this is the real problem: Some people simply refuse to believe that if you remove 82% of the original, the result can still sound exactly like the original (judged by human ears).

Take into account though that even when you apply only lossless compression, most tracks can be reduced to about half their original size. Or let's say 60%, to be on the safe side. So adding the lossy compression to the equation actually only adds an additional ~ 40% of compression. You can also take into account that today tracks are usually encoded using VBR (variable bitrate). This means that for each frame the mp3 encoder can analyze it and decide - depending on the complexity - which bitrate to use. The encoder can also select bitrates for left/right channels separately. In a nutshell: There are many ways the mp3 encoder can use to optimize the compression rate.

The bottom line for me: I have yet to hear a properly ripped ~256kbit MP3 that I could tell apart from the CD. I haven't done any real double blind tests, but I *never* found myself listening to such MP3 files and thinking "that sounds distorted/harsh/hollow, let's put in the CD".

^ exactly. Personally, I don't ever think about that when I listen to music. A few years ago when I started to rip all my 1000+ albums as MP3, I was concerned ... but whenever I listened to the tracks - especially those that I had listened to on CD countless times over the years and knew in and out - I found no difference to the originals, and that really encouraged me to accept MP3. 

^ No doubt, some people will invest a lot of money into such esoteric devices.I never would. Today even your average low cost computer mainboard contains high quality DAC circuitry. Unacceptable by most audiophiles, I know. But again I trust my ears.

Think about it: DAC circuits have been around for 30 years. Today computers are much faster, and more importantly: Manufacturing processes have been optimized. *Maybe* in the late 80s/early 90s it would have made sense to buy a $4000 CD player, but today? Come on.

I have an ASUS P5N-E SLI motherboard. When I first got this new computer I also used my Creative X-Fi soundcard, but I soon got rid of it because of driver problems (I use Vista 64bit).

Ever since I've been using the onboard Realtek hi-def sound card ... and I'm very happy with it. There is absolutely no audible interference from any component of the computer ... IMO it is simply an esoteric belief of many audiophiles that they have to protected their equipment from each other all the time ... every gadget needs its own power supply, the power supply needs power filters, the speaker cables need active shielding etc..

Come on people ... trust your ears!

^ I really am. Just finished listening to this fine album on my computer:

http://www.emusic.com/album/Kotebel-Ouroboros-MP3-Download/11447196.html

Heavily recommended to ... anyone really, I cannot imagine any Prog fan not liking it. And it surely put my Logitech speakers to the limit.

I purchased Wavering Radiant both on vinyl (excellent gatefold cover!) and as MP3 from eMusic.com:

http://www.emusic.com/album/Isis-Wavering-Radiant-MP3-Download/11412845.html

Maybe I'll get around to conducting a little experiment tomorrow: First I'll listen to the album on vinyl on my hi-fi, then I'll listen to the mp3s on my computer. I'll let you know if I hear a difference!

[QUOTE=Dominic]So, considering that we're on the subject of audiophilia, my longtime headphones have finally betrayed me; anyone have some phones they're especially proud of or feel like making a special recommendation?


Sennheiser 650. Trust me. :)


Edited by Rottenhat - August 06 2009 at 13:30