[budyog]

I have been extracting my CDs on to my computer using the EZ-CD extractor at a 128 bit rate. Now I am only using this to play through my cheap little system in my garage through my computer but I am wondering if I am missing a lot of quality extracting at that rate versus my choices of 128, 160,192,224,256,320,near CD-quality and CD quality.

I do not know why I chose this rate, my friend set the program up and set it to that rate and I dont remember his reasoning. It seems to be fine for my garage system so maybe that is all that matters. I am sure it had something to do with the amount of songs. Isn't it true I can get more songs at a lower bit rate?

[kfr01]

The higher the bitrate, the better the quality; end of story, more or less.

I believe that anything less than lossless is a waste of time, quality issues aside, given how inexpensive disk space is.  If you're going to spend the time archiving music, why not do it right the first time?

[Rob Babcock]

That's probably a loaded question, and my answer will probably be picked apart, but yeah- in my experience I'd say 128 is missing a whole bunch of music.  I will make the usual disclaimers; YMMV, it depends on your system, I'm not an expert at CODECs (ie there may be better ways to code that sound better than mine), etc etc.  I've found 128 to be generally pretty poor on anything even close to decent gear.  Rates in the low 200-ish range sound a lot better, at least with "easy to encode" music (ie, no acoustic piano or strings, no heavily distorted guitars).  At the 320 kbps, or using the highest quality setting and VBR (Variable Bit Rate), the sound starts to approach the original CD pretty closely.  You have to be listening pretty carefully on even good equipment to notice overt flaws in a 320 kbps MP3 that's properly encoded. 

Not that it's perfect, mind you, I'm just saying it can sound good.  Acoustic classical music is harder to "get right" with less bits.  And of course, your standards may he higher or lower than mine.  You can also legitimately point out that 320 takes so much room that you may as well just use wav.files, and to that I'd probably have to agree with you.  The only use I now have for MP3 is to burn the occasional long compilation to take to work with me (I'm the Sous Chef at a hotel/restaurant).  I keep a shelf system with MP3 playback capabilities in the banquet facility for times I have to help out over there, and it's nice to be able to have a whole days worth of music loaded up in the changer.

[LightFire]

I have been extracting my CDs on to my computer using the EZ-CD extractor at a 128 bit rate. Now I am only using this to play through my cheap little system in my garage through my computer but I am wondering if I am missing a lot of quality extracting at that rate versus my choices of 128, 160,192,224,256,320,near CD-quality and CD quality.

I do not know why I chose this rate, my friend set the program up and set it to that rate and I dont remember his reasoning. It seems to be fine for my garage system so maybe that is all that matters. I am sure it had something to do with the amount of songs. Isn't it true I can get more songs at a lower bit rate?

The reality is that a little bit over 128 kbps (using lame as a encoder) and almost nobody is able to distinguish between CD and a MP3 in a double blind test. If you feel insecure you can go all the way up to 320 kbps. For more reliable information about MP3's go to those web sites:

http://www.hydrogenaudio.org/

http://www.mp3-tech.org/

http://www.soundexpert.info/

[Rob Babcock]

I think those tests are of limited usefulness.  Sound & Vision (if I recall correctly) did an article a ways back about "CODEC-Busters", songs that, due to certain characterists that trip up most CODECs, are very easy to tell apart.  Again, I don't know which issue it's in nor do I know if their mag is searchable online, but nearly everyone in the tests could hear the codecs fall apart.  Some did better and some worse, of course.  Note that these songs weren't something off-the-wall that no one would really listen to- one was a Nirvana song.  They were songs that most of us probably have.

As you move to higher bit rates, the MP3 gets "tripped up" less, but almost inevitably you find something that sounds sour.  No, I haven't done a DBT that would satisfy guys like LightFire, but I've conducted single blind tests on volunteers with very good level matching (I don't have any way to do it without the person doing the actual test seeing the switch, although I know there's software to let you do this on a PC.  I don't have that software, nor can I recall the name).

As for DBT, I'd feel pretty confident in being able to pick the MP3 out vs the wav.file @ 128 kbps.  If I could pick the music I'd consider "racing you for pink slips."   At 320 kbps, properly encoded, that would be a potentially difficult test, depending upon the testing criterion.

[LightFire]

Those things that trip the codecs are the so called artifacts. They are not really related to quality of sound and more related to programming errors. Inability of a codec to process certain sounds under certain circumstances. They can be corrected as they are found, by the programmers that write  the software. That's why I use Lame, because it is constantly being developed. So if there is a problem today it will probably be corrected tomorrow. As Bob mentioned before,  using high bit rates is a good way to avoid them. MP3 at 320 kbps is still a lot smaller than a lossless codec.
It is a lot easier to perform blind tests with software than with hardware. I performed some with the help of some friends ripping the samples to wav with iTunes and then converting them from wav  to different mp3 bit rates (again using iTunes). Organized the samples in a play list order only known to myself and burned a CDDA from the list. They took the CDs home to listen in whatever environment/equipment they wanted, but promised not to extract wav files and compare visually. The 3 of them failed completely. The standards I used were NOT ABX double blind (that would be a lot harder).
I have to admit I failed similar single blind test as well.

[kfr01]

The reality is that a little bit over 128 kbps (using lame as a encoder) and almost nobody is able to distinguish between CD and a MP3 in a double blind test. If you feel insecure you can go all the way up to 320 kbps. For more reliable information about MP3's go to those web sites:

http://www.hydrogenaudio.org/

http://www.mp3-tech.org/

http://www.soundexpert.info/

You, like all the other lossy compression fans, fail to see the point.  Drive space is dirt cheap.  Generation loss is real.  Why waste time archiving in less than lossless quality?  If I want to burn a CD ten years from now, long after the disk itself has been lost or scratched to death, do I want a compressed copy or an uncompressed copy?  The answer is painfully clear, in my opinion.  I could, frankly, care less about the one hundred million ABX tests that John Doe has completed with his buddies.  It is about the purpose of archiving, not the audible difference between two 30 second clips.

http://www.newegg.com/Product/Product.asp?Item=N82E16822148140

Seriously, $100 for a state of the art 320GB HD.  What in the world is anyone's motivation (other than to fit an entire music collection on their iPod) to archive with a lossy compression format?  I simply don't see it.  I don't care if it doesn't make a testable audible difference...  what's my motivation, man?

[LightFire]

The reality is that a little bit over 128 kbps (using lame as a encoder) and almost nobody is able to distinguish between CD and a MP3 in a double blind test. If you feel insecure you can go all the way up to 320 kbps. For more reliable information about MP3's go to those web sites:

http://www.hydrogenaudio.org/

http://www.mp3-tech.org/

http://www.soundexpert.info/

You, like all the other compression fans, fail to see the point.  Drive space is cheap.  Generation loss is real.  Why waste time archiving in less than lossless quality? 

Well. In my case compatibility plays a role. It is important to me to be able to play mp3 CDs in my car. So I don't need to replace CDs in long trips. One mp3 CD at 320 kbps still can hold more than 5 hours of music. And once I have those CDs burned to use in the car I would like to play them (or the songs) in the other equipment too: portable player and sound system. It is very hard to find players that support the lossless codecs. And the fact of being lossless is irrelevant for listening purpose if someone (99.999% of all people) can't differentiate the quality levels anyways.

[kfr01]

Well. In my case compatibility plays a role. It is important to me to be able to play mp3 CDs in my car. So I don't need to replace CDs in long trips. One mp3 CD at 320 kbps still can hold more than 5 hours of music. And once I have those CDs burned to use in the car I would like to play them (or the songs) in the other equipment too: portable player and sound system. It is very hard to find players that support the lossless codecs. And the fact of being lossless is irrelevant for listening purpose if someone (99.999% of all people) can't differentiate the quality levels anyways.

I'd rather cater to my primary system for the long term, rather than secondary systems (e.g., car stereo) for the short term.  If car audio is important to you, buy a deck or adapter with line level inputs. 

It is hard to find players that support lossless?  Since when?  .wav?  .flac?  .apple lossless?  Let's see, that covers at least ALL of the best selling devices on the market.

I still fail to see the motivation to purposefully reduce resolution when archiving.

[Rob Babcock]

Let's say, for the sake of argument, that most of the audibly crappy stuff in CODEC's are "glitches" that can be patched.  What happens when you rip, say, a thousand CDs and they don't patch your codec for a year?  Or even til the next day!   Do you re-rip all one thousand discs again?  What about the discs you've already burned- do you reburn them all again?  I'm not trying to be argumentative, but no "patch" or update to LAME has ever resulted in me being able to make 128 kbps discs indistinguishable from the original 44.1/16 bit original.  Although admittedly I'm not the most savvy CODEC guy on Earth, although I'm pretty competent in computer matters.

I guess that's the whole point- Bill Gates & MS have changed the paradigm to the point that all consumers are really just Beta testers.  Playback of wav.files may get better over time, but my wav.files don't need to be updated over the years.  The same 20 year old CDs I have now will still work the same in another 20 years, with no need to "update/patch" them.

MP3 is almost completely useless to me simply because storage is laughably cheap now.  Sure, it's still handy for portable use, like my example above (taking stuff to work). 

NOTE:  my above example is based on my brother's sad case.  He ripped about 600 CDs to his HD only to discover after he did that his combo of codec/bitrate resulted in subpar sound.  Now he's not a dyed-in-the-wool junkie like most of us, and he doesn't have particularly good hearing.  But he is extremely PC savvy (studying at a tech school now).  Unfortunately he must not have done enough research on the codec he was using (I think LAME, don't know any specifics).  It took so long to do it that he decided it wasn't worth the effort to redo them.

[Rob Babcock]

One thing a person can do, of course, is to use FLAC to rip all your stuff, then use that file to create MP3s for portable use.

[budyog]

Thanks LightFire for the links and everyone else who replied.
 It looks to me after reading the links and being that I am happy with the sound I am getting in my garage system at 128 that is the way I will continue. My only thought is that someday I may want to do something with my home system using I think something like the Olive music server (I think that is what that is for, correct me if I am wrong) for some odd reason. I preferr to play direct from Cd's for critical listening, but for a party or something it might be cool. I think it is similar but far superior to the Omni system using SimpleCenter in my garage.
I will have to do some more reading to see the difference in disc space used with 128 vrsus CD quality. I would have to beleive the is a lot of difference. I have extracted 700CDs/8000? songs.