IPB

Welcome Guest ( Log In | Register )

48 kbps AAC Encoders Test - Q1 2006 Edition
Ivan Dimkovic
post Dec 14 2005, 14:45
Post #1


Nero MPEG4 developer


Group: Developer
Posts: 1466
Joined: 22-September 01
Member No.: 8



Hello,

This is the invitation for a discussion about 48 Kbps AAC test, that should be done before the next 48 kbps multiformat listening test.

Gabriel from LAME development team and myself decided to organize this test - there are few reasons why this test should be organized, and why it should be organized before the next 48 kbps multiformat test:

Reasons:

* Test the state-of-the-art HE-AAC v1 and HE-AAC v2 (Parametric Stereo) commercial encoders, as well as non-commercial 3GPP reference encoder

* Evaluate the usefullnes of Parametric Stereo tool at bit rate of 48 Kb/s

* Make the pre-selection of the best HE-AAC/HE-AAC v2 solutions for the multiformat listening test

* Compare all these solutions with the low-anchored Low Complexity (LC) AAC, in order to verify how the technology has improved over past few years

* Compare against higher bit rate MP3 to verify how much could state-of-the-art 48 kbps codec approach the good old MP3 technology at significantly higher bit rate.

Which high bit rate of MP3 should be used is debatable - should it be 80, 96 or 128 kbps - as stated above, this is open for debate.

Codecs

At this point (alphabetically):

- 3GPP Reference Encoder
- Coding Technologies aacPlus (HE-AAC)
- Coding Technologies Enhanced aacPlus (HE-AAC v2)
- Nero Digital HE-AAC v1
- Nero Digital HE-AAC v2

Anchors:
- LAME MP3 (high) - bit rate yet TBD / High Anchor
- QuickTime/iTunes LC-AAC @48 kbps / Low Anchor

Samples

My idea was to use combined sample set from multiformat 128 kbps listening test and previous Roberto's 64 kbps multiformat test - I thought of chosing several (5-6) samples from both of these two sets by using the worst scored samples.

Worst scored = Lowest average subjective difference grade (SDG) for all codecs in the particular set, except the low and (where applicable) high anchors.

Gabriel's addition was:

* 1 spoken (german male speech?)
* 1 with strong L/R differences, in order to know how PS fares in bad conditions. Something like some old Beatles songs, with singer on 1 channel and instruments on the other one.

Sample selection is definitely open for further discussion here smile.gif

When

Test should happen at the end of January - to make enough break after the current 128 kbps multiformat test.

Encoding Settings, Encoders, Bit Rates

Like with all other tests, we won't be testing closed or produced solutions - encoders used in the test will be widely available - as well as encoding settings and encoding material. Regarding the Coding Technologies' aacPlus encoder, my proposal is to use their imlpementation in the Winamp, which is widely available product.

Prior to the test, we will conduct public bit rate verification - in order to make sure codecs behave within the bitrate constraints.
Go to the top of the page
+Quote Post
 
Start new topic
Replies
Gabriel
post Dec 14 2005, 18:43
Post #2


LAME developer


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



VBR/CBR issue:
As you know, modern codecs are using a bit reservoir. It can be directly handled in frames (like mp3), or handled by a regulation part (like some mpeg4 modes).
If your encoder respects the normative constraints regarding CBR, then it is cbr, altough its instantaneous bitrate can vary.

CBR is a way to be sure that you can transmit the stream using a given bandwidth, provided that the decoder has a specific buffer size.

The mp3 cbr is very restricted, as the bit reservoir is only 4088 bits. The AAC reservoir is way higher than that, allowing more large instantaneous fluctuations while still beeing cbr.
If you are using VBR but are enforcing a given average bitrate over a given sliding time slice, then the same bit allocation could be achieved in cbr with a bit reservoir (or whatever its name) that would be the size of your given time slice.

IE: if allowed instantaneous variation is big enough, VBR does not provide any advantage while trying to keep a resonable decoding buffer and a given average bitrate.

question: what is the typical aac buffer size for cbr @48kbps?

edit: bytes -> bits

This post has been edited by Gabriel: Dec 14 2005, 23:43
Go to the top of the page
+Quote Post
guruboolez
post Dec 14 2005, 18:51
Post #3





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



QUOTE (Gabriel @ Dec 14 2005, 06:43 PM)
CBR is a way to be sure that you can transmit the stream using a given bandwidth, provided that the decoder has a specific buffer size.
*

This is true. The question is: do we need to ensure streaming conditions for 48 kbps encodings? If not, I don't see any reason to arbitrary discard VBR. Or is there something special with 48 kbps compared to 128 or 32 kbps?
Go to the top of the page
+Quote Post

Posts in this topic
- Ivan Dimkovic   48 kbps AAC Encoders Test - Q1 2006 Edition   Dec 14 2005, 14:45
- - elmar3rd   QUOTE Prior to the test, we will conduct public bi...   Dec 14 2005, 15:15
|- - guruboolez   QUOTE (elmar3rd @ Dec 14 2005, 03:15 PM)So th...   Dec 14 2005, 16:52
|- - IgorC   For many people here 48 kbit/s has sence. It was m...   Dec 14 2005, 17:07
||- - guruboolez   QUOTE (IgorC @ Dec 14 2005, 05:07 PM)For many...   Dec 14 2005, 17:19
||- - IgorC   Somebody want test at 32 kbps , another at 64 and ...   Dec 14 2005, 17:27
||- - guruboolez   QUOTE (IgorC @ Dec 14 2005, 05:27 PM)Somebody...   Dec 14 2005, 17:33
||- - dimzon   QUOTE (guruboolez @ Dec 14 2005, 08:33 PM)Why...   Dec 14 2005, 17:40
||- - Shade[ST]   QUOTE (guruboolez @ Dec 14 2005, 10:33 AM)If ...   Dec 14 2005, 17:41
|||- - guruboolez   QUOTE (Shade[ST] @ Dec 14 2005, 05:4...   Dec 14 2005, 17:53
||||- - Shade[ST]   QUOTE (guruboolez @ Dec 14 2005, 10:53 AM)QUO...   Dec 14 2005, 18:02
|||- - Atlantis   QUOTE (Shade[ST] @ Dec 14 2005, 06:4...   Dec 14 2005, 18:03
||- - IgorC   QUOTE (guruboolez @ Dec 14 2005, 08:33 AM)QUO...   Dec 14 2005, 17:42
||- - guruboolez   QUOTE (IgorC @ Dec 14 2005, 05:42 PM)I dont m...   Dec 14 2005, 18:00
||- - IgorC   QUOTE (guruboolez @ Dec 14 2005, 09:00 AM)QUO...   Dec 14 2005, 18:10
||- - dimzon   QUOTE (IgorC @ Dec 14 2005, 09:10 PM)Itīs not...   Dec 14 2005, 18:13
|||- - IgorC   QUOTE (dimzon @ Dec 14 2005, 09:13 AM)QUOTE (...   Dec 14 2005, 18:16
|||- - dimzon   QUOTE (IgorC @ Dec 14 2005, 09:16 PM)+/- 8 kb...   Dec 14 2005, 18:21
||- - guruboolez   QUOTE (IgorC @ Dec 14 2005, 06:10 PM)The idea...   Dec 14 2005, 18:56
||- - dimzon   QUOTE (guruboolez @ Dec 14 2005, 09:56 PM)QUO...   Dec 14 2005, 19:04
|- - Shade[ST]   QUOTE (guruboolez @ Dec 14 2005, 09:52 AM)I...   Dec 14 2005, 17:12
|- - Sebastian Mares   QUOTE (guruboolez @ Dec 14 2005, 04:52 PM)Are...   Dec 14 2005, 20:44
|- - Serge Smirnoff   QUOTE (Sebastian Mares @ Dec 14 2005, 10:44 P...   Dec 15 2005, 00:21
- - Shade[ST]   I have a few samples which you might like, and are...   Dec 14 2005, 15:21
|- - [JAZ]   About High Anchor: If MP3, then, a CBR (or ABR) 1...   Dec 14 2005, 16:22
- - dimzon   And please add avoTuVo Vorbis @ 48kbsp to - just t...   Dec 14 2005, 16:26
|- - Shade[ST]   QUOTE (dimzon @ Dec 14 2005, 09:26 AM)And ple...   Dec 14 2005, 16:28
|- - dimzon   QUOTE (Shade[ST] @ Dec 14 2005, 07:2...   Dec 14 2005, 16:30
|- - Shade[ST]   There are over 4 implementations of AAC-HE, and a ...   Dec 14 2005, 16:32
|- - dimzon   Oh! And don't forget - Coding Technologie...   Dec 14 2005, 16:37
- - Gabriel   QUOTE So this test will be a mix of CBR (aacPlus) ...   Dec 14 2005, 16:38
- - Ivan Dimkovic   QUOTE 48 kb/s ist very interesting for streaming i...   Dec 14 2005, 16:40
|- - dimzon   QUOTE (Ivan Dimkovic @ Dec 14 2005, 07:40 PM)...   Dec 14 2005, 17:02
- - Ivan Dimkovic   Hmm - nobody really tested WMA @48 kbps vs. anythi...   Dec 14 2005, 17:09
|- - dimzon   QUOTE (Ivan Dimkovic @ Dec 14 2005, 08:09 PM)...   Dec 14 2005, 17:15
- - Ivan Dimkovic   I wouldn't go interpolating results that easil...   Dec 14 2005, 17:18
- - dimzon   I beleive it wood be much more usable to add old H...   Dec 14 2005, 17:18
- - stephanV   At the 32 kbps test WMA and Real were tied with Vo...   Dec 14 2005, 17:23
- - Ivan Dimkovic   QUOTE guruboolez Right. But if people have a 128 k...   Dec 14 2005, 17:33
|- - IgorC   QUOTE (Ivan Dimkovic @ Dec 14 2005, 08:33 AM)...   Dec 14 2005, 17:36
|- - dimzon   QUOTE (IgorC @ Dec 14 2005, 08:36 PM)Finally ...   Dec 14 2005, 17:43
- - Ivan Dimkovic   Well - theoretically we (Nero) could do CBR, quali...   Dec 14 2005, 17:43
|- - dimzon   QUOTE (Ivan Dimkovic @ Dec 14 2005, 08:43 PM)...   Dec 14 2005, 17:48
- - Ivan Dimkovic   QUOTE WOW!!!!!!!! ...   Dec 14 2005, 18:07
- - Gabriel   VBR/CBR issue: As you know, modern codecs are usin...   Dec 14 2005, 18:43
|- - guruboolez   QUOTE (Gabriel @ Dec 14 2005, 06:43 PM)CBR is...   Dec 14 2005, 18:51
|- - IgorC   QUOTE (Gabriel @ Dec 14 2005, 09:43 AM)The mp...   Dec 15 2005, 15:39
- - Ivan Dimkovic   QUOTE question: what is the typical aac buffer siz...   Dec 14 2005, 19:04
- - guruboolez   Something is worrying me. The purpose of this test...   Dec 14 2005, 19:14
|- - Shade[ST]   I don't mind preparing a Vorbis + MP3 test at ...   Dec 14 2005, 19:37
|- - [JAZ]   48kbps Streaming IS reasonable : http://www.polsk...   Dec 14 2005, 20:34
- - NoXFeR   Well, in my opinion, the test should be about how ...   Dec 15 2005, 02:00
- - guruboolez   [JAZ] & Serge Smirnoff> I know that 48 kbps...   Dec 15 2005, 10:35
|- - IgorC   So you are pruposing a 32 kbit/s test, right?   Dec 15 2005, 14:54
|- - guruboolez   QUOTE (IgorC @ Dec 15 2005, 02:54 PM)So you a...   Dec 15 2005, 15:01
|- - IgorC   It was mentioned 48 kbit/s is very interesting t...   Dec 15 2005, 15:09
||- - guruboolez   QUOTE (IgorC @ Dec 15 2005, 03:09 PM)It was m...   Dec 15 2005, 15:21
||- - IgorC   QUOTE (guruboolez @ Dec 15 2005, 06:21 AM)It ...   Dec 15 2005, 15:26
||- - IgorC   I don't agree to use VBR at least for high a...   Dec 15 2005, 15:33
|||- - guruboolez   QUOTE (IgorC @ Dec 15 2005, 03:33 PM)I don...   Dec 15 2005, 16:27
|||- - IgorC   QUOTE (guruboolez @ Dec 15 2005, 07:27 AM)Hig...   Dec 15 2005, 17:36
||- - ATWindsor   QUOTE (IgorC @ Dec 15 2005, 06:26 AM)QUOTE (g...   Dec 15 2005, 16:15
||- - guruboolez   QUOTE (IgorC @ Dec 15 2005, 03:26 PM)QUOTE (g...   Dec 15 2005, 16:24
|- - JohnV   QUOTE (guruboolez @ Dec 15 2005, 05:01 PM)QUO...   Dec 15 2005, 15:21
|- - guruboolez   QUOTE (JohnV @ Dec 15 2005, 03:21 PM)QUOTE (g...   Dec 15 2005, 15:22
|- - JohnV   QUOTE (guruboolez @ Dec 15 2005, 05:22 PM)Doe...   Dec 15 2005, 15:24
- - Gabriel   QUOTE So why we should give a prior for mp3? If mp...   Dec 15 2005, 16:20
|- - JohnV   QUOTE (Gabriel @ Dec 15 2005, 06:20 PM)To me ...   Dec 16 2005, 14:41
- - NoXFeR   Gabriel: Is there any reason why streaming must be...   Dec 15 2005, 17:40
|- - Shade[ST]   QUOTE (NoXFeR @ Dec 15 2005, 10:40 AM)Gabriel...   Dec 15 2005, 17:45
- - NoXFeR   Well... Lowering the average bitrate of the VBR to...   Dec 15 2005, 17:57
|- - Shade[ST]   QUOTE (NoXFeR @ Dec 15 2005, 10:57 AM)Well......   Dec 15 2005, 18:36
- - NoXFeR   The bitrates were examples, modify them to the 48k...   Dec 15 2005, 18:51
|- - Shade[ST]   If a 40kbps VBR never goes above 48 kbps, it will ...   Dec 15 2005, 20:06
- - Ivan Dimkovic   One thing which worries me here - if the part of t...   Dec 16 2005, 12:28
|- - IgorC   Coding Technologies Enhanced aacPlus (HE-AAC v2) ?...   Dec 16 2005, 14:01
|- - guruboolez   QUOTE (Ivan Dimkovic @ Dec 16 2005, 12:28 PM)...   Dec 22 2005, 14:59
- - Ivan Dimkovic   I think Winamp was the only software actually bein...   Dec 16 2005, 14:41
|- - IgorC   QUOTE (Ivan Dimkovic @ Dec 16 2005, 05:41 AM)...   Dec 16 2005, 14:45
|- - benski   QUOTE (IgorC @ Dec 16 2005, 09:45 AM)QUOTE (I...   Dec 19 2005, 17:11
- - Ivan Dimkovic   Can more people confirm this? We would use best w...   Dec 16 2005, 15:07
- - Gabriel   QUOTE I think Winamp was the only software actuall...   Dec 16 2005, 15:18
|- - IgorC   Does helix support HE-AAC2?   Dec 16 2005, 15:19
|- - IgorC   More probably itīs the same updated audio encoder ...   Dec 16 2005, 15:29
|- - kurtnoise   QUOTE (IgorC @ Dec 16 2005, 03:19 PM)Does hel...   Dec 16 2005, 18:33
- - ckjnigel   I wish very much that someone who can translate Ja...   Dec 16 2005, 19:43
- - PatchWorKs   Hey, guyz... be calm. In my opinion "legal...   Dec 19 2005, 12:20
- - stephanV   That are not multiples of 128, that's just 2^n...   Dec 19 2005, 12:59
- - Ivan Dimkovic   Okay, we will use whatever has the latest CT encod...   Dec 19 2005, 18:14
|- - IgorC   Winamp 5.12 has a newer version than previos 5.11...   Dec 19 2005, 22:54
- - Ivan Dimkovic   Test is anyway scheduled for the late January, so ...   Dec 20 2005, 01:16
- - rjamorim   QUOTE (JohnV @ Dec 15 2005, 12:21 PM)One area...   Dec 20 2005, 01:34
|- - JohnV   QUOTE (rjamorim @ Dec 20 2005, 03:34 AM)Isn...   Dec 20 2005, 14:16
|- - rjamorim   QUOTE (JohnV @ Dec 20 2005, 11:16 AM)That...   Dec 21 2005, 00:32
- - digitalradiotech   A good reason to use CBR is that the new mobile br...   Dec 21 2005, 12:48
- - Ivan Dimkovic   Basically it all goes down to three use models: 1...   Dec 21 2005, 12:55
- - JohnV   One issue is also that in principle we should test...   Dec 21 2005, 20:32
|- - IgorC   QUOTE (JohnV @ Dec 21 2005, 11:32 AM)One issu...   Dec 22 2005, 14:51
- - Ivan Dimkovic   Well I know for sure that Nero could supply a vari...   Dec 22 2005, 15:01
- - Ivan Dimkovic   Actually a smart idea will be to do a small pre-te...   Dec 22 2005, 16:59
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: 21st April 2014 - 01:49