I'm quite new on these forums. Something like a year ago I started to encode my own MP3's but didn't figure out to contact any forum to learn the best settings.
I'm gonna tell you exactly how i do my MP3's and, please, comment without any respect for my newbie naivity... Any advice?
Ripper
CDex - currently 1.51
method: cdparanoia - full (WAY better handling of errors and "copy protection" than standard CDex...)
Default Normalization, no clipping.
Encoder
CDex-LAME
Joint Stereo
VBR: 96-224 kbps
VBR quality: 3
Encoding Quality: 2
I'm using >>that<< heavy VBR becouse I use my files on my MP3-cd and would like as many files as possible to fit... Usually I don't see the bitrate go down to 96 kbps in Winamp and if it does so just for max 1-2 seconds per song.
On songs that I know are recorded in mono (elvis presley - heartbreak hotel etc.) or I have checked with a vu-meter I use.
Everything same exept
Mono
VBR 60-112
Soooo, feel free to call me names as long as you explain why, as technically as possible. How bad does my files sound?
Full Version: How Stupid Am I
Welcome to the board!
The "don't worry about hurting a newbie response" is simple: why don't you go and read the FAQ!?!
But here's a more helpful response, with things to think about...
It depends on exactly what lame version and command line options you're using as to exactly what results you are getting. So no one knows what your files sound like except you at the moment!
There are two different psychoacoustic models in lame. Both have their faults. You're probably using gpsycho (the older default one) and it struggles with particular sounds. The --alt-presets use the newer nspsytune, and it's generally better. More importantly, the --alt-presets tune the internal workings of lame to try and remove the (different, smaller) problems with nspsytune. That's why they're usually recommended above any setting you choose to type in yourself.
If you've read the FAQ (which says, in essence, use lame 3.90.3 with --alt-preset standard for excellent results), you may decide that's what you want.
If you want a lower bitrate, well, it's always debatable how best to do this, but the suggestions in the FAQ are probably a good place to start - maybe better than your command line.
But "How bad does my files sound?" - I don't know - you encoded them, and you're listening to them. How does the bitrate compare with commands suggested in the FAQ? How does the sound compare on the most problematic samples (see the FAQ for archives of these).
You'll probably find that the suggested settings deliver the same sound at a lower bitrate, or a better sound at the same bitrate (depending which setting you use). Or you might find that it all sounds the same to you, and you really don't care.
btw, someone will probably come and tell you to use EAC instead of CDex.
Have fun reading the FAQ!
Cheers,
David.
The "don't worry about hurting a newbie response" is simple: why don't you go and read the FAQ!?!
But here's a more helpful response, with things to think about...
It depends on exactly what lame version and command line options you're using as to exactly what results you are getting. So no one knows what your files sound like except you at the moment!
There are two different psychoacoustic models in lame. Both have their faults. You're probably using gpsycho (the older default one) and it struggles with particular sounds. The --alt-presets use the newer nspsytune, and it's generally better. More importantly, the --alt-presets tune the internal workings of lame to try and remove the (different, smaller) problems with nspsytune. That's why they're usually recommended above any setting you choose to type in yourself.
If you've read the FAQ (which says, in essence, use lame 3.90.3 with --alt-preset standard for excellent results), you may decide that's what you want.
If you want a lower bitrate, well, it's always debatable how best to do this, but the suggestions in the FAQ are probably a good place to start - maybe better than your command line.
But "How bad does my files sound?" - I don't know - you encoded them, and you're listening to them. How does the bitrate compare with commands suggested in the FAQ? How does the sound compare on the most problematic samples (see the FAQ for archives of these).
You'll probably find that the suggested settings deliver the same sound at a lower bitrate, or a better sound at the same bitrate (depending which setting you use). Or you might find that it all sounds the same to you, and you really don't care.
btw, someone will probably come and tell you to use EAC instead of CDex.
Have fun reading the FAQ!
Cheers,
David.
David-
THink you hit on just about all th important subjects...
THink you hit on just about all th important subjects...
actually iam surprised there is no usual 'normalization is evil' feedback 
, anyway use replaygain after the ripping-encoding process instead of normalization.
, anyway use replaygain after the ripping-encoding process instead of normalization.
This is a "lo-fi" version of our main content. To view the full version with more information, formatting and images, please click here.