IPB

Welcome Guest ( Log In | Register )

2 Pages V   1 2 >  
Reply to this topicStart new topic
LAME and ipod shuffle, bitrate
KevinB52379
post Jun 5 2006, 01:21
Post #1





Group: Members
Posts: 24
Joined: 27-April 03
Member No.: 6223



Hi all,
After encoding my collection to AAC 160kbps using iTunes, I found a lot of the songs had problems..even with error correction. They would start skipping.

I found another program that I like to use called Easy CD-DA Extractor. I wanted to use MP3 for compatibility reasons.

I was curious to hear opinions..how does 160kpbs in latest lame recommendation (3.97 beta 2) compare to itunes AAC at 160kbps. Are they pretty similar? The AAC quality sounded good to me at that bitrate. I know lame developers have made vast improvements to its encoder and it competes with AAC very well.

I was thinking of using mp3 so it can work both with my shuffle and my CD player that supports mp3 and wma (i refuse to use wma).

I tried to do a search on this but not much was coming up.

Thanks for your help.
Kevin
PS I thought I remembered reading about some listening tests of bitrates of 160kbps..but I couldn't find that either.

If 160kbps lame isn't comprable to 160kbps aac files with itunes, then i will probably use alt preset standar.
Go to the top of the page
+Quote Post
ConCave
post Jun 5 2006, 01:41
Post #2





Group: Members
Posts: 47
Joined: 16-October 05
Member No.: 25153



If your using a shuffle i would go for lame -V 5 it will be transparent through your headphones especially with background noise and you could fit a lot more mp3s on it. Also think about using EAC with secure mode (See the Wiki for a set up guide Wiki) you wont get those annoying skips, this is due to your cd having slight scratchs and iTunes wouldnt have as good an error correction as EAC.

Buy the way you can also try the Nero AAC encoder if you want AAC. Just try stuff out and see what suites you best. Hope this helps.

ConCave


Edit: Spelling

This post has been edited by ConCave: Jun 5 2006, 01:46
Go to the top of the page
+Quote Post
amobms
post Jun 5 2006, 01:57
Post #3





Group: Members
Posts: 15
Joined: 11-April 06
Member No.: 29418



Kevin - I suggest you peruse the listening tests as a start. The most recent 128kbps multiformat test will probably be of interest to you:

Multiformat at 128kbps public listening test

LAME -V5 and Apple AAC 128kbps were statistically tied in this most recent 128k test. I know a lot of people who use LAME -V4 (target 165kbps) on their iPods as a compatibilty alternative to AAC. I switched from LAME -V2 to the lower -V4 bitrate VBR setting after quite a bit of my own ABX testing. I've been completely happy with my Shuffle's performance with -V4 encodes.

I don't recommend you use a CBR setting with LAME unless you are making your encoding decision at gunpoint. Best of luck...
Go to the top of the page
+Quote Post
KevinB52379
post Jun 5 2006, 06:34
Post #4





Group: Members
Posts: 24
Joined: 27-April 03
Member No.: 6223



QUOTE (amobms @ Jun 4 2006, 16:57) *
Kevin - I suggest you peruse the listening tests as a start. The most recent 128kbps multiformat test will probably be of interest to you:

Multiformat at 128kbps public listening test

LAME -V5 and Apple AAC 128kbps were statistically tied in this most recent 128k test. I know a lot of people who use LAME -V4 (target 165kbps) on their iPods as a compatibilty alternative to AAC. I switched from LAME -V2 to the lower -V4 bitrate VBR setting after quite a bit of my own ABX testing. I've been completely happy with my Shuffle's performance with -V4 encodes.

I don't recommend you use a CBR setting with LAME unless you are making your encoding decision at gunpoint. Best of luck...


Thank you so much. The information you provided is exactly what I was looking for. You have been most helpful. smile.gif If anyone wants to add some more comments, please feel free to do so.

QUOTE (KevinB52379 @ Jun 4 2006, 21:26) *
QUOTE (amobms @ Jun 4 2006, 16:57) *

Kevin - I suggest you peruse the listening tests as a start. The most recent 128kbps multiformat test will probably be of interest to you:

Multiformat at 128kbps public listening test

LAME -V5 and Apple AAC 128kbps were statistically tied in this most recent 128k test. I know a lot of people who use LAME -V4 (target 165kbps) on their iPods as a compatibilty alternative to AAC. I switched from LAME -V2 to the lower -V4 bitrate VBR setting after quite a bit of my own ABX testing. I've been completely happy with my Shuffle's performance with -V4 encodes.

I don't recommend you use a CBR setting with LAME unless you are making your encoding decision at gunpoint. Best of luck...


Thank you so much. The information you provided is exactly what I was looking for. You have been most helpful. smile.gif If anyone wants to add some more comments, please feel free to do so.


EDIT: Maybe I'm totally missing something, but where are the results to this test on the page, or have they not been done yet? thanks
Go to the top of the page
+Quote Post
audiomars
post Jun 5 2006, 07:06
Post #5





Group: Members
Posts: 135
Joined: 16-September 05
From: India
Member No.: 24505



QUOTE (KevinB52379 @ Jun 5 2006, 11:04) *
EDIT: Maybe I'm totally missing something, but where are the results to this test on the page, or have they not been done yet? thanks


The results are here


--------------------
Reason is immortal, all else mortal
- Pythagoras
Go to the top of the page
+Quote Post
odious_m
post Jun 5 2006, 07:15
Post #6





Group: Members
Posts: 78
Joined: 4-June 06
Member No.: 31428



QUOTE (audiomars @ Jun 5 2006, 01:06) *
QUOTE (KevinB52379 @ Jun 5 2006, 11:04) *

EDIT: Maybe I'm totally missing something, but where are the results to this test on the page, or have they not been done yet? thanks


The results are here


Nope, that's an older test using LAME 3.96. Here are AFAIK the most recent test results.
Go to the top of the page
+Quote Post
Sebastian Mares
post Jun 5 2006, 07:16
Post #7





Group: Members
Posts: 3633
Joined: 14-May 03
From: Bad Herrenalb
Member No.: 6613



You guys know that you didn't post the links to the most recent multiformat listening test, but to the last one that Roberto conducted?

Most recent test is http://www.maresweb.de/listening-tests/mf-128-1/ and the results are http://www.maresweb.de/listening-tests/mf-128-1/results.htm

This post has been edited by Sebastian Mares: Jun 5 2006, 07:16


--------------------
http://listening-tests.hydrogenaudio.org/sebastian/
Go to the top of the page
+Quote Post
odious_m
post Jun 5 2006, 07:20
Post #8





Group: Members
Posts: 78
Joined: 4-June 06
Member No.: 31428



QUOTE (Sebastian Mares @ Jun 5 2006, 01:16) *
You guys know that you didn't post the links to the most recent multiformat listening test, but to the last one that Roberto conducted?


Nope, I linked to yours.
Go to the top of the page
+Quote Post
Sebastian Mares
post Jun 5 2006, 08:38
Post #9





Group: Members
Posts: 3633
Joined: 14-May 03
From: Bad Herrenalb
Member No.: 6613



QUOTE (odious_m @ Jun 5 2006, 08:20) *
QUOTE (Sebastian Mares @ Jun 5 2006, 01:16) *

You guys know that you didn't post the links to the most recent multiformat listening test, but to the last one that Roberto conducted?


Nope, I linked to yours.


I was referring to the other two posters. I didn't see your post - we must've posted at the same time. smile.gif


--------------------
http://listening-tests.hydrogenaudio.org/sebastian/
Go to the top of the page
+Quote Post
amobms
post Jun 5 2006, 14:06
Post #10





Group: Members
Posts: 15
Joined: 11-April 06
Member No.: 29418



QUOTE (Sebastian Mares @ Jun 5 2006, 02:16) *
You guys know that you didn't post the links to the most recent multiformat listening test, but to the last one that Roberto conducted?

Most recent test is http://www.maresweb.de/listening-tests/mf-128-1/ and the results are http://www.maresweb.de/listening-tests/mf-128-1/results.htm

Thanks, Sebastian. Do these tests have a central repository? Perhaps if they don't, this is the place to archive them...
Go to the top of the page
+Quote Post
Firon
post Jun 5 2006, 14:36
Post #11





Group: Members
Posts: 830
Joined: 3-November 05
Member No.: 25526



The Listening Tests subforum has pretty much all the tests.
Go to the top of the page
+Quote Post
amobms
post Jun 5 2006, 14:48
Post #12





Group: Members
Posts: 15
Joined: 11-April 06
Member No.: 29418



QUOTE (Firon @ Jun 5 2006, 09:36) *
The Listening Tests subforum has pretty much all the tests.

Here they are...

Listening Tests
Go to the top of the page
+Quote Post
KevinB52379
post Jun 5 2006, 22:24
Post #13





Group: Members
Posts: 24
Joined: 27-April 03
Member No.: 6223



First I apologize for not posting this in the correct forum (I see it has been moved).

I was just playing around with the -V settings. I tried -V4 but wasn't happy with the low bitrate results. I then tried -V3 and it seems to hover more around the 160kbps to 180kbps range.

I was playing around with Easy CD-DA Extractor and it has tons of audio formats (vorbis, wma, mp3 (via lame and fraunhofer acm) lossless compression, and then i notice it has AAC and M4a support.

I know that the ipod shuffle uses the m4a format when encoding aac files in itunes.

So I guess my new question would be, should I use 160kbps in the m4a format? I liked apple's 160kbps aac files. My only problem is that I don't know what encoder this program is using.

Does anyone have any ideas?
Thanks
Kevin
Go to the top of the page
+Quote Post
amobms
post Jun 6 2006, 14:13
Post #14





Group: Members
Posts: 15
Joined: 11-April 06
Member No.: 29418



QUOTE (KevinB52379 @ Jun 5 2006, 17:24) *
I was just playing around with the -V settings. I tried -V4 but wasn't happy with the low bitrate results. I then tried -V3 and it seems to hover more around the 160kbps to 180kbps range.

Kevin - Did you perform a double blind comparison to reach that conclusion? If not I suggest you do. There is a reasonable likelyhood that you are being fooled by the placebo effect.

Placebo Effect
Blind Test

While most folks around here are happy to help, there is also an expectation that the person seeking help or guidance will also use objective means to form an opinion or reach a conclusion. If it's clear that this isn't the case, your requests will likely be passed over by this dominantly objective community.
Go to the top of the page
+Quote Post
kornchild2002
post Jun 6 2006, 17:03
Post #15





Group: Members
Posts: 2080
Joined: 8-April 05
From: Cincinnati, OH
Member No.: 21277



QUOTE (KevinB52379 @ Jun 5 2006, 15:24) *
First I apologize for not posting this in the correct forum (I see it has been moved).

I was just playing around with the -V settings. I tried -V4 but wasn't happy with the low bitrate results. I then tried -V3 and it seems to hover more around the 160kbps to 180kbps range.

I was playing around with Easy CD-DA Extractor and it has tons of audio formats (vorbis, wma, mp3 (via lame and fraunhofer acm) lossless compression, and then i notice it has AAC and M4a support.

I know that the ipod shuffle uses the m4a format when encoding aac files in itunes.

So I guess my new question would be, should I use 160kbps in the m4a format? I liked apple's 160kbps aac files. My only problem is that I don't know what encoder this program is using.

Does anyone have any ideas?
Thanks
Kevin


OK, it doesn't really matter what the bitrate comes out to. For example, I encode all my music with Lame 3.97b2 at the -V 2 --vbr-new setting (you always want to use --vbr-new). This is supposed to have a target bitrate of around 190kbps VBR. Well, sometimes songs actually come out with 142kbps or 160kbps VBR bitrates even though I have selected a higher bitrate setting. This doesn't matter as the quality level of these songs is at the -V 2 setting. That is the thing about the different Lame -V X settings. You aren't picking a target bitrate, you are picking a target quality. I have even had some songs come out to 190kbps VBR using -V 4, its bitrate should be around 160kbps VBR.

As previously pointed out, do a blind ABX listening test. Your eyes and brain are fooling your ears, this is the placebo affect. Where you think you can hear a difference because you know the bitrate is low. I am willing to bet that you will fail a ABX test at -V 5 --vbr-new with Lame 3.97b2 or you will fail when using 128kbps VBR mpeg-4 AAC with iTunes. Seriously, if you are using the stock Apple earbuds, you won't hear a difference. Those earbuds are physically incapable of expressing the flaws of lossy formats at bitrates higher than 96kbps.

You should really look at using EAC. EAC offers all of the same features as Easy CD-DA and then some. Easy CD-DA has some error correction implementations but they are no where near the quality of EAC's. If you were having trouble with iTunes then you might have trouble with Easy CD-DA.

If you want to use the iTunes mpeg-4 AAC encoder, then there is a EAC command line program called iTunes Encode. You will use EAC to rip a CD to your hard drive in the wav format, then it will open up iTunes and convert those wav files to mpeg-4 AAC (or whatever CD ripping settings you have in iTunes), and then it will delete those source wav files and add track tag information to the lossy files. I suggest that you take a look at Nero's AAC encoder. Nero recently released a command line version of their encoder for free. This means that you don't have to download Nero 7 to experience Nero's AAC encoder. Nero has completely re-written their AAC encoder from the ground up. I am not aware of any listening tests on the new Nero AAC encoder but I have heard good things from other people.

Trust me, I switched from using the mpeg-4 AAC format (iTunes AAC) and went to Lame mp3. I have not looked back since. mp3 has greater compatiblity with hardware devices and the Lame mp3 encoder continues to push the mp3 format past its limits. If you plan on letting friends listen to your music, carry around about 100 songs on a CD so that it can be played on any mp3 CD car CD deck, or plan on getting a different portable digital audio player, then use the Lame mp3 format. If you don't plan on sharing, don't care about not having mp3 CD's, or only plan on using Sony or Apple portable digital audio players, then use the AAC format (either iTunes' or Nero's AAC encoder). I use Lame mp3 because I want that "universal" compatibility. I have a car CD deck that plays mp3 CD's. This comes in really handy as I can throw in a mp3 CD and not have to change it for a month. I can fit around 100 songs on a 700MB CD of which are encoded at -V 2 --vbr-new with Lame 3.97b2/b1. From my experience, on the mp3 CD car decks that I have messed around with, playing a mp3 CD provides higher quality than hooking up a iPod (or other portable) through a line input on the CD deck.
Go to the top of the page
+Quote Post
KevinB52379
post Jun 6 2006, 21:02
Post #16





Group: Members
Posts: 24
Joined: 27-April 03
Member No.: 6223



I decided to take your advice and listen to the mp3 results. I used -V4 and it sounded absolutely fine to me...just like the 160kbps AAC itune files. smile.gif

I apologize for not listening first.

I'm not using the stock ibuds by the way, but still using a cheap pair of philips headphones ($10 dollar foldable ones..I can't use buds because I wear hearing aids).

One last question. I notice in the -V4 setting, the mp3 quality is set to normal. Should I set it there or should the quality be, higher or best quality? I'm not talking about the actual -V choices, but how you can choose within them whether to have normal, higher or highest quality.

I remember reading a long time ago that using higher or highest quality setting used to produce nasty artifacts..is this still true today? Any pros or cons in changing the quality settings?

I apologize for all the questions, and I hope someone is willing to help me. All the suggestions have been helpful so far.
Thanks
Kevin
Go to the top of the page
+Quote Post
amobms
post Jun 6 2006, 22:14
Post #17





Group: Members
Posts: 15
Joined: 11-April 06
Member No.: 29418



QUOTE (KevinB52379 @ Jun 6 2006, 16:02) *
One last question. I notice in the -V4 setting, the mp3 quality is set to normal. Should I set it there or should the quality be, higher or best quality? I'm not talking about the actual -V choices, but how you can choose within them whether to have normal, higher or highest quality.

What software are you referring to? EAC does offer some default quality settings, but these are overridden by your "-V4 --vbr-new" command line. If you're referring to Easy CD-DA Extractor, I strongly suggest you switch to EAC - it is simply a better ripper.

There are a few thousand HA members here who will support that recommendation. We can point you to resources for configuration settings, encoder settings, etc...

Edit: typo

This post has been edited by amobms: Jun 6 2006, 22:15
Go to the top of the page
+Quote Post
KevinB52379
post Jun 7 2006, 03:10
Post #18





Group: Members
Posts: 24
Joined: 27-April 03
Member No.: 6223



Here are a couple of snapshots of Easy CD-DA Extractor. I hope they will clarify my statements:

http://www.uploadfile.info/uploads/ee78546282.jpg

http://www.uploadfile.info/uploads/e9fe0e9aeb.jpg

As you can see, there are quality settings for each -V setting. Easy CD-DA Extractor uses lame dll (3.97).

Also easy cd-da extractor has "error correction and recovery" option, which I think will suffice. I paid for this program and really enjoy it..so I'd like to stick with it, but thanks for the EAC suggestions.

EDITED to fix picture urls and add comments.

This post has been edited by KevinB52379: Jun 7 2006, 03:58
Go to the top of the page
+Quote Post
saratoga
post Jun 7 2006, 06:01
Post #19





Group: Members
Posts: 5119
Joined: 2-September 02
Member No.: 3264



QUOTE (KevinB52379 @ Jun 6 2006, 19:10) *
Here are a couple of snapshots of Easy CD-DA Extractor. I hope they will clarify my statements:

http://www.uploadfile.info/uploads/ee78546282.jpg

http://www.uploadfile.info/uploads/e9fe0e9aeb.jpg

As you can see, there are quality settings for each -V setting. Easy CD-DA Extractor uses lame dll (3.97).


Those quality settings don't corrispond to anything in LAME, AFAIK, so hopefully they don't do anything. I suppose you could try changing them and seeing if the output MP3 changes.

Not sure I'd bother with it though. I don't trust programs that use the lame dll. Its better to use command line encoder so that you know whats actually happening.
Go to the top of the page
+Quote Post
Gabriel
post Jun 7 2006, 09:22
Post #20


LAME developer


Group: Developer
Posts: 2950
Joined: 1-October 01
From: Nanterre, France
Member No.: 138



QUOTE (KevinB52379 @ Jun 6 2006, 22:02) *
I'm not using the stock ibuds by the way, but still using a cheap pair of philips headphones ($10 dollar foldable ones..I can't use buds because I wear hearing aids).

A little warning regarding audio encoders and hearing aids. If you have hearing aids, then it is very likely that your audition, even once corrected/assisted, does not really follow the "typical" audition. Thus, as encoders are using models based on "typical/unaltered" audition, your perception of encoded files is likely to be different than the perception from other people.

In your specific case, I'd suggest you to always test yourself encoder settings instead of relying on group listening tests.
Go to the top of the page
+Quote Post
kornchild2002
post Jun 7 2006, 12:27
Post #21





Group: Members
Posts: 2080
Joined: 8-April 05
From: Cincinnati, OH
Member No.: 21277



As others have stated, those quality settings shouldn't change anything regarding Lame mp3 encoding. For example, when typing in a Lame command, you are using -V 4 --vbr-new. You are just visually selecting different options instead of typing that in. Well, Lame has different quality settings. You take that command further by adding -V 4 --vbr-new -Q 1 (I think it goes 0 through 9 but don't quote me). With Lame 3.97b1/b2, these different -Q values don't really do anything in certain ranges. -Q 0 (which is the highest quality) produces the same quality as -Q 3, I think -Q 4 produces the same quality as -Q 7, etc. By default, Lame is set to -Q 2 which I think is concidered to be "normal" with Easy CD-DA.

In other words, changing the quality levels shouldn't change the quality of the Lame encoded mp3. You should see if Easy CD-DA has an option for a command line encoder as you would be much better off using the Lame.exe command line program instead of using the Lame.dll.

Eventhough you paid for Easy CD-DA, I would still look at using EAC. You could use Easy CD-DA for most of your material but there might be a few CD's that Easy CD-DA has trouble with, thus you would either have to use EAC (which is 100% free) or run out and purchase a disc re-surfacing tool (~$30).
Go to the top of the page
+Quote Post
AstralStorm
post Jun 7 2006, 13:09
Post #22





Group: Members
Posts: 745
Joined: 22-April 03
From: /dev/null
Member No.: 6130



QUOTE (kornchild2002 @ Jun 7 2006, 13:27) *
As others have stated, those quality settings shouldn't change anything regarding Lame mp3 encoding. For example, when typing in a Lame command, you are using -V 4 --vbr-new. You are just visually selecting different options instead of typing that in. Well, Lame has different quality settings. You take that command further by adding -V 4 --vbr-new -Q 1 (I think it goes 0 through 9 but don't quote me). With Lame 3.97b1/b2, these different -Q values don't really do anything in certain ranges. -Q 0 (which is the highest quality) produces the same quality as -Q 3, I think -Q 4 produces the same quality as -Q 7, etc. By default, Lame is set to -Q 2 which I think is concidered to be "normal" with Easy CD-DA.


Wrong wrt Q-levels. They really do different things. With -q 0, the best I can get is about 1x encoding speed and the files are really slightly higher quality (-V 6, ABXed) and don't take more space. -q 0 and -q 1 did little/nothing in the days of LAME 3.90.3. Not anymore.
The default is -q 3 and --high = -q 2. (and still uses reasonable time for encoding)

This post has been edited by AstralStorm: Jun 7 2006, 13:12


--------------------
ruxvilti'a
Go to the top of the page
+Quote Post
robert
post Jun 7 2006, 14:04
Post #23


LAME developer


Group: Developer
Posts: 788
Joined: 22-September 01
Member No.: 5



@AstralStorm
kornchild2002 spoke about the new VBR mode of LAME 3.97.

This post has been edited by robert: Jun 7 2006, 14:05
Go to the top of the page
+Quote Post
AstralStorm
post Jun 7 2006, 15:21
Post #24





Group: Members
Posts: 745
Joined: 22-April 03
From: /dev/null
Member No.: 6130



That I've yet to check, sorry about that then.


--------------------
ruxvilti'a
Go to the top of the page
+Quote Post
KevinB52379
post Jun 7 2006, 18:36
Post #25





Group: Members
Posts: 24
Joined: 27-April 03
Member No.: 6223



I guess I don't understand. I tried using lame_enc.dll in audiograbber (3.97b2) by choosing vbr level 4, vbr mode as new, quality as normal, and i chose joint stereo.

The results came out exactly the same as when I used the commandline -V 4 --vbr-new in lame 3.97b2 executable (found in the same package). It had the same stereo frames, bitrate histogram, same number of frames...everything was exactly the same. So why can't the lame_enc.dll be used?

The funny thing is the lame v4 mode in easy cd-da extractor seems to be very identical to the above but just slightly different (avg bitrate of a song was 156kbps and it's lame header says lame 3.97, meanwhile the lame 3.97b2 bundle (with header 3.97b2) has an avg bitrate of 154kbps).

I am assuming these slight differences are from slight differences from complilers used.

As you can see I'm a little confused.

I can however use -V 4 --vbr-new in lame v 3.97b2 executable with easy cd-da extractor (telling it to send to external encoder) and it tags the songs for me. Would I be better off just using the command line?
Go to the top of the page
+Quote Post

2 Pages V   1 2 >
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: 28th November 2014 - 08:14