AAC trouble until 350k (q0.85)
LAME problems -V presets.
Full Version: Emese sample
Is that supposed to be.... music ??
Anyway, nice sample.
Anyway, nice sample.
Really is this music? Who is the performer of this?
Its music to my ears with subliminal messages too.
QUOTE(shadowking @ Jun 4 2007, 17:38) 
Its music to my ears with subliminal messages too.
Ok. I respect that.
Can you please tell me who is the performer?
Thank you for your interesting sample. It's rare that such a serious AAC problem comes up. Easy to hear.
This one is a nice "short block trigger" test signal.
Can anyone hear problems with this sample when using Nero AAC Encoder at 192kbps ABR?
QUOTE(muaddib @ Jun 4 2007, 17:50)
Can you please tell me who is the performer?
French composer Pierre Henry. The composition's name is Le Temple d'Émèse which is a part of the « Fragment pour Artaud » work.
CD is still available in a nice and not too expensive 6CD box which may help anyone collecting weird audio samples :-)
N.B. I haven't tried this at 192 kbps.
N.B(2). Quick ABX test (my first one since 7 months, very easy...) - check the parameters (I'm not sure that I have the latest Nero's encoder)
CODE
foo_abx 1.3.1 report
foobar2000 v0.9.4.2
2007/07/18 01:36:18
File A: C:\temp\A03_emese.flac
File B: C:\temp\A03_emese.mp4
01:36:18 : Test started.
01:36:35 : 01/01 50.0%
01:36:38 : 02/02 25.0%
01:36:41 : 03/03 12.5%
01:36:44 : 04/04 6.3%
01:36:46 : 05/05 3.1%
01:36:49 : 06/06 1.6%
01:36:52 : 07/07 0.8%
01:36:54 : 08/08 0.4%
01:37:02 : 09/09 0.2%
01:37:09 : 10/10 0.1%
01:37:12 : 11/11 0.0%
01:37:17 : 12/12 0.0%
01:37:24 : 13/13 0.0%
01:37:27 : 14/14 0.0%
01:37:36 : 15/15 0.0%
01:37:39 : 16/16 0.0%
01:37:55 : Test finished.
----------
Total: 16/16 (0.0%)
Duration : 0:06.000 (264600 samples)
Sample Rate : 44100 Hz
Channels : 2
Bitrate : 205 kbps
Codec : AAC
Codec Profile : AAC LC
Encoding : lossy
Tool : Nero AAC codec / Feb 12 2007
The 'attacks' are noisy, buggy (weirds sounds) and highly smeared.
N.B(3). Confirmation of shadowking's test at ~350 kbps (VBR 0.85)
CODE
foo_abx 1.3.1 report
foobar2000 v0.9.4.2
2007/07/18 01:44:11
File A: C:\temp\A03_emese.flac
File B: C:\temp\A03_emese.mp4
01:44:11 : Test started.
01:44:45 : 01/01 50.0%
01:44:49 : 02/02 25.0%
01:44:54 : 02/03 50.0%
01:44:59 : 03/04 31.3%
01:45:06 : 04/05 18.8%
01:45:12 : 05/06 10.9%
01:45:15 : 06/07 6.3%
01:45:20 : 07/08 3.5%
01:45:24 : 08/09 2.0%
01:45:30 : 09/10 1.1%
01:45:45 : 10/11 0.6%
01:45:48 : 11/12 0.3%
01:45:56 : 12/13 0.2%
01:46:09 : 12/14 0.6%
01:46:14 : 13/15 0.4%
01:46:18 : 14/16 0.2%
01:46:24 : 15/17 0.1%
01:46:26 : 16/18 0.1%
01:46:29 : 17/19 0.0%
01:46:32 : 18/20 0.0%
01:46:36 : Test finished.
----------
Total: 18/20 (0.0%)
Duration : 0:06.000 (264600 samples)
Sample Rate : 44100 Hz
Channels : 2
Bitrate : 361 kbps
Codec : AAC
Codec Profile : AAC LC
Encoding : lossy
Tool : Nero AAC codec / Feb 12 2007
Sorry, Guruboolez, I didn't catch which settings did you use for the first ABX test?
Thank you very much for information about author!
Thank you very much for information about author!
Nero AAC codec / Feb 12 2007
ABR (1 pass) 192 kbps (with foobar2000's GUI)
I wasn't simply sure that I got the latest Nero's version (and I was a bit tired for googling yesterday).
ABR (1 pass) 192 kbps (with foobar2000's GUI)
I wasn't simply sure that I got the latest Nero's version (and I was a bit tired for googling yesterday).
Yes, that is the latest Nero version. Thank you for testing.
I can hear the problems at abr 192, but my ears are not good enough for -q 0.85.
We will check what causes bad quality.
I can hear the problems at abr 192, but my ears are not good enough for -q 0.85.
We will check what causes bad quality.
I just pushed the encoder to its maximum (-q1.00, 406 kbps). The problem is slightly perceptible to my ears :
foo_abx 1.3.1 report
foobar2000 v0.9.4.2
2007/07/18 12:31:17
File A: C:\temp\A03_emese.flac
File B: C:\temp\A03_emese.mp4
12:31:17 : Test started.
12:31:40 : 01/01 50.0%
12:31:49 : 01/02 75.0%
12:31:53 : 02/03 50.0%
12:32:01 : 03/04 31.3%
12:32:03 : 04/05 18.8%
12:32:14 : 05/06 10.9%
12:32:23 : 06/07 6.3%
12:32:50 : 07/08 3.5%
12:32:56 : 08/09 2.0%
12:33:08 : 09/10 1.1%
12:33:13 : 10/11 0.6%
12:33:32 : 11/12 0.3%
12:33:39 : Test finished.
----------
Total: 11/12 (0.3%)
CODE
foo_abx 1.3.1 report
foobar2000 v0.9.4.2
2007/07/18 12:31:17
File A: C:\temp\A03_emese.flac
File B: C:\temp\A03_emese.mp4
12:31:17 : Test started.
12:31:40 : 01/01 50.0%
12:31:49 : 01/02 75.0%
12:31:53 : 02/03 50.0%
12:32:01 : 03/04 31.3%
12:32:03 : 04/05 18.8%
12:32:14 : 05/06 10.9%
12:32:23 : 06/07 6.3%
12:32:50 : 07/08 3.5%
12:32:56 : 08/09 2.0%
12:33:08 : 09/10 1.1%
12:33:13 : 10/11 0.6%
12:33:32 : 11/12 0.3%
12:33:39 : Test finished.
----------
Total: 11/12 (0.3%)
Did anybody test this sample with other encoders (mp3,ogg,mp4) at ~192kbps?
Pretty easy with MP3 Lame 3.97 @ V0. A most interesting sample.
edit: With Nero AAC codec / Aug 6 2007 I'm finding it hard to ABX at Q0.8 under the same conditions as Lame. Q0.7 was ABXable but an improvement on Lame V0 for me.
CODE
foo_abx 1.3.1 report
foobar2000 v0.9.5 beta 7
2007/12/19 15:23:47
File A: \test\A03_emese.flac
File B: \test\V0. A03_emese.mp3
15:23:47 : Test started.
15:24:10 : 01/01 50.0%
15:24:19 : 02/02 25.0%
15:24:28 : 03/03 12.5%
15:24:34 : 04/04 6.3%
15:24:40 : 05/05 3.1%
15:24:46 : 06/06 1.6%
15:24:55 : 07/07 0.8%
15:25:03 : 08/08 0.4%
15:25:17 : 09/09 0.2%
15:25:26 : 10/10 0.1%
15:25:27 : Test finished.
----------
Total: 10/10 (0.1%)
foobar2000 v0.9.5 beta 7
2007/12/19 15:23:47
File A: \test\A03_emese.flac
File B: \test\V0. A03_emese.mp3
15:23:47 : Test started.
15:24:10 : 01/01 50.0%
15:24:19 : 02/02 25.0%
15:24:28 : 03/03 12.5%
15:24:34 : 04/04 6.3%
15:24:40 : 05/05 3.1%
15:24:46 : 06/06 1.6%
15:24:55 : 07/07 0.8%
15:25:03 : 08/08 0.4%
15:25:17 : 09/09 0.2%
15:25:26 : 10/10 0.1%
15:25:27 : Test finished.
----------
Total: 10/10 (0.1%)
edit: With Nero AAC codec / Aug 6 2007 I'm finding it hard to ABX at Q0.8 under the same conditions as Lame. Q0.7 was ABXable but an improvement on Lame V0 for me.
This is a "lo-fi" version of our main content. To view the full version with more information, formatting and images, please click here.