320kbps VBR vs 320kbps CBR MP3's?

Dec 23, 2009 at 8:20 PM Post #16 of 34
Quote:

Originally Posted by userlander /img/forum/go_quote.gif
Actually VBR is shown to give higher sound quality, and at lower file size. Someone posted a really good link recently that I'm trying to find, with graphs comparing all kinds of different encodings. From doing my own listening tests, I had found that I was consistently picking the high bitrate VBR over the CBR, and when I read that article it really sealed it for me. Don't always assume that more bits necessarily = better quality.
smily_headphones1.gif



For bitrates lower than 320kbps that makes sense, ex. 256kbps. Since then CBR will be fixed at 256kbps, while VBR may go all the way up to 320kbps if needed.
But for 320kbps it can not go any higher, just lower. Hence CBR always have the maximum bits, while VBR may not.

But, what do I know...
wink.gif
 
Dec 23, 2009 at 8:35 PM Post #17 of 34
Quote:

Originally Posted by krmathis /img/forum/go_quote.gif
For bitrates lower than 320kbps that makes sense, ex. 256kbps. Since then CBR will be fixed at 256kbps, while VBR may go all the way up to 320kbps if needed.
But for 320kbps it can not go any higher, just lower. Hence CBR always have the maximum bits, while VBR may not.

But, what do I know...
wink.gif



No, that's all true, of course. I am merely questioning whether that automatically means that 320CBR is going to sound better. It's not better compression, at least, if by better we mean the best SQ to size ratio.
wink.gif
 
Dec 23, 2009 at 8:54 PM Post #18 of 34
Quote:

Originally Posted by userlander /img/forum/go_quote.gif
I actually think VBR sounds better than CBR. I'll even take a 256 VBR over a 320 CBR.


Same here - but because VBR generally requires a good LAME encoder, instead of some crappy old encoder that could have been used to do 320 CBR. Since it can't go above 320 anyway, the encoder matters.
 
Dec 23, 2009 at 9:10 PM Post #20 of 34
Quote:

Originally Posted by Pageygeeza /img/forum/go_quote.gif
I've noticed on Exact Audio Copy that Lame defaults to VBR Joint Stereo. I know It's saving a few bytes, but is it seriously affecting audio quality?


Quote:

(j)oint stereo
In this mode, the encoder will make use of a correlation between both channels.
The signal will be matrixed into a sum ("mid"), computed by L+R, and difference
("side") signal, computed by L-R, and more bits are allocated to the mid chan‐
nel. This will effectively increase the bandwidth if the signal does not have
too much stereo separation, thus giving a significant gain in encoding quality.



...
 
Dec 23, 2009 at 10:26 PM Post #21 of 34
From the Lame documentation:

Quote:

JOINT STEREO is the default mode for stereo files with fixed bitrates of 128 kbps or less. At higher fixed bitrates, the default is stereo.
For VBR encoding, jstereo is the default for VBR_q >4, and stereo
is the default for VBR_q <=4. You can override all of these defaults
by specifing the mode on the command line.

jstereo means the encoder can use (on a frame by frame bases) either
regular stereo (just encode left and right channels independently)
or mid/side stereo. In mid/side stereo, the mid (L+R) and side (L-R)
channels are encoded, and more bits are allocated to the mid channel
than the side channel. This will effectively increase the bandwidth
if the signal does not have too much stereo separation.

Mid/side stereo is basically a trick to increase bandwidth. At 128 kbps,
it is clearly worth while. At higher bitrates it is less useful.

FORCED MID/SIDE STEREO forces all frames to be encoded mid/side stereo. It
should only be used if you are sure every frame of the input file
has very little stereo seperation.


Long story short -- you shouldn't need to specify the stereo mode on the command-line; Lame will take care of that for you to give you the best average combination of quality and file size.
 
Dec 24, 2009 at 7:05 AM Post #22 of 34
So, I downloaded RazorLame and it seems rather simple, however when I try to encode, it says it cannot find "Lame.EXE", I've downloaded the file "lame-398-2.tar.gz" from Browse LAME (Lame Aint an MP3 Encoder) Files on SourceForge.net

However, I could not find an EXE file at all amongst the files included on the folder. Am I missing something or did I downloaded the wrong file?

I apologize for the dumb questions, I've only recently discovered the joys of listening to high-quality music as a general means, therefore I don't know too much so far.

Thanks!
 
Dec 24, 2009 at 7:12 AM Post #23 of 34
I would try using EAC for encoding. Here is an excelent tutorial in setting it up. Exact Audio Copy Guide Once you have that set you are good to go.
 
Dec 24, 2009 at 9:17 AM Post #24 of 34
I have to agree with the view that LAME VBR sounds better than 320kbps CBR. I can't explain that with any tech talk, it's just my subjective feeling.

I had a collection of 100% CBR 320kbps before I found LAME. Now I'm re-ripping my CDs into LAME and find it sounds better. So much so that I cannot tell the difference between MP3s and WAV/Flac files now, whereas with the 320kbps CBR I felt I could detect a slight difference. Also when I'm done I'm going to be able to fit a LOT more music on my dap than before
smily_headphones1.gif


Placebo? Possibly, but that's how I feel about it.
 
Dec 24, 2009 at 10:12 AM Post #25 of 34
I can tell the difference between 320 mp3 and FLAC on my setup, but I really have to strain to do so. If I'm just enjoying the music, I can't tell the difference at all.
 
Dec 24, 2009 at 12:25 PM Post #26 of 34
Quote:

Originally Posted by EddieE /img/forum/go_quote.gif
I have to agree with the view that LAME VBR sounds better than 320kbps CBR. I can't explain that with any tech talk, it's just my subjective feeling.

I had a collection of 100% CBR 320kbps before I found LAME. Now I'm re-ripping my CDs into LAME and find it sounds better.



Could it be that you were comparing LAME with non-LAME?
.
 
Dec 24, 2009 at 12:36 PM Post #27 of 34
Quote:

Originally Posted by Raikoh /img/forum/go_quote.gif
So, I downloaded RazorLame and it seems rather simple, however when I try to encode, it says it cannot find "Lame.EXE", I've downloaded the file "lame-398-2.tar.gz" from Browse LAME (Lame Aint an MP3 Encoder) Files on SourceForge.net

However, I could not find an EXE file at all amongst the files included on the folder. Am I missing something or did I downloaded the wrong file?



Yes, you've downloaded the wrong file. Here's the .exe you need. (I would take version 3.98.2 on top.)
.
 
Dec 24, 2009 at 2:29 PM Post #28 of 34
Quote:

Originally Posted by JaZZ /img/forum/go_quote.gif
Could it be that you were comparing LAME with non-LAME?
.



Not sure how you mean. Before I was listening to wmp ripped 320kbps and sometimes on some tracks I felt I could tell a difference between lossless or a cd and it.

Now I'm using Eac ripped lame vbr and i've done a few tests and am yet to spot a difference in any way. I guess eac could have as much to do with it as lame...
 
Dec 24, 2009 at 7:48 PM Post #29 of 34
Quote:

Originally Posted by Raikoh /img/forum/go_quote.gif
So, I downloaded RazorLame and it seems rather simple, however when I try to encode, it says it cannot find "Lame.EXE", I've downloaded the file "lame-398-2.tar.gz" from Browse LAME (Lame Aint an MP3 Encoder) Files on SourceForge.net

However, I could not find an EXE file at all amongst the files included on the folder. Am I missing something or did I downloaded the wrong file?



What you've downloaded is the LAME source code. If the code is compiled, you will get an executable.

JaZZ gave a URL that will work, but pretty much the canonical place to get encoder binaries is Rarewares:
RareWares
The guy who runs Rarewares knows how to compile encoders to run their best.

Quote:

I have to agree with the view that LAME VBR sounds better than 320kbps CBR. I can't explain that with any tech talk, it's just my subjective feeling.


I agree with JaZZ on this one; if you're comparing non-Lame 320CBR against Lame VBR, it's no contest -- Lame will almost always win.
smily_headphones1.gif
However, using the same algorithm, Lame's 320CBR will beat Lame's VBR, because Lame cannot use more than 320kbps for any packet in VBR mode, so there's inherently less information. It's impossible to get a better reproduction with the same algorithm with fewer bits.
 
Dec 24, 2009 at 8:11 PM Post #30 of 34
UNHchabo,
Ah, I get what JaZZ was saying now, thanks for the clarification.

I'll have a go ripping 320 LAME CBR and LAME VBR with EAC and do some listening tests.

EDIT: But could I forward speculatively, that LAME VBR could of course decide to encode every part of the file at 320kbps - 320kbps is available there for it to use. The reason that it decides to encode so much of the file at less than 320kbps, ending in a smaller file size, is that it deems it can meet the criteria of transparency without it?

If that is the case, why would there be any audible difference?

Anyway, just an idea, I will do a test with the ears rather than logic soon.
wink.gif
 

Users who are viewing this thread

Back
Top