I'm quite new to all this mp3 converting stuff so please be gentle

I've just started converting my cds to mp3 using chrs Mydens guide to mp3s.
However I've just been browsing on another forum and someone says that VBR is unstable and fools you into thinking that the mp3s are better quality by just increasing the treble. I don't have the best of ears, but I think that my mp3s do sound better then the ones that my music match software used to produce ( I don't use this any more). I'm a bit confused. Is there a "battle" between VBR and CBR like there is with intel and amd ect with people taking sides?

"someone" has no idea whatsoever what he is talking about.



No battle. VBR was proven to be superior more than 3 years ago (original Dibrom test at r3mix), and VBR algorithms have improved significantly since then.

There are also logical reasons for preferring VBR - the demands on the MP3 encoder vary with the style of music, and CBR cannot adjust to that.

Some of the more modern codecs are inherently VBR for the same reason.

Please, refer to this sticky or to the wiki.

As a rule of thumb for quality: CBR<ABR<VBR at the same bitrate (except 320kbps).

VBR is very stable with LAME. Ask to the person that said that VBR quality is inferior to provide proof through a double blind test. I bet he or she will be silent.

"increasing the treble"

more like you can hear the same treble as it is on the CD with VBR than 192 CBR...

More like a battle between VBR and idiots. Just ignore whatever that blowhard was making up and stick to the recommended settings.

Thanks for the reply guys.

If there was a battle going on between CBR and VBR, wouln't it seem weird to you that the the same encoder can produce VBR files as well as CBR files?

Not at all, because there is no doubting room for "preference". As far as quality/size goes, VBR is provably superior, and is a fact.

Unstability? Well, sort of. There are some problems with some portable MP3 devices playing VBR. But that's the portable not being good enough, not a matter of VBR itself. You won't see such a problem playing VBR mp3's on the computer - VBR is standard.

Increasing treble the only reason of quality? Absolute and utter ignorance.

Time to start working on the new magic encoder that boosts treble!

It's called an EQ

VBR is the way to go if you are looking for quality vs size. However, be aware that some iPods seem to have a problem with LAME VBR files. The problem is with the iPod and not with LAME

keep in mind, these ipod/lame problems are relatively isolated and rare. afaik, the stuttering only occurs on certain songs, at certain points. imho, this is no reason to shy away from vbr. if it is causing you problems on your ipod, i would try modifying your commandline settings before you ditch lame vbr all together. there is a thread floating around here somewhere that discusses this...

Thanks for the replies guys,
I fell for the many post = must know what they are talking about. Things that he post is truely laughable.

Just for fun, what forum was that?

The vast majority of my MP3s are 192 kbps CBR. Then someone opened my mind (and ears!) to VBR, so I'm very much interested (and just getting started) w/ VBR. I am using LAME 3.96.1 and VBR'd a WAV file to MP3. But the time length displayed on Winamp 5.11 is incorrect. Someone in another thread noticed the same "phenomenom". The bug is either with VBR or Winamp. Since both have been around for many years, who knows.

I don't think it is a LAME VBR issue in case the files are correctly encoded. I tried to reproduce this with Winamp 5.11. I used the default Nullsoft v.3.31 decoder, Thompson mp3PRO decoder and Otachan's version of the mpg123 decoder. They all showed the same correct track lengths. It seems that Winamp reads the VBR header info correctly with all decoders. I have no idea how Winamp works, but some other programs read a small amount of frames in the beginning of the files and estimate the bitrate and track length if a compatible VBR header is not available.

I don't know if this helps anything, but you could try the Otachan's mpg123 decoder: http://www.hydrogenaudio.org/forums/index....18530&hl=mpg123.

The problem is very likely caused by a missing Xing VBR Tag in the mp3 file. So, it's the encodings fault (IIRC it's somehow possible to add such a tag later with some tool). Check your lame settings, and, if you're using some weird GUI, try encoding with LamedropXPd or just commandline lame.exe encoder (both found on rarewares.org, best use 3.97b1). This should solve your problems.
Winamp does properly detect VBR MP3's length per default if a Xing tag is inside. (Unlike old sh*** WMP versions)

http://www.foobar2000.org

I used LAME 3.96.1 and tried various "--preset" options for VBR and even ABR. Both displayed incorrect times. I have a 72:43 mix and while both showed "72:43" on Winamp, when I actually played the files, by the time the last track finished, the finish times were around 72:46 or 72:47. This is really bad for me b/c I'm an e-DJ and having the proper track time is a vital requirement, especially for the cue files that I make.

So as much as what I'm about to say will be painful to many eyes and ears on the board, I've decided to stay with CBR 192 for my mixes. For individual tracks, where the displayed track time is not as important, I'll probably go with "--preset extreme", which is a VBR around the 220-250 kbps area.

Maybe the official release of LAME 3.97, whenever that comes out, will fix this. I assume the beta version of LAME 3.97 has the same bug. Yes, I'm blaming this on the VBR heading encoding of LAME rather than Winamp.

I used the command line interface of LAME 3.96.1 and am using a variety of presets, which I'm learning about via "lame --preset help".

Honestly if Winamp has such a bug now (which is easily fixed by either switching decoders or reverting back to an older one as in the past no such bug existed) then it's not LAME's fault it's Winamp fault.

Also --preset insane = -b320 in 3.96.1 = 320kbps CBR. --preset extreme = -V0 in 3.96.1 = 250-260kbps or so VBR is what you're thinking.

Also as a user of 3.97b and decoding VBR files generated by it with both foobar2000 and a portable device I don't believe it has any header generating issues. I could be wrong but I've not noticed any whatsoever.

If I had an old version of Winamp (say, from the 3.x series), then I can further investigate this VBR issue. Even simpler, I should try playing these VBR files (that show the wrong time on Winamp) on Windows Media Player. I'll try that sometime this weekend.

You're right, I meant to say "--present extreme". That is what I meant, but I got my adjectives mixed up. Incidentally, I took a ~5 min. track and encoded it as "--preset insane", "--preset extreme", "--preset 192", and "--preset cbr 192". Interestingly enough, I could hear the difference between the original WAV file and the MP3 files. But I had a hard time discerning between the MP3s. I have very good computer speakers (Logitech Z-5500 via digital fiber optic Toslink to my sound card).

Even better, try a non retarded player, like foobar2000.


Do you use the EQ in Winamp? Because that one is crap and is the reason for the difference, if in fact there is one and it's not just placebo.

I never heard of Foobar2000 before I signed up here w/ HA. I will check it out. I've used Winamp for several years (since the 2.x days), so I'm quite familiar with it; hence my hesitation to discard it.


Yes, I am using the EQ on Winamp. But the EQ also applies to WAV files and not just MP3s. Maybe the EQ is applied differently on a WAV than an MP3. I will re-attempt this (non-blind) listening test and try to determine if I hear a difference between the WAV and the various encoded MP3s.

Read this thread:
http://www.hydrogenaudio.org/forums/index....topic=37954&hl=

Ahh, thank you for the link! That explains a lot. So the Cliffs Note answer to my question is that Winamp uses different EQ'ing for WAV and MP3. You're right that when I disabled the EQ, I couldn't hear a difference, from the very brief testing I did. I'll continue to look into this some more when I get more time.