Skip to main content

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: New Public Listening Test (July-August 2011) (Read 79917 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

New Public Listening Test (July-August 2011)

Reply #25
I'm actually more interested on multiformat.


In case of multiformat there are next possibilties:

Vorbis AoTuV, AAC  (which one ?), CELT/Opus (?), MP3 (?)

Let's decide if test will be multiformat or AAC. Deadline was set until 1st of July

Right now Kennedyb4, Kohlrabi(?), /mnt are interested in AAC test.
Kwanbis and Zarggg - multiformat


Who else?

The list of codecs for ~100 kbps AAC test:
Code: [Select]
Candidates:
1. Nero
2. At least one Apple encoder (CVBR or TVBR) or two of them (?)
3. Fraunhofer
4. Winamp's Coding Technologies and Divx (Pre-Test)

As I understand it, transparency for the latest batch of codecs is believed to be somewhere between 128 kbit and 64 kbit, thus the proposed testing at 96 kbit.

As we try to resolve this further, we should also try to deal with the fact that outside scientific and developer circles, transparency at 128 kbit is not well appreciated. Is there some testing we can do that will bring a greater appreciation?

Many people base their disdain for lossy coding on the performance of early encoders. Can we, to show how far the art has progressed, compare some crusty but widely-used encoder at high bit rate to newer ones at lower bit rates?

Last Sebastian's test had the old l3enc MP3 encoder. It showed how bad can be this old mp3 encoder in comparison with the modern MP3 encoder.
http://listening-tests.hydrogenaudio.org/sebastian/

New Public Listening Test (July-August 2011)

Reply #26
I will strongly suggest to perform first LC-AAC test and only then go for Multiformat.
Because later it can turn into a lot of discussion of not right choice of AAC encoder for Multiformat test.
Although I am solely interested in the multiformat test, I have to agree with you here, IgorC. I'd say each audioformat in the multiformat test needs to be represented by the best encoder around.
Another aspect I'm really interested in; besides all ~96kbps samples, including some 128kbps samples encoded with an old mp3 encoder. I think this even more interests me than a battle between ~96kbps samples encoded with all sorts of state-of-the-art encoders. How cool would it be seeing a ~96kbps sample outperforming an "old" 128kbps sample!

New Public Listening Test (July-August 2011)

Reply #27
Hi. Sorry for my bad English.

In my opinion we should perform 96 kbps multiformat testing with next encoders:

Vorbis aoTuV 6.03
Nero AAC 1.5.4.0 (or maybe QT AAC, unfortunately I'm not sure too)
CELT, of course
LAME 3.98.4 (but something tells me that we should try FhG for this bitrate. at least for me it sounds better at 128 kbps)

Can we include more then one encoder per format into the test? Otherwise we must first of all determine the best encoder @96kbps for mp3 and AAC. I see no other ways.

And how about WMA (or even WMA 10 Pro)?
🇺🇦 Glory to Ukraine!

New Public Listening Test (July-August 2011)

Reply #28
Quote
How cool would it be seeing a ~96kbps sample outperforming an "old" 128kbps sample!

I cannot see a practical side in this...


BTW, Nero also has several modes: vbr, cbr, abr, abr 2pass...

New Public Listening Test (July-August 2011)

Reply #29
Let's decide if test will be multiformat or AAC. Deadline was set until 1st of July

Right now Kennedyb4, Kohlrabi(?), /mnt are interested in AAC test.
Kwanbis and Zarggg - multiformat


Who else?
I'm also only interested in a multiformat test.

New Public Listening Test (July-August 2011)

Reply #30
Can we include more then one encoder per format into the test? Otherwise we must first of all determine the best encoder @96kbps for mp3 and AAC. I see no other ways.

Yes, it's possbile to include 2 encoders of the same format but then we should exclude one of the formats. The max. number of codecs should be 4. The previous tests indicates that it's affordable number of codecs to test.


And how about WMA (or even WMA 10 Pro)

Statistics indicates that very very few persons who use it. http://www.hydrogenaudio.org/forums/index....showtopic=86830

I'm also only interested in a multiformat test.

ok, but I will ask people to specify also encoders for multiformat and/or AAC candidate tests.

It should be highlighted that if multiformat will be conducted first then AAC test will be in next year.
The only thing that changes is chronological order.

Current update:
Multiformat                                                                                                              - Zarggg, Kwanbis, dsimcha - (3)
AAC                                                                                                              - Kennedyb4, Kohlrabi, /mnt  - (3)
Multiformat / but realize that it will be difficult to choose best encoder per format  -  Corone, Steve Forte Rio - (2)

Who else?

New Public Listening Test (July-August 2011)

Reply #31
I do not like WMA at all, but i find it more interesting to test it than lets say CELT. I would do

MP3 LAME
AAC iTunes
Vorbis AoTuv
WMA


New Public Listening Test (July-August 2011)

Reply #32
I am for an AAC test.

New Public Listening Test (July-August 2011)

Reply #33
I am for an AAC test.

Me too. There a lot of news for AAC world. Last Nero, iTunes and new FhG. Damn, it would be great.
I've received preview version of FhG. it looks interesting.

What about Nero encoder? 1.5.4 is ok, right?

I do not like WMA at all, but i find it more interesting to test it than lets say CELT. I would do

MP3 LAME
AAC iTunes
Vorbis AoTuv
WMA


I guess it's WMA PRO 10. Right?


New Public Listening Test (July-August 2011)

Reply #35
If so, I'm for an AAC test too.

But what dou you think about competition between:

Nero AAC, QT AAC, LAME MP3, FhG MP3?

Thus we can find out the winner for mp3 and aac using the results of only one test. Or someone insist on iclusion other encoders to the AAC test?
🇺🇦 Glory to Ukraine!

New Public Listening Test (July-August 2011)

Reply #36
AAC test would be great.

If so, I'm for an AAC test too.

Great.

But what dou you think about competition between:

Nero AAC, QT AAC, LAME MP3, FhG MP3?

Thus we can find out the winner for mp3 and aac using the results of only one test. Or someone insist on iclusion other encoders to the AAC test?

Here I really don't know what to say. You can suggest this set of codecs and we will see if some other members will be interested to test these encoders.
But I think LAME and FhG encoder have been already compared between themselves in one of Sebastian's public test. Both were tied.

New Public Listening Test (July-August 2011)

Reply #37
So what encoder we should choose for our test - LAME or FhG? Сonsidering the results for 128 kbps test, I suppose it is LAME 3.98

And I'm still don't see the final solution about the list of formats (for future multiformat test) - will it be aac/celt/mp3/vorbis or aac/celt/vorbis/wma or some else combination?
🇺🇦 Glory to Ukraine!

New Public Listening Test (July-August 2011)

Reply #38
I vote for doing a AAC test only first, even though I currently don't use it.
Later on, multiformat would be fine.

Cheers
johnb

 

New Public Listening Test (July-August 2011)

Reply #39
So what encoder we should choose for our test - LAME or FhG?

And I'm still don't see the final solution about the list of formats (for future multiformat test) - will it be aac/celt/mp3/vorbis or aac/celt/vorbis/wma or some else combination?

Well, we are in the middle of suggestion period for set of codecs. People make suggestions and the next weekend all of us will analyse suggestions and will make final conclusion. Let's wait for suggestions of other members.

New Public Listening Test (July-August 2011)

Reply #40
Ok, then my final suggestion:

1) AAC Test:

Nero AAC
QT AAC
Fraunhofer
CT AAC (Winamp)

2) Multiformat test:

AAC (winner)
CELT
Vorbis aoTuV
WMA Pro

🇺🇦 Glory to Ukraine!

New Public Listening Test (July-August 2011)

Reply #41
How can a multi format test be considered complete without testing MP3 alongside the others?
lossyWAV -q X -a 4 -s h -A --feedback 2 --limit 15848 --scale 0.5 | FLAC -5 -e -p -b 512 -P=4096 -S- (having set foobar to output 24-bit PCM; scaling by 0.5 gives the ANS headroom to work)

New Public Listening Test (July-August 2011)

Reply #42
I'm also in favour of an (LC-)AAC test.  As on the AAC side, there are probably only 3 or 4 contenders for the crown, amongst which Nero and Apple, therefore I agree that it would be interesting to compare them to some vintage MP3 codec at 128k CBR, such as Fraunhofer's L3enc or MP3enc, as a 4th or 5th contender in the test.

I also still volunteer to host the test samples on my web server, as I did a couple of years ago for the 128k MP3 test.

New Public Listening Test (July-August 2011)

Reply #43
I think an LC-AAC test would nice and then we can use the winner in a mult-format test

For the multi-format test I would prefer constrained VBR or CBR at 96Kb/s for streaming purposes.

- CELT/OPUS
- Vorbis aoTuV
- LAME MP3
- Winner of the LC-AAC test

New Public Listening Test (July-August 2011)

Reply #44
I guess it's WMA PRO 10. Right?

Correct.

How can a multi format test be considered complete without testing MP3 alongside the others?

Exactly my thoughts.

For the multi-format test I would prefer constrained VBR or CBR at 96Kb/s for streaming purposes.

Do people actually care bout streaming quality that much? I tend to think that people that care about quality would not be listening to 96kbps music.

New Public Listening Test (July-August 2011)

Reply #45
I will include suggestion of developers as well. It won't be right just don't count them. Later we will see what should be done in this case.

Also if members have voted for Apple encoder I submit them to iTunes (CVBR) because until now TVBR produces very low bitrate for the test (~92 kbps).

Current votes:

Multiformat - Zarggg, Kwanbis, dsimcha, (3)
AAC - Kennedyb4, , /mnt, muaddib, Steve Forte Rio, IgorC, Pri3st, johnb, Polar, LithosZA - (8 +1 developer)
Multiformat / but realizes that it will be difficult to choose best encoder per format -  Corone (1)

Not sure what Kohlrabi wants.

Codecs:

MP3 LAME – Kwanbis,  Nick.C, LithosZA, IgorC (4)
Vorbis AoTuV – Kwanbis, LithosZA, IgorC, Steve Forte Rio (4)
CELT – Steve Forte Rio,  LithosZA, IgorC, (3)
WMA/WMA Pro – Kwanbis,  Steve Forte Rio,  (2)

iTunes AAC (constrained VBR) - Kennedyb4, Zarggg, Kwanbis, Steve Forte Rio,  IgorC, Polar (6)
Nero – Zarggg,  Steve Forte Rio,  IgorC, Polar (4)
FhG AAC - /mnt,  Steve Forte Rio, IgorC (3)
Winamp’s CT AAC – Steve Forte Rio, IgorC (2)
QuickTime AAC (true VBR) – Kennedyb4,  benski (1 + 1 developer against)
FAAC – Kohlrabi (1)

Old encoder with higher  bitrate – Notat,  Corone,  Polar,  (3)
To include only one Apple encoder in AAC test (CVBR or TVBR) – IgorC, Benski (2) (1 + 1 developer)

Who else?

New Public Listening Test (July-August 2011)

Reply #46
Quote
Also if members have voted for Apple encoder I submit them to iTunes (CVBR) because until now TVBR produces very low bitrate for the test (~92 kbps).


Hm, didn't know it. But in my opinion constraining of the bitrate isn't a good idea. Please, move my vote from CVBR to TVBR.
🇺🇦 Glory to Ukraine!

New Public Listening Test (July-August 2011)

Reply #47
Me multiformat including at least:
MP3 LAME
Nero
QuickTime AAC (true VBR)

and optionally, preferring Vorbis:
Vorbis AoTuV
iTunes AAC (constrained VBR)

New Public Listening Test (July-August 2011)

Reply #48
QuickTime AAC (true VBR) or iTunes AAC (constrained VBR), prefer TVBR
Nero
FhG AAC
Lame mp3