#64539 - 29/01/2002 19:58
VBR question(s)
|
member
Registered: 15/01/2002
Posts: 122
|
Ok, so I read the FAQ and learned a little something about VBR, which I knew nothing about before. Now I may be dense, so I just want to clarify. Is this a true statement: VBR ensures better audio quality than CBR and at reduced size. Is that right or did I not fully comprehend that section? It sounds to me like anytime I am doing my own ripping, encoding, etc, I should do so in VBR.
Finally, I'm using to encoding my files at 192 or 224 kbps. What percentage setting on VBR will give me a comparable audio quality?
Thanks for any/all help you can provide.
|
Top
|
|
|
|
#64540 - 29/01/2002 20:07
Re: VBR question(s)
[Re: nikko]
|
carpal tunnel
Registered: 12/11/2001
Posts: 7738
Loc: Toronto, CANADA
|
If you're using a good MP3 encoder, then VBR is not specified as a percentage. (example: LAME)
Next, VBR will not be better than a very high bitrate CBR (256/320Kbit).
That said, VBR can be very good. Take a look at www.r3mix.net. There are other places to read. Check out this thread for a small discussion and pointers on what to search the net for (to get info on an updated LAME preset).
Bruno
|
Top
|
|
|
|
#64541 - 29/01/2002 20:07
Re: VBR question(s)
[Re: nikko]
|
journeyman
Registered: 04/11/2001
Posts: 59
Loc: Texas
|
Yes, VBR will give you better overall quality while lowering the file size. It allows for the most complex parts of the music to use up to 320kbps and the silent or simple parts to use as little as is needed to insure accurate reproduction.
Not all VBR encoders are the same (or CBR for that matter).
I use LAME 3.91 with the "--r3mix" switch (which is a macro for a popular MP3 website's recommended settings for CD quality).
My MP3's enocded with this have an average size usually between 165 and 190 kbps, and I am totally happy with the results. Prior to this I used Music Match at 192 kbps and still could hear encoding artifacts.
_________________________
[green]-Matt Pritchard[/green]
Riocar 60gig S/N : 010102081
|
Top
|
|
|
|
#64542 - 29/01/2002 20:09
Re: VBR question(s)
[Re: nikko]
|
carpal tunnel
Registered: 20/12/1999
Posts: 31600
Loc: Seattle, WA
|
Your statement is essentially correct, but over-simplified.
VBR gives you "more bang for your buck" in terms of file size. In other words, if you had two 5-megabyte files of typical music, one of them a constant-bit-rate file and another one a variable-bit-rate file, the VBR file will generally have better quality. Because it can crank up the bitrate for complex (high-frequency) passages, and reduce the bit rate for less complex passages.
I don't know about "percentage" settings because my VBR encoder software doesn't use a "percentage". With VBR, the resulting bit rate and file size are kind of a crap-shoot, you never know exactly what you're going to get. You just choose a setting and try it out for a while and see what kind of file sizes you're getting.
|
Top
|
|
|
|
#64543 - 29/01/2002 20:21
Re: VBR question(s)
[Re: tfabris]
|
carpal tunnel
Registered: 25/12/2000
Posts: 16706
Loc: Raleigh, NC US
|
Some encoders also offer ABR (Average Bit Rate), where the results are similar to VBR, but you do know (almost) what size file you'll get beforehand, while sacrificing some quality from VBR.
_________________________
Bitt Faulk
|
Top
|
|
|
|
#64544 - 29/01/2002 20:35
Re: VBR question(s)
[Re: nikko]
|
journeyman
Registered: 22/07/1999
Posts: 60
Loc: St. Paul, MN, USA
|
Ignore some of the subtle elitism inherent to this conversation -- the encoder you're using is probably more than adequate for excellent sounding mp3s. It's MusicMatch or AudioCatalyst, right? Anyhow, you'll find that 100% VBR gives you files of approximately the same size as files created at 192 kps. A single album will be about 100 MB. This will vary from album to album, of course, but this is a generalization that seems to hold in real life.
I find the size to be totally acceptable and the sound quality meets CD quality to my ears.
_________________________
your fiend,
mafisto
|
Top
|
|
|
|
#64545 - 29/01/2002 20:36
Re: VBR question(s)
[Re: The_Optimizer]
|
carpal tunnel
Registered: 12/11/2001
Posts: 7738
Loc: Toronto, CANADA
|
That's the important part. At a lower file size, VBR will give you an overall better sounding file. At a high bitrate, VBR cannot be any better than CBR (obviously if a VBR file is maxing out at your CBR rate, it can't possibly be any better). This is basic info right in the LAME documentation.
Bruno
|
Top
|
|
|
|
#64546 - 29/01/2002 20:38
Re: VBR question(s)
[Re: mafisto]
|
carpal tunnel
Registered: 25/12/2000
Posts: 16706
Loc: Raleigh, NC US
|
It's not really elitism. It's just that a percentage means absolutely nothing. It's their own proprietary scale, and we just have to guess as to what that means, since it's not apparently documented anywhere.
_________________________
Bitt Faulk
|
Top
|
|
|
|
#64547 - 29/01/2002 20:45
Re: VBR question(s)
[Re: wfaulk]
|
member
Registered: 01/01/2002
Posts: 144
|
Well, you're half right
It's not elitism, but for different reasons.
"Ignore some of the subtle elitism inherent to this conversation -- the encoder you're using is probably more than adequate for excellent sounding mp3s."
But why not suggest to him the best FREE (as opposed to the others you mentioned which may require registration, I haven't used them in years :P ) encoder, which just happens to have better sound than the pay encoders?
I find nothing elitist about that. EAC+LAME, or W2L+LAME is a great combination. I used to use the Fraunhofer codec with Sound Forge to encode, and this really does seem to sound better. Of course, that' s my elitist opinion, so i'm biased
|
Top
|
|
|
|
#64548 - 29/01/2002 20:47
Re: VBR question(s)
[Re: wfaulk]
|
carpal tunnel
Registered: 12/11/2001
Posts: 7738
Loc: Toronto, CANADA
|
That and if you look at graphs comparing XING to LAME or Fraunhoffer at high bitrates you'll see how bad XING is.
XING is FAST. Doesn't mean it's any good though. And I'm pretty certain that LAME VBR is more sophisticated than XING's by a greater factor than LAME's CBR is better than XING's.
Bruno
|
Top
|
|
|
|
#64549 - 29/01/2002 20:54
Re: VBR question(s)
[Re: wfaulk]
|
journeyman
Registered: 22/07/1999
Posts: 60
Loc: St. Paul, MN, USA
|
*sigh* I should have just kept my keyboard shut on that one...
I've been reading this board for two and a half years. I understand the arguments. Some people use mainstream rippers with nebulous googly buttons and dials with no documentation -- and don't want them. I answered his question, to the best of my ability, regarding "percentages".
Okay, now there's six responses and counting. Jesus, now I remember why I just read the email summaries in the morning.
Fine, ask a question about how to operate your software, and get a recommendation for another set of software. I've learned about all sorts of nice things that way, and cannot begrudge the common voice its best practice. I hope Nikko gets some useful information from this conversation at the very least.
_________________________
your fiend,
mafisto
|
Top
|
|
|
|
#64550 - 29/01/2002 20:58
Re: VBR question(s)
[Re: mafisto]
|
carpal tunnel
Registered: 12/11/2001
Posts: 7738
Loc: Toronto, CANADA
|
I answered his question simply without going into specific application software. If you like AudioCatalyst, then you'd probably like AudioGrabber, which AC is based on. As easy to use and will allow you to use LAME or any number of other codecs.
Bruno
|
Top
|
|
|
|
#64551 - 29/01/2002 21:01
Re: VBR question(s)
[Re: tfabris]
|
old hand
Registered: 12/01/2000
Posts: 1079
Loc: Dallas, TX
|
Fraunhofer VBR ( which uses the percentage) is actually more like ABR, or at least thats the way it worked in the past. Not sure if that has changed.
|
Top
|
|
|
|
#64552 - 29/01/2002 21:13
Re: VBR question(s)
[Re: hybrid8]
|
journeyman
Registered: 22/07/1999
Posts: 60
Loc: St. Paul, MN, USA
|
Okay, now I'm REALLY regretting that offhand comment.
Everyone is officially obsolved of whatever sin you think that I think you may have committed. My comment was directed to Tony's vaguely patronizing tone. Which wasn't bad by the measure of similar conversations.
I love you all, and hope you can forgive our collective trespasses. I DON'T wish you would burn in hell.
_________________________
your fiend,
mafisto
|
Top
|
|
|
|
#64553 - 29/01/2002 21:42
Re: VBR question(s)
[Re: mafisto]
|
new poster
Registered: 28/01/2002
Posts: 5
Loc: Sent back to Hell
|
I DON'T wish you would burn in hell
I don't see why not. It sure is nice down here, and I'm sure there's room somewhere...
The Angel of the Abyss
|
Top
|
|
|
|
#64554 - 30/01/2002 11:21
Re: VBR question(s)
[Re: hybrid8]
|
addict
Registered: 24/08/1999
Posts: 564
Loc: TX
|
aren't Xing the guys who started the whole DeCCS thing by writing such crappy code that the security keys where visible?
_________________________
==========================
the chewtoy for the dog of Life
|
Top
|
|
|
|
#64555 - 30/01/2002 16:03
Re: VBR question(s)
[Re: ashmoore]
|
member
Registered: 22/09/2000
Posts: 195
Loc: Copenhagen, Denmark
|
Xing were the ones getting the ball rolling. Bad code or not, as i understood it they left in something by mistake, but not excatly sure what it was.
\\Kaare
|
Top
|
|
|
|
#64556 - 30/01/2002 16:13
Re: VBR question(s)
[Re: _hardcore_]
|
enthusiast
Registered: 28/01/2002
Posts: 265
Loc: MI, USA
|
it was bad code..
they left OUT something when they forgot to encrypt the decryption key
_________________________
guardian__J MKIIa 20g Smoke
|
Top
|
|
|
|
#64557 - 30/01/2002 22:20
Re: VBR question(s)
[Re: mafisto]
|
carpal tunnel
Registered: 20/12/1999
Posts: 31600
Loc: Seattle, WA
|
My comment was directed to Tony's vaguely patronizing tone.
Although I will admit to using a patronizing tone many times in the past, I didn't think I was doing so in that particular post. I genuinely meant that I know nothing about percentage settings. I have never used a piece of software which has that option, so I really and truly don't know what they mean.
|
Top
|
|
|
|
|
|