IPB

Welcome Guest ( Log In | Register )

 
Reply to this topicStart new topic
best codec for CBR 128 kbps encoding?, 3.97b2???
memomai
post May 26 2006, 16:29
Post #1





Group: Members
Posts: 264
Joined: 13-February 05
From: Germany, Kempten
Member No.: 19808



Hi folks, please don't ask me WHY 128 cbr, I just want to know which mp3 encoder is the recommended one for this work... since I heard a few bugs of 3.97b2 in low CBR encoding (like definetly audible artifacts) I doubt on this encoder... or is it nevertheless the best one to choose?

thx 4 answers smile.gif


--------------------
FB2K,APE&LAME
Go to the top of the page
+Quote Post
kornchild2002
post May 26 2006, 16:32
Post #2





Group: Members
Posts: 2043
Joined: 8-April 05
From: Cincinnati, OH
Member No.: 21277



If you want to choose from popular encoders then yes, Lame 3.97b2 would be your best bet when compared to the Fraunhoffer encoder (sorry for spelling) that is offered in the free version of Winamp, Windows Media Player, MusicMatch, and iTunes. I am not sure if listening tests have been conducted on the Helix mp3 encoder which is included with RealPlayer.

You should really, really, really go with -V 5 --vbr-new Lame 3.97b2 if you want a target bitrate of 128kbps. Then again, your hardware may not support VBR mp3's. In order to get a better understanding of your situtation, you may want to summarize why you want CBR encoding instead of VBR or even ABR.
Go to the top of the page
+Quote Post
memomai
post May 26 2006, 16:46
Post #3





Group: Members
Posts: 264
Joined: 13-February 05
From: Germany, Kempten
Member No.: 19808



QUOTE
You should really, really, really go with -V 5 --vbr-new Lame 3.97b2 if you want a target bitrate of 128kbps. Then again, your hardware may not support VBR mp3's. In order to get a better understanding of your situtation, you may want to summarize why you want CBR encoding instead of VBR or even ABR.


just interested in CBR. I know VBR and ABR would be more useful. Stay on my question please. thank you anyway for your answer smile.gif


--------------------
FB2K,APE&LAME
Go to the top of the page
+Quote Post
jmartis
post May 26 2006, 17:15
Post #4





Group: Members
Posts: 381
Joined: 9-April 06
From: Czech Republic
Member No.: 29311



[I will try to write my post so that I don't violate T.O.S. 8. It is just hard to answer this type of question without violating the T.O.S.]
My favourite encoder at 128kbps cbr is Helix encoder. Main reasons I think it is better than LAME at 128k cbr: Better pre-echo. Tested mainly on "scooter" sample. Lower low-frequency "sandpaper" distortion. Tested on "Birds" sample and others. Faster. Encodes a 5min. song in 5 seconds.
I think it sounds much like the FastEnc encoder, but it's free.

I mainly encode electronic music, so for other genres the encoder choose could be different.

Again, I'm really just recommending what I'm using and what I think is the best.

J.M.

This post has been edited by jmartis: May 26 2006, 17:18
Go to the top of the page
+Quote Post
memomai
post May 26 2006, 17:21
Post #5





Group: Members
Posts: 264
Joined: 13-February 05
From: Germany, Kempten
Member No.: 19808



QUOTE
[I will try to write my post so that I don't violate T.O.S. 8. It is just hard to answer this type of question without violating the T.O.S.]


yes that's right this is a question hard to answer without violating the T.O.S. I'm sorry for that.


--------------------
FB2K,APE&LAME
Go to the top of the page
+Quote Post
shadowking
post May 26 2006, 18:19
Post #6





Group: Members
Posts: 1523
Joined: 31-January 04
Member No.: 11664



QUOTE (memomai @ May 26 2006, 07:29) *
Hi folks, please don't ask me WHY 128 cbr, I just want to know which mp3 encoder is the recommended one for this work... since I heard a few bugs of 3.97b2 in low CBR encoding (like definetly audible artifacts) I doubt on this encoder... or is it nevertheless the best one to choose?

thx 4 answers smile.gif



Did you abx what you heard ?. Either way mp3 isn't great at 128k cbr. 3.97 is the most stable for that bitrate in my tests.

PS - I can't imagine anything not supporting ABR, --abr 128 should be considered. If quality isn't enough then up it to 144~160k


--------------------
Wavpack -b450x1
Go to the top of the page
+Quote Post
Remedial Sound
post May 26 2006, 19:05
Post #7





Group: Members
Posts: 501
Joined: 5-January 06
From: Dublin
Member No.: 26898



To the O.P. - jmartis raises a good point. You're not going to hear a huge difference in quality between mp3 encoders at 128 CBR, so perhaps a choice should be made on encoding speed. Don't know where I saw the encoding speed test results (thinking it was someone from the HA forums that conducted them) but the Helix encoder is PDQ (Gogo is "built for speed" too). If quality is of paramaount importance then you might want to try some ABX tests with all the available encoders (FhG, Lame, Helix, Gogo). FYI Helix is available with mediacoder.

Just for the record it appears clear that the O.P. is aware that CBR is inferior to VBR and ABR. Let's remember that there are indeed valid applications for CBR, including streaming, audio within AVI video container (VBR is non-standard), and predictable file sizes (yes, ABR is predictable too tongue.gif ).
Go to the top of the page
+Quote Post
JunkieXL
post May 26 2006, 20:13
Post #8





Group: Members
Posts: 359
Joined: 3-April 05
Member No.: 21165



I've read several places that the AAC encoder is a good choice for making 128 kb/s CBR files.

I have no personal experience with AAC, but it would probably be worth your while to read up on the AAC forum as they did several listening tests at the 128 kb/s range.
JXL

Edit: Woops, missed the part where he wanted a mp3 encoder

This post has been edited by JunkieXL: May 26 2006, 20:14
Go to the top of the page
+Quote Post
Andavari
post May 26 2006, 21:54
Post #9





Group: Members
Posts: 935
Joined: 3-June 02
From: USA
Member No.: 2204



QUOTE (Remedial Sound @ May 26 2006, 13:05) *
You're not going to hear a huge difference in quality between mp3 encoders at 128 CBR, so perhaps a choice should be made on encoding speed.

Well I know for sure that I can't stand anything for more than a few seconds that was encoded at 128 k/bits CBR using Blade as the encoder, let alone any other bitrate it encodes at.

I can however put up with properly encoded recent LAME versions to LAME 3.97b2 128 k/bits mp3's (properly meaning someone, or some software like Audacity hasn't disabled Joint-Stereo). I can also put up with more recent (WMP10, etc) FhG Pro 128 k/bit mp3's.


--------------------
Complexity of incoherent design.
Go to the top of the page
+Quote Post
skelly831
post May 27 2006, 01:44
Post #10





Group: Members (Donating)
Posts: 782
Joined: 11-April 05
From: México
Member No.: 21361



What about FhG mp3senc?


--------------------
we was young an' full of beans
Go to the top of the page
+Quote Post
guruboolez
post May 27 2006, 08:35
Post #11





Group: Members (Donating)
Posts: 3474
Joined: 7-November 01
From: Strasbourg (France)
Member No.: 420



QUOTE (Remedial Sound @ May 26 2006, 19:05) *
To the O.P. - jmartis raises a good point. You're not going to hear a huge difference in quality between mp3 encoders at 128 CBR, so perhaps a choice should be made on encoding speed.

blink.gif

Take a look on the 128 kbps MP3 collective listening test:



http://www.rjamorim.com/test/mp3-128/results.html

There are strong differences in quality between different encoder - and as usual the difference is more obvious with low bitrate (like 128 kbps). It may be interesting to note that LAME made serious improvement with ABR/CBR since this test (3.97 introduced an adaptive ATH adjustement which increased quality on several samples - tests in the MP3-TECH forum are proving this).
Go to the top of the page
+Quote Post
jmartis
post May 27 2006, 09:29
Post #12





Group: Members
Posts: 381
Joined: 9-April 06
From: Czech Republic
Member No.: 29311



QUOTE (guruboolez @ May 27 2006, 09:35) *
Take a look on the 128 kbps MP3 collective listening test:

The "FhG" encoder, which one is it? (Fast/mp3enc, the new mp3sEnc (surround), build63i -family, other?)
looks like some bad one if it is beaten by Xing.

J.M.
Go to the top of the page
+Quote Post
guruboolez
post May 27 2006, 09:35
Post #13





Group: Members (Donating)
Posts: 3474
Joined: 7-November 01
From: Strasbourg (France)
Member No.: 420



http://www.rjamorim.com/test/mp3-128/presentation.html

Quality of Fhg "fastenc" encoders is overestimated IMO (and Xing "crapness" was also exagerated - as shown in the results).
Go to the top of the page
+Quote Post
jmartis
post May 27 2006, 09:40
Post #14





Group: Members
Posts: 381
Joined: 9-April 06
From: Czech Republic
Member No.: 29311



QUOTE (guruboolez @ May 27 2006, 10:35) *
Quality of Fhg "fastenc" encoders is overestimated IMO (and Xing "crapness" was also exagerated - as shown in the results).

If I'm not mistaken, the FastEnc VBR only encodes in plain Stereo. In my testing it seems to be really better in CBR. (just to mention, the "AudioActive" encoder is actually FhG 1.2 build 63i smile.gif )

J.M.

This post has been edited by jmartis: May 27 2006, 09:56
Go to the top of the page
+Quote Post
euphonic
post May 27 2006, 10:12
Post #15





Group: Members
Posts: 86
Joined: 12-February 06
Member No.: 27711



QUOTE (jmartis @ May 27 2006, 01:40) *
QUOTE (guruboolez @ May 27 2006, 10:35) *

Quality of Fhg "fastenc" encoders is overestimated IMO (and Xing "crapness" was also exagerated - as shown in the results).

If I'm not mistaken, the FastEnc VBR only encodes in plain Stereo. In my testing it seems to be really better in CBR. (just to mention, the "AudioActive" encoder is actually FhG 1.2 build 63i smile.gif )

J.M.


FastEnc's VBR also encodes in joint stereo, depending on the quality setting.
Go to the top of the page
+Quote Post
Remedial Sound
post May 27 2006, 18:58
Post #16





Group: Members
Posts: 501
Joined: 5-January 06
From: Dublin
Member No.: 26898



QUOTE (Andavari @ May 26 2006, 16:54) *
Well I know for sure that I can't stand anything for more than a few seconds that was encoded at 128 k/bits CBR using Blade as the encoder, let alone any other bitrate it encodes at.

Appologies. I assumed that old, crappy encoders aren't under consideration. biggrin.gif

QUOTE (guruboolez @ May 27 2006, 03:35) *
blink.gif

Take a look on the 128 kbps MP3 collective listening test:

http://www.rjamorim.com/test/mp3-128/results.html

There are strong differences in quality between different encoder - and as usual the difference is more obvious with low bitrate (like 128 kbps). It may be interesting to note that LAME made serious improvement with ABR/CBR since this test (3.97 introduced an adaptive ATH adjustement which increased quality on several samples - tests in the MP3-TECH forum are proving this).


Sorry, I should have been clearer here. My point was that by going with mp3 at 128kbps CBR (instead of a different codec / higher bitrate / VBR) that most of the decision on quality is already made right there. Yes, formal listening tests (the one you reference is for VBR btw) have revealed the differences in quality between encoders, however I think most would agree that current mp3 encoders at this bitrate are going to sound "decent", though probably not transparent or acceptable to most audiophiles.

So based on the listening test, if we're going strictly for highest quality we'd choose Lame. Though my personal preference (and hence suggestion) would be to trade a small, possibly imperceivable reduction in quality (keeping in mind that we've already limited ourselves by chosing 128kbps CBR mp3) of a for more appreciable gain in encoding speed.

This post has been edited by Remedial Sound: May 27 2006, 19:10
Go to the top of the page
+Quote Post
guruboolez
post May 27 2006, 19:48
Post #17





Group: Members (Donating)
Posts: 3474
Joined: 7-November 01
From: Strasbourg (France)
Member No.: 420



QUOTE (Remedial Sound @ May 27 2006, 18:58) *
however I think most would agree that current mp3 encoders at this bitrate are going to sound "decent", though probably not transparent or acceptable to most audiophiles.

Mmmhhhh... some encoders are clearly less decent than LAME at this bitrate I would say. Anyway, the question was: what is the best MP3 encoder at CBR 128, and not which is the fastest-but-decent one wink.gif
LAME is IMO the obvious answer to the original question. For faster encoder, HELIX is maybe as good as fastenc and also faster. Gogo is -according to the same test- a working contender too.
Go to the top of the page
+Quote Post
gameplaya15143
post May 27 2006, 20:40
Post #18





Group: Members
Posts: 484
Joined: 8-January 06
From: Earth
Member No.: 26978



[Opinion]
If I ever need to use cbr 128kbps I will either use
lame 3.93.1 -q 2 -b 128 --nspsytune
or
l3codecp.acm 1.2.0.63
[/Opinion]

The only way to find out which is 'best' is to test a bunch of different encoders, and see which one you like the best.. then use it.


--------------------
Vorbis-q0-lowpass99
lame3.93.1-q5-V9-k-nspsytune
Go to the top of the page
+Quote Post
jmartis
post May 27 2006, 20:48
Post #19





Group: Members
Posts: 381
Joined: 9-April 06
From: Czech Republic
Member No.: 29311



QUOTE (gameplaya15143 @ May 27 2006, 21:40) *
[Opinion]
If I ever need to use cbr 128kbps I will either use
lame 3.93.1 -q 2 -b 128 --nspsytune
or
l3codecp.acm 1.2.0.63
[/Opinion]

The fhg 1.2 build 63 is buggy (it was mentioned earlier; try to encode ma11short sample, really unlistenable blink.gif )

J.M.
Go to the top of the page
+Quote Post
memomai
post May 28 2006, 15:44
Post #20





Group: Members
Posts: 264
Joined: 13-February 05
From: Germany, Kempten
Member No.: 19808



QUOTE
Take a look on the 128 kbps MP3 collective listening test:


guruboolez, how old is this test? which lame version was tested there? why isn't helix tested yet in 128 kbps CBR ?


--------------------
FB2K,APE&LAME
Go to the top of the page
+Quote Post
guruboolez
post May 28 2006, 16:30
Post #21





Group: Members (Donating)
Posts: 3474
Joined: 7-November 01
From: Strasbourg (France)
Member No.: 420



http://www.rjamorim.com/test/index.html
MP3 at 128kbps
from 2004-01-14 to 2004-02-01

Additional info:
http://www.rjamorim.com/test/mp3-128/presentation.html

Real (Helix) encoder was tested. The CLI encoder was released 18 months later.
Go to the top of the page
+Quote Post

Reply to this topicStart new topic
1 User(s) are reading this topic (1 Guests and 0 Anonymous Users)
0 Members:

 



RSS Lo-Fi Version Time is now: 17th April 2014 - 15:27