Skip to main content ↑ ↓

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: Overall MP3 and OGG encoder quality comparison (Read 5166 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Overall MP3 and OGG encoder quality comparison

I recently got a new set of speakers (Logitech Z-2300) and all of a sudden I can tell the difference between "not so high" and high quality lossy encoded music files. Now I'm on a quest to get rid of my poor quality music files and replace them with better ones.

How do I determine whether a music file is good or not? I understand it has both to do with the encoder and the bitrate (and, of course, the preset, but that information is almost always lost). Can you say that a Xing file at 160 kbps is equal in quality to a LAME at 128 kbps? And a Shine at 192 equal to a GoGo or Blade at 160? Or how would one go about automatically determining whether it is good or poor quality when you've got the encoder and bitrate? I'm quite sure that a Xing at 128 is not the same as a Fraunhofer at 128. And when it comes to the differences between libvorbis versions I'm clueless.

I would appreciate your input on this matter.

I've created a small console app (Linux), based on Mr. QuestionMan, and a Python script. This way I can recursively check all my MP3 and Ogg Vorbis files for both bitrate and encoder, and make a list from all my music that needs to be replaced.

(http://www.christian.daven.name/software/mp3qm.php)

Overall MP3 and OGG encoder quality comparison

Reply #1 –
The latest Xing encoder (togegher with Real Player) supposed to be very competitive to lame because it uses short blocks eventually. Maby you can check this encoder (need $, sigh) out. I myself can not identify the original track from 128 k mp3 from this encoder with my not so good speaker.

Overall MP3 and OGG encoder quality comparison

Reply #2 –
Adding short block feature is not enough to make any mp3 encoder competitive, especially against lame 3.9x. Blade also uses shortblocks (at the wrong place, though), and quality is really bad.

cyner> encspot, MrQuestionMan... can't tell you if the audio quality is good or bad. Especially when files are coming from P2P networks. For exemple, some nice lame 320 kbps were reencoded from a crap plugger@128 source... Encspot can't tell you the history of the mp3 stream and therefore can't detect the crapiness of some files.
In my opinion, all files detected as plugger, shine, blade should be deleted without verification. For others, test (quickly) by yourself.
Wavpack Hybrid -c4hx6

Overall MP3 and OGG encoder quality comparison

Reply #3 –
Quote
Adding short block feature is not enough to make any mp3 encoder competitive, especially against lame 3.9x. Blade also uses shortblocks (at the wrong place, though), and quality is really bad.


Guruboolez, I'm not compare that way. I just mean adding short block makes the new Xing encoder better than older one.

Also I'm glad to hear your comment on this encoder. If I want to make some 128 k cbr mp3 for mp3 portable player, should I choose LAME 3.96? The music genre is classical, my favourite.

Overall MP3 and OGG encoder quality comparison

Reply #4 –
Lame is probably the best choice if quality is your main concern. Try also lame 3.90.3: [proxima] (I'm not sure) reported ringing problems introduced by 3.96 on some samples. 3.90.3 is nevertheless slower.

ProducerPro was also interesting with some classical samples (see Roberto's test for details), but it's slow and very expensive.

N.B. ABR is probably better than CBR. Lame 3.96.1 also have a VBR mode for ~128 kbps, which is really interesting for quality (but also problematic with some samples, especially at low volume after mp3gaining).
Wavpack Hybrid -c4hx6

Overall MP3 and OGG encoder quality comparison

Reply #5 –
Thank you, guruboolez. I will try lame 3.90.3 abr 128. Maby mid bitrate should be the most concentration for lossy formats. Much high bitrate make less sense to lossy format. That is just my own opinion.

Some guys on forums of my native language (Chinese) choose MPC as there lossy format for classical music. I don't think MPC is efficient with frequency domain and the psychostic model it use is not suit for classical music (too agressive). Am I right?

Overall MP3 and OGG encoder quality comparison

Reply #6 –
Quote
Some guys on forums of my native language (Chinese) choose MPC as there lossy format for classical music. I don't think MPC is efficient with frequency domain and the psychostic model it use is not suit for classical music (too agressive). Am I right?
[{POST_SNAPBACK}][/a]


[a href="http://www.rjamorim.com/test/multiformat128/results.html]http://www.rjamorim.com/test/multiformat128/results.html[/url]

well look here
surprisingly, MPC is better than MP3 in double blind test
and its one of the best overall!

you can safely trust double blind, since its "blind" test, so the result is fair!

Cyner, if you want compability, then LAME mp3,
if you want the highest quality possible for 128kbps, why not try new formats (OGG, AAC)?

Overall MP3 and OGG encoder quality comparison

Reply #7 –
Quote
Some guys on forums of my native language (Chinese) choose MPC as there lossy format for classical music. I don't think MPC is efficient with frequency domain and the psychostic model it use is not suit for classical music (too agressive). Am I right?
[a href="index.php?act=findpost&pid=248051"][{POST_SNAPBACK}][/a]


IIRC, guruboolez prefers MPC over other lossy codecs for classical music and he's the expert with that genre.  I might be wrong.

Overall MP3 and OGG encoder quality comparison

Reply #8 –
Quote
IIRC, guruboolez prefers MPC over other lossy codecs for classical music and he's the expert with that genre.  I might be wrong.
[a href="index.php?act=findpost&pid=248060"][{POST_SNAPBACK}][/a]


He's also the expert with various formats  . That's why I ask him for recommendation.

Overall MP3 and OGG encoder quality comparison

Reply #9 –
Quote
ProducerPro was also interesting with some classical samples (see Roberto's test for details), but it's slow and very expensive.


What kind of classical music? Can you give more detailed opinion of yours?

Hope it's harpsichord. 

Actually I'll never convert harpsichord music in any one of lossy encoders.

Overall MP3 and OGG encoder quality comparison

Reply #10 –
Quote
I don't think MPC is efficient with frequency domain and the psychostic model it use is not suit for classical music (too agressive). Am I right?
[{POST_SNAPBACK}][/a]


 
Take a look at my [a href="http://www.hydrogenaudio.org/forums/index.php?showtopic=23355]MPC vs OGG VORBIS vs MP3 at 175 kbps, listening test on non-killer samples - classical music only[/url] and to additional results...

NB: I'm not an expert. I can't say what format is best, but only what format is better to my ears, at some moments on my life. That's all.

P.S. mpc at ~128 kbps is probably not a good choice for classical music. See my two olds listening tests on classical music:
http://www.hydrogenaudio.org/forums/index....topic=14091&hl=
http://www.hydrogenaudio.org/forums/index....topic=16395&hl=
Wavpack Hybrid -c4hx6

Overall MP3 and OGG encoder quality comparison

Reply #11 –
Quote
encspot, MrQuestionMan... can't tell you if the audio quality is good or bad. Especially when files are coming from P2P networks. For exemple, some nice lame 320 kbps were reencoded from a crap plugger@128 source... Encspot can't tell you the history of the mp3 stream and therefore can't detect the crapiness of some files.
[a href="index.php?act=findpost&pid=248045"][{POST_SNAPBACK}][/a]


Ah, that's right. Thanks, I didn't think of that! But, seeing that I'm not certifying hifi music but instead using a script to pick out the worst sounding pieces I can live with the occasional reencoded files. I will detect them myself, eventually.

What I'm looking for is a quick way to identify the crap and get rid of it, no questions asked. Hence my script.

Quote
In my opinion, all files detected as plugger, shine, blade should be deleted without verification. For others, test (quickly) by yourself.


Ok, sounds good. Could I have a second opinion? What about e.g. Shine-encoded music at 256 kbps? I actually got one of those from a P2P network yesterday... It sounds ok, but I don't know what the original is like.

I guess my question was unclear, but if it's possible I would like to know approximately at what bitrate a Xing, Blade, Shine and Plugger file is equivalent in quality to a LAME or Fraunhofer at 128, 160, 192 and 256, respectively. Or some guidelines, I know this is inflexible, but my script uses no artificial intelligence.
 

Overall MP3 and OGG encoder quality comparison

Reply #12 –
If you're happy with a file, don't delete it.
Shine is really primitive: it could be OK at 256 kbps, but keep in mind that the ~same quality is possible with a better encoder at half bitrate. That's why it's probably better to delete all plugger/shine/blade, whatever the bitrate, and encode CD tracks again with a better encoder.
Wavpack Hybrid -c4hx6