HydrogenAudio

Lossy Audio Compression => MP3 => MP3 - General => Topic started by: EcKo72DB on 2002-11-18 19:47:28

Title: Well now that Lame 3.93 has been released...
Post by: EcKo72DB on 2002-11-18 19:47:28
I keep hearing bad things about lame 3.93 and the --alt-presets , now that it's released should I upgrade from 3.92? I encode using --alt-preset cbr 192 . Will using lame 3.93 downgrade the quality at all?

Thx for the replies. 
Title: Well now that Lame 3.93 has been released...
Post by: JohnV on 2002-11-18 20:09:54
Well, I can't say for sure, but I don't think cbr should be any lower quality than before.
It's just that the quality has not been checked. I tried --alt-preset fast standard a little bit, and in some cases the quality was actually higher than before, although the bitrate is also clearly higher.
Anyway, it has not been troughly tested.

Dibrom has some interesting ideas for the future for mp3.
Title: Well now that Lame 3.93 has been released...
Post by: FinCoder on 2002-11-18 21:40:22
"Dibrom has some interesting ideas for the future for mp3."

Like what? 
Title: Well now that Lame 3.93 has been released...
Post by: Andavari on 2002-11-18 21:50:14
I see that --alt-preset medium is now an option. 3.93 encodes much faster than 3.92, I was used to .7x to .8x speed now it hums along at 1.1 to 1.2x.
Title: Well now that Lame 3.93 has been released...
Post by: JohnV on 2002-11-18 21:53:38
Quote
"Dibrom has some interesting ideas for the future for mp3."

Like what?  

Too early to tell.. Maybe he will tell soon, if there's a possibility for the plans to happen.
Title: Well now that Lame 3.93 has been released...
Post by: Dibrom on 2002-11-18 23:07:28
Quote
I see that --alt-preset medium is now an option. 3.93 encodes much faster than 3.92, I was used to .7x to .8x speed now it hums along at 1.1 to 1.2x.

Just for the record, --alt-preset medium is not something that I created (not to say that it's necessarily bad.. it is also at least based very largely on aps).  I haven't tested it on any of the normal test samples or anything that.  I believe this is Gabriels creation, and I don't know anything about how he does his testing (ie.  What samples he uses, does he do double blind tests, does he get feedback from other people, etc.).

Just want to say that, because I think there's going to be some confusion from people about this.
Title: Well now that Lame 3.93 has been released...
Post by: frozenspeed on 2002-11-18 23:22:17
What kinds of changes can we expect between 3.93 & 3.94?  With all the hub-ub going on over these two releases it isn't quite clear what's going into 3.94 that isn't in 3.93.

-Jeff
Title: Well now that Lame 3.93 has been released...
Post by: Dibrom on 2002-11-19 00:22:13
Quote
What kinds of changes can we expect between 3.93 & 3.94?  With all the hub-ub going on over these two releases it isn't quite clear what's going into 3.94 that isn't in 3.93.

-Jeff

I have no idea.
Title: Well now that Lame 3.93 has been released...
Post by: JohnV on 2002-11-19 02:09:28
Quote
What kinds of changes can we expect between 3.93 & 3.94?  With all the hub-ub going on over these two releases it isn't quite clear what's going into 3.94 that isn't in 3.93.

-Jeff

Well.. you can get somekind of idea if you check this and see what was done for Tak's branch and what was done for the main branch:
http://cvs.sourceforge.net/cgi-bin/viewcvs...l.c?sortby=date (http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/lame/lame/libmp3lame/psymodel.c?sortby=date)
http://cvs.sourceforge.net/cgi-bin/viewcvs...e.c?sortby=date (http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/lame/lame/libmp3lame/lame.c?sortby=date)
http://cvs.sourceforge.net/cgi-bin/viewcvs...e.c?sortby=date (http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/lame/lame/libmp3lame/vbrquantize.c?sortby=date)

Of course not everything is necessarely going for 3.94..