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: Winamp AAC on castanets (Read 5450 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Winamp AAC on castanets

I decided to have a play with the AAC encoder in Winamp and see how well it performs on castanets at different bitrates (128, 160, 192, 224 kbps).  I burnt the sample into a CD image and ripped using Winamp.  Then I used faad 2.1 to decode to wav.

It seems between 1.5 and 2.5 seconds, there is a large boost of HF in the right channel that makes ABXing very easy even up to 192 kbps.  At 224 kbps, it became harder though.

Code: [Select]
-------------------------------------
WinABX v0.23 test report
07/25/2004 11:30:01

A file: E:\aactest\castanets.wav
B file: E:\aactest\castanets128.wav

11:31:18    1/1  p=50.0%
11:31:27    2/2  p=25.0%
11:31:45    3/3  p=12.5%
11:31:52    4/4  p= 6.2%
11:32:00    5/5  p= 3.1%
11:32:07    6/6  p= 1.6%
11:32:20    7/7  p= 0.8%
11:32:26    8/8  p= 0.4%
11:32:33    9/9  p= 0.2%
11:32:38  10/10  p< 0.1%
11:32:44  11/11  p< 0.1%
11:32:51  12/12  p< 0.1%
11:33:00  test finished

-------------------------------------
WinABX v0.23 test report
07/25/2004 11:34:45

A file: E:\aactest\castanets.wav
B file: E:\aactest\castanets160.wav

11:35:11    1/1  p=50.0%
11:35:24    2/2  p=25.0%
11:35:29    3/3  p=12.5%
11:35:38    4/4  p= 6.2%
11:35:44    5/5  p= 3.1%
11:35:52    6/6  p= 1.6%
11:36:01    7/7  p= 0.8%
11:36:05    8/8  p= 0.4%
11:36:14    9/9  p= 0.2%
11:36:24  10/10  p< 0.1%
11:36:44  11/11  p< 0.1%
11:36:49  12/12  p< 0.1%
11:36:50  test finished

-------------------------------------
WinABX v0.23 test report
07/25/2004 11:36:52

A file: E:\aactest\castanets.wav
B file: E:\aactest\castanets192.wav

11:37:12    1/1  p=50.0%
11:37:17    2/2  p=25.0%
11:37:33    3/3  p=12.5%
11:37:42    4/4  p= 6.2%
11:37:47    5/5  p= 3.1%
11:37:55    6/6  p= 1.6%
11:38:02    7/7  p= 0.8%
11:38:07    8/8  p= 0.4%
11:38:14    9/9  p= 0.2%
11:38:22  10/10  p< 0.1%
11:38:27  11/11  p< 0.1%
11:38:30  12/12  p< 0.1%
11:38:32  test finished

-------------------------------------
WinABX v0.23 test report
07/25/2004 11:48:27

A file: E:\aactest\castanets.wav
B file: E:\aactest\castanets224.wav

11:48:55    1/1  p=50.0%
11:49:16    2/2  p=25.0%
11:49:44    2/3  p=50.0%
11:50:38    3/4  p=31.2%
11:51:12    4/5  p=18.8%
11:51:26    5/6  p=10.9%
11:51:32    6/7  p= 6.2%
11:51:37    7/8  p= 3.5%
11:51:57    8/9  p= 2.0%
11:52:39   9/10  p= 1.1%
11:52:50  10/11  p= 0.6%
11:53:57  11/12  p= 0.3%
11:53:59  test finished

Winamp AAC on castanets

Reply #1 –
That could be due to Winamps TNS implementation. Can you try with the Winamp decoder and see if the same thing happens?

Menno

Winamp AAC on castanets

Reply #2 –
I've performed in novmeber a multiformat listening test at ~192 kbps with pre-echo as object. Results were not published (I've used an unofficial release of Nero AAC encoder, which had some problems).
But the artifacts you've described for Winamp5 AAC encoder is something I remember very clearly. It was especially annoying and obvious with "micro-attacks samples": awe32_20, creaking and another one I've maybe never uploaded. I've checked with winamp decoder, and the problem is the same. The HF boost on attacks is also clearly visible in a sound editor.

I don't know if Winamp AAC encoder has changed since (mine was 1.00 I think). But I still have the .aac files on my laptop (I could upload them if needed).
Wavpack Hybrid -c4hx6

Winamp AAC on castanets

Reply #3 –
Oh I forgot to mention that the version of the AAC encoder in Winamp is 1.00 (came with Winamp 5.03).

I'll also try with the Winamp decoder when I have time.