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: Problematic sample (ABX'ble @ 192kbps), audibly reduced quality fr (Read 2794 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Problematic sample (ABX'ble @ 192kbps), audibly reduced quality fr

Code: [Select]
foo_abx 1.3.4 report
foobar2000 v1.2.9
2015/01/19 03:09:21

File A: H:\Workings\artifact\TransSiberian.flac
File B: H:\Workings\artifact\TransSiberian128.opus

03:09:21 : Test started.
03:09:47 : 01/01  50.0%
03:10:05 : 02/02  25.0%
03:10:17 : 03/03  12.5%
03:10:29 : 04/04  6.3%
03:10:59 : 05/05  3.1%
03:11:19 : 06/06  1.6%
03:11:32 : 07/07  0.8%
03:11:41 : 08/08  0.4%
03:11:48 : Test finished.

 ---------- 
Total: 8/8 (0.4%)


Code: [Select]
foo_abx 1.3.4 report
foobar2000 v1.2.9
2015/01/19 03:13:16

File A: H:\Workings\artifact\TransSiberian.flac
File B: H:\Workings\artifact\TransSiberian192.opus

03:13:16 : Test started.
03:15:54 : 01/01  50.0%
03:16:26 : 02/02  25.0%
03:17:39 : 03/03  12.5%
03:18:57 : 04/04  6.3%
03:19:43 : 05/05  3.1%
03:20:20 : 06/06  1.6%
03:20:51 : 06/07  6.3%
03:23:16 : 07/08  3.5%
03:28:19 : 08/09  2.0%
03:32:32 : 09/10  1.1%
03:32:39 : Test finished.

 ---------- 
Total: 9/10 (1.1%)


Encoding log shows the codec is already working overtime (+20~% from nominal bitrate)

Problematic sample (ABX'ble @ 192kbps), audibly reduced quality fr

Reply #1 –
To me, 96k seems slightly different from the source. 128k one is transparent.

OT:
There is a problem of ABXing the 44.1/16k source and the resulting opus file.
(see http://www.hydrogenaud.io/forums/index.php?showtopic=108133)

I succeeded in ABXing with the trick.

Code: [Select]
foo_abx 2.0 report
foobar2000 v1.3.7
2015-01-19 00:47:53

File A: TransSiberian.flac
SHA1: 42adabb2669831be3fdff833e934a01c50467b13
Gain adjustment: -4.92 dB
File B: TransSiberian128.opus
SHA1: 3cdcff96bc5aa772d1c9e95a045783ab7df42833
Gain adjustment: -4.91 dB

Output:
DS : 스피커(Envy24 Family Audio Controller WDM)
Crossfading: NO

00:47:53 : Test started.
00:47:57 : 01/01
00:48:01 : 02/02
00:48:05 : 03/03
00:48:09 : 04/04
00:48:12 : 05/05
00:48:17 : 06/06
00:48:33 : 07/07
00:48:38 : 08/08
00:48:48 : 09/09
00:48:53 : 10/10
00:48:56 : 11/11
00:48:59 : 12/12
00:49:02 : 13/13
00:49:06 : 14/14
00:49:09 : 15/15
00:49:13 : 16/16
00:49:13 : Test finished.

 ---------- 
Total: 16/16
Probability that you were guessing: 0.0%

 -- signature -- 
b8e90d8d915b19a7adf93134d7a9eccfd98f1233


Maybe we need to clarify ABX conditions (windows samplerate, use of DSP, etc.) when comparing a 44.1k source and the opus file...