I recently became again a fan of CD-quality music!

Yes, loosy formats as the beloved MP3 are great for archiving music. They allow you buying a hard disk and filling it with hundreds or thousands of audio files. Nonetheless in most situations I enjoy music, lossless formats and CD-quality music is something I can't live without.

I'd like to know your experiences enjoying lossless and CD-quality formats. Why do you prefer them, and in which situations lossless is a must, or you can perfectly go with loosy formats.

Any ideas, comments or suggestions welcome!

Just because the technology exists, and I can encode that way. (To be sure it makes it possible to have an archive that can be quickly transcoded to whatever lossy format I might need for whatever player or device I might have now or in the future without re-ripping, but that's probably less important.)



I no situation, I suppose. It's just I like to.

The converse is that in any situation where I haven't much room lossy is a "must". That's why I have it on my iPod.

They're actually a terrible idea for archiving music in the true sense of the word "archive". You can't reconstruct an exact 1:1 copy of the original source material from them.

Have you performed ABX testing to prove this for yourself?

I'm perfectly happy to go lossy with all of my listening material as I've carried out extensive ABX testing to prove to myself that MP3 encoded with LAME in VBR at -V3 is transparent to me in all listening situations with all of the source material I've tested.

Cheers, Slipstreem.

ivalladt: I don't really get what it is that you're enjoying / not enjoying.

A simple example: Kind of Blue by Miles Davis. I downloaded the album, MP3 encoded, VBR 192 Kbps. The best jazz music ever recorded. But this copy had a problem: Double bass can't simply be heard. And on the first track of this album, double bass is the instrument carrying the melody. So I bought a CD copy, and the double bass is back!

Another example: MP3 encoding is by some means removing harmonic content. Get an MP3 album to your car, and engine noise of the car is enough so music is not listeneable. The same album, CD quality, is listeneable again in the car, just because the whole harmonic content is back.

Another example: Bach by Glenn Gould. One thing that makes his recordings great is listening to him singing the melody while he's playing piano. This listenin to CD-quality. Encode it to MP3 and you can say goodbye to Gould's voice. It simply disappears.

How do you know the copy you bought was from the same master as the source of the music you downloaded?

Pure nonsense.

Pure nonsense.

Please read the Terms Of Service, particularly #8:
http://www.hydrogenaudio.org/forums/index....974#entry149481

I want to see some double-blind test results.

Definitely sorry I disturbed you.

For me, having spent hours listening to the same So What! from Davis' record MP3 and CD and clearly noticing the difference from the first heard is equivalent to any blind test. Sorry if it isn't useful for anyone except me.

Feel free to delete this thread from this forum.

It most certainly is not equivalent to conducting a blind test.

"Lack of bass" is not an artifact caused by mp3.

If you want to participate in this forum you must play by the rules. This means that statements about sound quality must be backed up by results from blind listening tests.

I hate to beat a dead horse but what was the point of this thread? Lossless is great for having a 1:1 digital archive but lossy encoders have more than proven themselves over the years as providing great transparent files for general listening (ie you listening in your house using your computer) and portable listening (in the car, with a portable device, in public, etc.).

Now you could provide us with some blind ABX results and then lets the developers on Lame know that their encoder is performing so badly that it is removing actual instruments and voices from tracks. Hell, I would want to know if my lossy encoder was doing that with your music especially since no one else in the world is reporting the problem. That sounds like a serious issue to me.

Either way, I don't see the harm in conducting a blind ABX test just to prove your point about quality. You should provide us with the version of Lame you are using in that test. That was you can redeem yourself.

I can't imagine a music archive consisting of lossy files. I use lossless because this way I'm freed of making the choice, what to throw away. So, instead of second guessing my choice, I can start enjoying the music. In the past I have burned my fingers with lossy video, no more. "Archive" is the keyword here.

I have a lot of music. Mainly mp3 lame encoded. Lately, I have been converting all my cd's to wavpack. I suppose that anyone who enjoys music would want an exact duplicate of their discs. Thats what I tell myself anyway.

Thesedays, I only listen to mp3 in the car. My car engine does not make instruments disappear thankfully!

Edit:

ivalladt: please provide samples for testing.

You are right, Kind of blue is one of the very best jazz ever produced.
If you can’t hear the double bass I’m wondering what’s going on.
Are you exaggerating or do you have the most lousiest MP3 player in this world?


You are right, if differences are so obvious why ABX it?
At the same time you are wrong, prejudices can be so strong that we are simply not able to hear the facts. That’s what making ABX so useful, it won’t yield absolute truth but at least it cancels out prejudices. Beside technology is rather complex, maybe the differences are not due to MP3 design but because of a very lousy implementation.



I appreciate but what is wrong by putting it to test?
May I suggest?
Take out the CD and rip the first track to WAV so you have a true 44.1 kHz/16 bit copy on your hard disk. Lets call this A
Rip it to MP3 say 192 kbs VBR, lets call this B
Listen to A
Listen to B
Now ask a friend to play 10 time this track and ask him to choose A or B at random and let him jot down his choice. Lets call this X
You, of course, don’t know what he has chosen but for each track you make a note X=A or x=B
Then compare the notes and let us know the outcome.

I would not find strange the lack of quality while listening to my MP3 if I weren't using a good soundcard. Here's what I am using:

- At home: RME Digi96/8 which are considered to be the ones with best D/A converters around, connected to a pair of Behringer MS16 speakers, not the best around but nice enough.
- At work: A Sharp minidisc connected via S/PDIF to the laptop, using a pair of Sony speakers, again not the best around but more than nice enough.



You're right. I think what I am doing is extracting myself two MP3 copies of the CD quality file, one with production-ready LAME and another with testing LAME. I am comparing them. Also if I find who helps me, I'll do ABX. I'll get back to you with the results. Definitely maybe I was not doing things correctly. "The truth is out there."

Pure crap. The reason you can't hear stuff in the car is because of the engine noise/road noise/crap stereo, not because it's an MP3.

In fact MP3 is more suitable in a car environment because all the noise masks a lot of the sound anyways....especially when you have a noisy exhaust

I'd say it depends on your hearing abilities. Some can hear the difference between lossy and losless, some can't. Another issue is the low "mastering quality" (I call it that way in lack of another term) which makes me ask if going lossless on bad sounding original CDs is worth the hd-space.

From my experience: I cannot tell high quality mp3 from flac when abx'd. I learned that my ears need time to adjust - given this time, I can easily blind-spot even high quality mp3 from flac.

ivalladt

Glad to hear you are willing to test, you captured the spirit of this forum!

However, I don’t understand why you are comparing MP3 with MP3.
You’re claim is that there are big differences between CD and MP3 (bass gone!).

As you have LAME (excellent choice) I suggest rip the CD to WAV.
CDs contains audio in PCM (pulse code modulation) with a 16 bit resolution and a 44.1 sampling rate.
WAV is audio in PCM (pulse code modulation) with a 16 bit resolution and a 44.1 sampling rate.
So WAV is straight CD quality
Rip the same track to MP3 say 192 kbs VBR.
If you are willing to put your claims to test, you should compare 100% CD quality with MP3

Having a track in WAV and in MP3 you have the right experimental setup:
- both files are on the same computer
- both files will be played on exactly the same hardware
- you compare CD quality with MP3

Keep us posted!

So what are you telling us? That you failed to abx flac vs. mp3, but you still claim to be able to hear the difference?

I claim exactly that but yes, being that it's not me who encoded the MP3 being tested in the past, I'll encode it myself using LAME VBR192 and do ABX. Sure I'll learn a lot and stop praying "nonsense" and "crap".

I guess that by "LAME VBR192" you mean LAME 3.97 in VBR at -V2?

Unless you have exceptionally good equipment, ears like a bat, and happen to have stumbled across a new problem sample, I'll be personally amazed if you can hear any difference at all. Awaiting your results with interest though.

Cheers, Slipstreem.

Yep.

Yes!

So you simultaneously can and can't ABX the same things? Thats a new one on me!

Cheers, Slipstreem.

Is it April the 1st or May the 2nd? I can't tell anymore...

Edit: You may find mp3's sound better when the moon is waxing rather than waning.

You do know that ABX tests can last as long as the song, right? You don't have to try to judge within the first 10 seconds of the song playing. I guess the placebo affect hits everyone though and this is a prime example.

ivalladt, I am glad you decided to conduct a blind ABX test and I am awaiting your results. It isn't very common that someone comes in and decided to correct the errors of their ways. Lossless is great for archiving, no one is arguing against that. But lossy is more than acceptable when it comes to actually listening to music. Good luck with the ABX tests.

If differences are so obvious to everyone else too, then an ABX test is not necessary.

Otherwise, it is.

You do have a point, why reinventing the wheel?

On the other hand: a majority vote is a majority vote and next year they vote different.
ABX is simply about a method to cancel out your own prejudices.
Nothing wrong applying it all of the times,