Crusher said:
I didn't make that statement at all, stop trying to put words in my mouth. Please re-read what I wrote, then look at
the sourceforge page for LAME for confirmation of what I said. I find it hard to believe one can read those forums and not find a noticable amount of LAME f@nboyism in the comments.
The LAME sourceforge page has no relevance to anything in this thread, and certainly not to Hydrogenaudio. Period. The only f@nboy here appears to be you (how tiresome).
HA is a community that strictly adheres to objectivity via double blind testing. People who makes assertions regarding sound quality without backing them up with ABX tests are first warned and eventually banned.
If you want an example, my search in their forums for "blade" turned up a poll titled "Which MP3 codec do you prefer?" and the options were:
1) LAME 3.90.3
2) LAME 3.96.1
3) yet another LAME version (please specify)
4) another MP3 codec than LAME (please specify)
As if the poll itself wasn't biased enough, 97.5% of the people who responded chose one of the first three options. That hardly sounds objective to me.
Look, it isn't that Blade is bad, it's that it's
so bad that it is common to use it as a low anchor when performing ABX testing of multiple samples! Blade is so easy to ABX that it's boring.
Given that tests using solid objective methods have showed that LAME is, for a fact, the best MP3 encoder I see nothing wrong with that poll. You even have a chance to vote for your encoder of choice, Blade...
What bitrate are these supposed tests done at? As I mentioned before, Blade cuts out high frequencies at 128kbps and 160kbps (and to a lesser extent at 192), which is why I said to use 256kbps (I use 320 myself).
Actually, and I know you won't get this, a good mp3 codec
should cut out audio > 16 kHz when using 160 kbps and lower. All codecs, including iTunes AAC use low pass filtering when using medium to low bitrates. There are several reasons for this:
- Given that few people hear much more than 16 kHz anyway, it's better to trade off the higher, less audible, frequencies in order to lessen the severity of the artifacts in the much more important sub 16 kHz.
- Mp3 has a design "flaw" that causes bitrates to become bloated if > 16 kHz audio is to be properly encoded --> see the previous point.
What bitrates? You see, the purpose of a lossy codec is to save space. LAME is transparent around ~ 190 kbps (for most people even lower). I don't know if Blade
ever becomes transparent, but even if it does become "OK" at 320 kbps that still wouldn't make Blade a good mp3 encoder.
ABX test with Blade as a low anchor:
http://www.rjamorim.com/test/128extension/comments/results01/phong-result01.txt
LAME vs Blade and others @ 256 kbps:
http://www.hydrogenaudio.org/forums/index.php?showtopic=1946&hl=
More can easily be found but I refuse to do your homework.
