I encode with preset extreme but I wonder if my mp3 would have better quality with -V 0 -q0?
Thanks for the info that you can give me.

Theoretically: Maybe.

Practically: No, and it will take much longer to encode.

In Lame 3.97b2, I believe the -q switches below 2 have been remapped to -q 2, so I think it's useless. In any case, you can always add it on if time is not an issue.

'If time is not an issue'...?
So, there's actually a benefit of setting the -q switch below 2?

There can be. that's the point.

Uh...right.
And what kind of benefit(s) 'there can be'...?

improved quality, eg, smaller size for same sonic reproduction (better huffman coding)

I think that's the potential better point.

With LAME 3.97b2, has the -q 0 problem been resolved?
I don't really follow LAME's development ('thanks to Vorbis', I guess...) and a short 'yes' or 'no' would suffice.
Thanks before. =)

Yes, It is said to be the case.

Ok, thanks for clearing it up.
Might as well doing some test encodings and see what kind of 'benefits' I could get out of -q 0.

When using --vbr-new, yes. Otherwise each -q value is different.

Lame really should have less switches.

Your car doesn't need a brake pedal either, I suppose?

LAME is great the way it is right now. Very good settings used by default, and plenty of switches available for people who are control freaks.

Does your car have switches that control the fuel injection? Or the alternator current? Probably not since a computer can control those better then you could. Most of the switches in Lame seem to exist mainly for control freaks to hang themselves with. But thats a topic for a different thread I guess.

In short, would -q 0 without --vbr-new make much more sense than pairing them up?
Any reasons why -q settings below 2 are remapped to -q 2 when using --vbr-new?

@Shade[ST}
Come to think of it, if -q settings below 2 are remapped, why bother 'taking some time' to try them out?

Take a look here.

Spot on.
Thanks for directing me to the right direction.

And it seems the reason for that is kinda unclear though.
Gabriel just said, "Intended.", but nothing on why it's intended to be like that w/ --vbr-new.
I'm guessing it's to 'preserve' --vbr-new's speed, but cmiiw.

it has nothing todo with "preserve speed", --vbr-new simply works different and uses different code. ie a q-level feature controling the temination of inner-loops has no impact, because the newer vbr code does not use that inner loop at all.

So, is the q level that's used in the new vbr mode controlling a different routine or is it just there so it won't produce error(s) when people using the -q switch?

Ah, and sorry if I ask too much, though...

Hello, first off I'd like to point out to all of you people here on HA what a great community you are and that I'm looking forward to become part of it.

Now - on with the topic:
Is it controlling a differnt routine in vbr new mode? Especially as -q from 0 to 4 are the same?
But then, why is -q0 taking so much longer if the outcome is the same?

ThanX