IPB

Welcome Guest ( Log In | Register )

 
Reply to this topicStart new topic
Converting from lossless to portable, any reason to prefer AAC or MP3?, Was “AAC or MP3 for mobile listening convertyed from Lossless AAC”
izzyfinhaifa
post Dec 4 2011, 00:01
Post #1





Group: Members
Posts: 13
Joined: 11-April 04
Member No.: 13421



I've got lots of CD's that have been backed up to lossless AAC. This is the archive.

I have a few music players, most do AAC, all do MP3. Is there any physical reason why I should choose one format over another, when i convert down for portable listening? I will probably go for 320 MP3, or the equivalent in AAC.

Thanks.
Go to the top of the page
+Quote Post
saratoga
post Dec 4 2011, 00:11
Post #2





Group: Members
Posts: 5162
Joined: 2-September 02
Member No.: 3264



QUOTE (izzyfinhaifa @ Dec 3 2011, 19:01) *
I've got lots of CD's that have been backed up to lossless AAC. This is the archive.


You mean ALAC? AAC isn't lossless. If its really AAC, you may not want to transcode since that would be lossy -> lossy.

QUOTE (izzyfinhaifa @ Dec 3 2011, 19:01) *
Is there any physical reason why I should choose one format over another, when i convert down for portable listening? I will probably go for 320 MP3, or the equivalent in AAC.


Doesn't really matter. Although I wouldn't use sucha ridiculous bitrate on a portable player.

This post has been edited by saratoga: Dec 4 2011, 00:12
Go to the top of the page
+Quote Post
izzyfinhaifa
post Dec 4 2011, 00:21
Post #3





Group: Members
Posts: 13
Joined: 11-April 04
Member No.: 13421



Sorry, I meant ALAC. Thanks for the opinion.

This post has been edited by db1989: Dec 6 2011, 21:42
Reason for edit: removing unnecessary full quote
Go to the top of the page
+Quote Post
Silversight
post Dec 6 2011, 16:14
Post #4





Group: Members
Posts: 310
Joined: 5-April 06
From: Aachen, Germany
Member No.: 29203



At bitrates that high, it really doesn't matter whether you use MP3 or AAC, but as saratoga said before, 320 kbit/s could be considered overkill for a portable player (or more specifically for the usual listening conditions).
AAC mostly has its advantages in bitrate regions below ~120 kbit/s. However, at a certain quality level, the AAC encoders use Spectral Band Replication for high frequency reproduction (a mode called HE-AAC), which might drain your battery faster than the "regular" LC-AAC mode.



--------------------
Nothing is impossible if you don't need to do it yourself.
Go to the top of the page
+Quote Post
izzyfinhaifa
post Dec 7 2011, 02:59
Post #5





Group: Members
Posts: 13
Joined: 11-April 04
Member No.: 13421



QUOTE (Silversight @ Dec 6 2011, 08:14) *
At bitrates that high, it really doesn't matter whether you use MP3 or AAC, but as saratoga said before, 320 kbit/s could be considered overkill for a portable player (or more specifically for the usual listening conditions).
AAC mostly has its advantages in bitrate regions below ~120 kbit/s. However, at a certain quality level, the AAC encoders use Spectral Band Replication for high frequency reproduction (a mode called HE-AAC), which might drain your battery faster than the "regular" LC-AAC mode.


Thanks. What would be the recommended encoder for AAC, and for MP3, (and front end, if appropraite) given that I am going from lossless AAC? The files are all in the format /artist/album,/song format, lots and lots of them.

This post has been edited by izzyfinhaifa: Dec 7 2011, 03:00
Go to the top of the page
+Quote Post
saratoga
post Dec 7 2011, 03:00
Post #6





Group: Members
Posts: 5162
Joined: 2-September 02
Member No.: 3264



160k VBR is probably a good balance. For MP3 lame is recommended, for AAC Nero or iTunes are both good.
Go to the top of the page
+Quote Post
izzyfinhaifa
post Dec 11 2011, 14:00
Post #7





Group: Members
Posts: 13
Joined: 11-April 04
Member No.: 13421



QUOTE (saratoga @ Dec 6 2011, 19:00) *
160k VBR is probably a good balance. For MP3 lame is recommended, for AAC Nero or iTunes are both good.


Thanks. I was thinking of my particular case. Many gigabytes of data stored in a directory structure. Is there a front end that will go up and down that structure with LAME?
Go to the top of the page
+Quote Post
kennedyb4
post Dec 11 2011, 14:41
Post #8





Group: Members
Posts: 772
Joined: 3-October 01
Member No.: 180



Please have a look at the results of the 96kbps second AAC test just held. This was a big eye(ear?) opener for me. Many reported great difficulty picking out the encoded samples even at this low bitrate.

I would test a bit with AAC before encoding at 320 mp3. AAC is a very good format.
Go to the top of the page
+Quote Post
Brand
post Dec 11 2011, 16:04
Post #9





Group: Members
Posts: 321
Joined: 27-November 09
Member No.: 75355



I haven't tried them at low bitrates, but I recently did some ABXing, aiming for transparency. While LAME MP3 was pretty much transparent at V3, AAC at a similar bitrate (170-180) was still easily ABXable, with both Nero and Quicktime.
It was only one sample so I can't draw any definite conclusions, all I'd advise is that you do some testing of your own to see/hear what works best for you.
Compatibility is still in favor of MP3, but eventually that depends on the devices you use.

BTW, here are the files I used for testing: http://www.mediafire.com/?p8t9786go1m4xda

This post has been edited by Brand: Dec 11 2011, 16:06
Go to the top of the page
+Quote Post
C.R.Helmrich
post Dec 11 2011, 16:39
Post #10





Group: Developer
Posts: 694
Joined: 6-December 08
From: Erlangen Germany
Member No.: 64012



Thanks a lot, Kennedy, for introducing the topic starter to an essential piece of information (for the lazy folks: this is what he's talking about). And I agree that anything above ~128 kb AAC is overkill for mobile use.

QUOTE (Brand @ Dec 11 2011, 17:04) *
I haven't tried them at low bitrates, but I recently did some ABXing, aiming for transparency. While LAME MP3 was pretty much transparent at V3, AAC at a similar bitrate (170-180) was still easily ABXable, with both Nero and Quicktime.

That surprises me. When you have time, would you mind trying the same sample on the Winamp AAC encoder, VBR mode 4?

Chris

This post has been edited by C.R.Helmrich: Dec 11 2011, 16:43


--------------------
If I don't reply to your reply, it means I agree with you.
Go to the top of the page
+Quote Post
Brand
post Dec 11 2011, 16:52
Post #11





Group: Members
Posts: 321
Joined: 27-November 09
Member No.: 75355



@ C.R.Helmrich: Sure. Would you mind converting it for me, since I don't have Winamp (or its encoder) at hand? There's a FLAC included in that Mediafire link, which is the same I also uploaded here (original).
(Or let me know how to do that in Foobar.)
Go to the top of the page
+Quote Post
IgorC
post Dec 11 2011, 17:21
Post #12





Group: Members
Posts: 1581
Joined: 3-January 05
From: ARG/RUS
Member No.: 18803



QUOTE (Brand @ Dec 11 2011, 12:04) *
I haven't tried them at low bitrates, but I recently did some ABXing, aiming for transparency. While LAME MP3 was pretty much transparent at V3, AAC at a similar bitrate (170-180) was still easily ABXable, with both Nero and Quicktime.
It was only one sample so I can't draw any definite conclusions, all I'd advise is that you do some testing of your own to see/hear what works best for you.
Compatibility is still in favor of MP3, but eventually that depends on the devices you use.

Well, neither 10 samples is enough to draw any valid conclusion.

The good AAC encoder is always superior to MP3 at any bitrates for my ears. I have tried like 30 samples (dirty in home tests) and my findings are :

Apple LC-AAC and LAME 3.97/3.98/3.99
AAC 96 kbps is better than MP3 V5 ~128 kbps.
AAC 128 kbps is on par with V2 . Another listener reports the same findings http://d.hatena.ne.jp/kamedo2/20111029/1319840519
AAC 160-192 is better than V0-320 kbps.

This post has been edited by IgorC: Dec 11 2011, 17:23
Go to the top of the page
+Quote Post
C.R.Helmrich
post Dec 11 2011, 22:38
Post #13





Group: Developer
Posts: 694
Joined: 6-December 08
From: Erlangen Germany
Member No.: 64012



QUOTE (Brand @ Dec 11 2011, 17:52) *
@ C.R.Helmrich: Sure. Would you mind converting it for me, since I don't have Winamp (or its encoder) at hand? There's a FLAC included in that Mediafire link, which is the same I also uploaded here (original).
(Or let me know how to do that in Foobar.)

I can do that tomorrow, since it seems a new Winamp (and AAC encoder) version will be available then.

Chris


--------------------
If I don't reply to your reply, it means I agree with you.
Go to the top of the page
+Quote Post
/mnt
post Dec 12 2011, 01:53
Post #14





Group: Members
Posts: 697
Joined: 22-April 06
Member No.: 29877



I find QuickTime AAC at 160kbps VBR (CVBR) to be just as good or better then LAME V2 in most cases.

My main reasons to prefer AAC over MP3:

No sbf21 bloat (huge benefit for Metal music).
Better support on iTunes (MP3 support on iTunes is pretty poor IMO).
Decodes faster on foobar2000.
More efficient at lower bitrates.
Support for more channels and freqs.
Less serious pre-echo problems (huge benefit for Electronic music).
A good choice of decent encoders such Nero AAC, QuickTime and FhG AAC.



--------------------
"I never thought I'd see this much candy in one mission!"
Go to the top of the page
+Quote Post
C.R.Helmrich
post Dec 13 2011, 22:46
Post #15





Group: Developer
Posts: 694
Joined: 6-December 08
From: Erlangen Germany
Member No.: 64012



QUOTE (C.R.Helmrich @ Dec 11 2011, 23:38) *

Sorry for the delay. Attached your sample, encoded from the FLAC in your Zip uploaded on mediafire, in the highest 2 VBR modes with Winamp 5.623.

Chris

Attached File  Nas___02___My_Way__original_winampv4.m4a ( 465.03K ) Number of downloads: 77
Attached File  Nas___02___My_Way__original_winampv5.m4a ( 709.29K ) Number of downloads: 77


--------------------
If I don't reply to your reply, it means I agree with you.
Go to the top of the page
+Quote Post
Brand
post Dec 14 2011, 21:09
Post #16





Group: Members
Posts: 321
Joined: 27-November 09
Member No.: 75355



Thanks C.R.

Judging by a quick test with this sample I'd say this Winamp encoder is very good, definitely the best AAC encoder and better than LAME as well.
(I also have to take back what I wrote about V3 being pretty much transparent, since I was later able to easily ABX it.)

For this sample, the Winamp 4 version is comparable to if not even a bit better than LAME V2 in transparency and to 190+ kbps AACs from Nero and Quicktime. It's quite an achievement and even a bit suspect* for such a low bitrate, I'll see if I'll be able to reproduce this other times and with other samples.

*I'm wondering if it could be due to something simple like loudness/amplitude, which perhaps other encoders don't convert as accurately, but Winamp does... or maybe I was focusing on a single sound, neglecting others.. BTW, for me the ABX spot I focused on for this sample are the snare hits. LAME and Nero/QT make them lose some of the clarity and punch/loudness, whereas Winamp seems to keep them almost intact.

This post has been edited by Brand: Dec 14 2011, 21:09
Go to the top of the page
+Quote Post
C.R.Helmrich
post Dec 15 2011, 09:09
Post #17





Group: Developer
Posts: 694
Joined: 6-December 08
From: Erlangen Germany
Member No.: 64012



QUOTE (Brand @ Dec 14 2011, 22:09) *
*I'm wondering if it could be due to something simple like loudness/amplitude, which perhaps other encoders don't convert as accurately, but Winamp does...

You might be right. The QuickTime encoder used to apply a - in some cases audible - level compressor to the input signal in previous versions (including the one in this year's 64-kb public listening test here on HA) to avoid clipping after decoding. Don't know if it still does, though. The Fraunhofer AAC encoder certainly doesn't.

Chris

This post has been edited by C.R.Helmrich: Dec 15 2011, 09:10


--------------------
If I don't reply to your reply, it means I agree with you.
Go to the top of the page
+Quote Post
lvqcl
post Dec 15 2011, 15:44
Post #18





Group: Developer
Posts: 3468
Joined: 2-December 07
Member No.: 49183



QUOTE (C.R.Helmrich @ Dec 15 2011, 12:09) *
Don't know if it still does, though.


It isn't. (Or at least I cannot find any trace of this)
Go to the top of the page
+Quote Post
kennedyb4
post Dec 15 2011, 23:35
Post #19





Group: Members
Posts: 772
Joined: 3-October 01
Member No.: 180



Just a quick re direct.

The samples in this test,
http://listening-tests.hydrogenaudio.org/i...-a/results.html

were codec killers and the QT scores were way above 4 meaning that any artifacts were perceptible but not annoying. This is frequently after multiple trials to find a suspected artifact and repeated headphone abx runs to confirm the error.

I have to concur that anything above 128 QT constrained or TVBR is overkill on a portable.

Samples from less complex passages re even harder to pick up. I found myself trying to pick out the lowpasses rather than actual artifacts.
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: 25th December 2014 - 20:19