Jump to content

Impressions of LAME Mp3 Codec

Rate this topic


Dinko

Recommended Posts

Ok, so I kept reading how LAME was the best to use to rip CDs to mp3, as much here as elsewhere.

I rarely used DBPowerAMP but I figured I'd try it. I upgraded to the latest version, and ripped a bunch of files using RealPlayer (FHG) and DBPowerAMP (Lame) using the same basic settings.

End result: a *very* audible advantage to FHG.

For illustration purposes, let's take one of the tracks I used: the second movement from Balakirev's First Symphony.

Difference #1: the oboe in the LAME recording has a digital echo and background artifacts.

Difference #2: the drums in the LAME recording produce clear distorsions and artifacts. These are loud and very obvious. To the point that the artifacts become louder than the drums themselves!

Difference #3: with LAME, the strings become acid, with distorsions, as if they were being torn to shreds.

Difference #4: the soundstage in LAME is reduced. Reverb is lacking. Instrument separation is lower.

I'm really not sure why everyone keeps touting LAME as the best thing since the wheel, but my personal experience just now indicates that it's a pretty poor performer for orchestral music. I tried the files on the PC, on a Samsung DAP and an iPod. The iPod showed the least difference between LAME and FHG. I haven't tried gapless, but I really couldn't care less. Given the audible distorsions I just experienced in LAME, I'll put up with gaps if needed to avoid the artifacts.

Link to comment
Share on other sites

1. Use the latest recommended version of lame: http://wiki.hydrogenaudio.org/index.php?title=Lame_Compiles

2. Use VBR. Recommended setting is -V2 --vbr-new. It's well known, even at Hydrogenaudio, that with CBR, FhG seems to be better. But at VBR, Lame is better.

3. Did you do ABX? Don't mean to be mean, but without ABX, your comments about "very audible" artifacts don't mean much.

Link to comment
Share on other sites

Thanks for the additional feedback. I'll try that in the future.

Truth be told, I wanted CBR for two reasons: faster ripping & more predictable file sizes. I rarely use VBR. But I might start.

As for ABX... it's a very good concept, which is statistically meaningful, but in the end, there were artifacts and distorsions in the CBR Lame encoded tracks that simply do not appear in the FHG versions. Twist and turn and randomly try as I might in blind listening tests, it's not going to change anything. The artifacts are so audible, that I'm pretty sure anyone could get 90% accuracy or more in guessing which file was Lame and which was FHG. This really wasn't like guessing between AAC, ATRAC or WMA files at 192kbps. The difference in sound was more like comparing 96kbps mp3s with 192kbps ATRACs, if you see what I mean.

In any case, I'll try some VBR rips. The results that I got were so weird compared to the general consensus on LAME, that I'm wondering if there was a momentary computer glitch that might have caused such a difference in quality.

BTW, you have an iPod, right? I suppose you mostly use VBR mp3s, but do you have any idea how (if at all) the iPod battery life would be affected by the use of VBR relative to CBR?

Link to comment
Share on other sites

Not sure if this is entirely relevant, but I tried to rip some some copied CDs using Sonicstage and it couldn't recognise the discs. I tried using CDex to get a LAME rip - next best thing to ATRAC ;) - and the discs did in fact read and rip, but with so many artifacts as to make them unlistenable. I don't think it was LAME that was at fault though - stuff from eMusic sounds great - but with the discs themselves. Perhaps you might have a similar problem, i.e. scratches etc? Not sure at all, but I thought I'd mention it just in case.

Nav

Link to comment
Share on other sites

BTW, you have an iPod, right? I suppose you mostly use VBR mp3s, but do you have any idea how (if at all) the iPod battery life would be affected by the use of VBR relative to CBR?

Decoding MP3s use so litle processing power that I doubt there will be a significant difference. I have 5G iPod 60GB, and it has a 64MB cache, so what matters is file-size, not CBR vs VBR. Considering VBR is more efficient than CBR, giving better quality/smaller file size, in the end I would think VBR is better. Battery life is more affected by having the LCD on and/or skipping tracks frequently.

Not sure if this is entirely relevant, but I tried to rip some some copied CDs using Sonicstage and it couldn't recognise the discs. I tried using CDex to get a LAME rip - next best thing to ATRAC ;) - and the discs did in fact read and rip, but with so many artifacts as to make them unlistenable. I don't think it was LAME that was at fault though - stuff from eMusic sounds great - but with the discs themselves. Perhaps you might have a similar problem, i.e. scratches etc? Not sure at all, but I thought I'd mention it just in case.

First, clean your CDs. 2nd, use EAC in secure mode to rip the tracks. Good luck.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
×
×
  • Create New...