New Public Multiformat Listening Test (Jan 2014), Discussion

[kamedo2]

IgorC, could you please prepare the sample02, I'll redo the graph once more, using sample other than 'Waiting'.

[IgorC]

oops, I will be busy in these days and will dedicate time to a test by weekends.

[darkbyte]

We've been stuck a little, didn't we?

[kennedyb4]

I have not read the entire thread so please forgive me if this is commonly known.

Someone over on H/A is reporting a change in the latest Itunes encoder and has posted a chart indicating substantial bitrate changes from the older encoder to the newer.

If it will affect the test or a discussion of which encoder to use is required, this is why I posted.

[kamedo2]

kennedyb4,

Thank you for telling the important info. I'll update the iTunes and check.

[Gainless]

Well, Igor hasn't been online for 2 weeks now. It's getting a bit creepy, I hope he is just busy with something.

[win32vb]

    If it helps anybody, IgorR1 was last seen on #opus on January 22 according to my IRC client.  I will add that it is normal for him to come and go on there though.

[kamedo2]

I tested the latest iTunes. The bitrate increased by about 0.1kbps. The Apple CoreAudio still doesn't allow fine-tuning of bitrates.

[kennedyb4]

Sorry if I wasted your time. The Bit rate differences rwported were much larger but I can't find the thread.

[IgorC]

Organization of the test is delayed. There were some unexpected events on my side which I should pay attention now.

I'm sorry. Will be back at first opportunity.

[kamedo2]

I'm preparing for the restart. Please wait few more days.

[kamedo2]

The three major codec to test: qaac, aoTuV, Opus.

qaac_2.35\qaac --cvbr 96 -o %o %i
# rate(kbps) title
1 103 SinceAlways
2 106 Waiting
3 114 velvet
4 101 trumpet
5 101 girl
6 99 Can't Wait Until Tonight (Dry Wurlitzer Mix)
7 85 35_SQAM_glockenspiel_cut
8 108 Robots_old

9 109 Asleep__4.11-4.31_
10 100 Greatest_Love_of_All_2min57
11 102 Hey Tonight
12 100 Severance__1.31-1.51_
13 109 Shinsho_pool_3min45_4min4
14 100 SlavesOfFear
15 109 The Chastising of Renegade

16 99 TrosYGareg
17 95 4-Sound-English-male.441
18 104 9-Have-big-expensive-car.441
19 95 12-German-male-speech.441
20 99 15-Good-evening.441
21 105 21-classic.441
22 101 24-Greensleeves-Korean-male-speech.441
23 99 25-This-is-the-end.441

24 105 27-last-song-drums-and-trampets.441
25 106 bonhemian_rhapsody
26 107 clapton_44k
27 107 Coral
28 101 ExitMusic
29 116 liberate
30 103 NewYorkCity
31 99 sandman

32 109 take_your_finger_frin_my_head
33 106 Changes
34 111 Girl_In_The_Fire__Sample_
35 110 Hotel California
36 103 Jupiter, the Bringer of Jolity
37 110 Last_Of_The_Mohicanz__Sample_
38 110 Only Time
39 99 Through The Fire And Flames
40 103 With Love (Outro)
average: 103.7kbps


aoTuV: venc603 -q2.2 %i %o
# rate(kbps) title
1 97 SinceAlways
2 107 Waiting
3 124 velvet
4 101 trumpet
5 134 girl
6 111 Can't Wait Until Tonight (Dry Wurlitzer Mix)
7 79 35_SQAM_glockenspiel_cut
8 105 Robots_old

9 116 Asleep__4.11-4.31_
10 89 Greatest_Love_of_All_2min57
11 96 Hey Tonight
12 94 Severance__1.31-1.51_
13 88 Shinsho_pool_3min45_4min4
14 107 SlavesOfFear
15 122 The Chastising of Renegade
16 96 TrosYGareg

17 99 4-Sound-English-male.441
18 113 9-Have-big-expensive-car.441
19 104 12-German-male-speech.441
20 100 15-Good-evening.441
21 89 21-classic.441
22 121 24-Greensleeves-Korean-male-speech.441
23 95 25-This-is-the-end.441
24 112 27-last-song-drums-and-trampets.441

25 94 bonhemian_rhapsody
26 122 clapton_44k
27 102 Coral
28 85 ExitMusic
29 143 liberate
30 102 NewYorkCity
31 106 sandman
32 133 take_your_finger_frin_my_head

33 119 Changes
34 108 Girl_In_The_Fire__Sample_
35 112 Hotel California
36 89 Jupiter, the Bringer of Jolity
37 114 Last_Of_The_Mohicanz__Sample_
38 100 Only Time
39 117 Through The Fire And Flames
40 99 With Love (Outro)
average: 106.1kbps


opus-tools-0.1.8-win64_SSE2\opusenc --bitrate 96 %i %o
# rate(kbps) title
1 82 SinceAlways
2 110 Waiting
3 112 velvet
4 132 trumpet
5 122 girl
6 116 Can't Wait Until Tonight (Dry Wurlitzer Mix)
7 127 35_SQAM_glockenspiel_cut
8 100 Robots_old

9 103 Asleep__4.11-4.31_
10 103 Greatest_Love_of_All_2min57
11 92 Hey Tonight
12 119 Severance__1.31-1.51_
13 104 Shinsho_pool_3min45_4min4
14 90 SlavesOfFear
15 104 The Chastising of Renegade
16 124 TrosYGareg

17 95 4-Sound-English-male.441
18 119 9-Have-big-expensive-car.441
19 99 12-German-male-speech.441
20 94 15-Good-evening.441
21 112 21-classic.441
22 132 24-Greensleeves-Korean-male-speech.441
23 98 25-This-is-the-end.441
24 104 27-last-song-drums-and-trampets.441

25 101 bonhemian_rhapsody
26 104 clapton_44k
27 104 Coral
28 111 ExitMusic
29 97 liberate
30 93 NewYorkCity
31 97 sandman
32 114 take_your_finger_frin_my_head

33 112 Changes
34 121 Girl_In_The_Fire__Sample_
35 106 Hotel California
36 104 Jupiter, the Bringer of Jolity
37 104 Last_Of_The_Mohicanz__Sample_
38 96 Only Time
39 97 Through The Fire And Flames
40 103 With Love (Outro)
average: 106.4kbps

[kamedo2]

[IgorC]

Kamedo2 is the main organizer of the test. It will be better for all.

[kamedo2]

Thank you, IgorC. From now, I'm going to organize the public listening test.

The current status is:
Agenda (can be changed)
A choice of codecs. December 8 – December 18 DONE
Bitrate verification, a choice of settings – December 19 – December 23-25. DONE
Sample selection - December 25-26 – January 8. DONE Selection
Checking all conditions, preparations, dummy packages  


We're very close to the goal. I'll prepare the prototype package, and do the final check of ABC/HR's normalization within several days.

What process should be used? I like this chain, but feel free to post your own opinion.

1. References Lossless
source 44.1kHz/16bits -> SoX 48kHz/16bits

2. AAC/MP3/Vorbis
source 44.1 kHz/16 bits -> lossy encoder -> lossy decoder .wav 44.1kHz/16 bits -> SoX 48kHz/16bits

3. Opus
source 44.1 kHz/16 bits -> Opus (internal 48 kHz) -> lossy decoder .wav 48kHz/16 bits