IPB

Welcome Guest ( Log In | Register )

Opus 1.1a BABYEATER build
NullC
post Feb 11 2013, 21:51
Post #1





Group: Developer
Posts: 200
Joined: 8-July 03
Member No.: 7653



Ever mindful of the world's population problems, the Opus development team is excited to present a new highly experimental build of Opus which is expected to eat 100% more babies than prior editions. It is also expected have (potentially) improved transient handling performance:

https://people.xiph.org/~greg/opus-tools_ba...ater-deluxe.zip (Edit: replaced with updated version, see below)

What this does is adapts the frame-size dynamically based on transient analysis prior to encoding. Leave the opus-tools framesize knob alone (in this test build it controls the amount of look-ahead used for the analysis, and 60ms is the default and only tested value).

It would be super-helpful to compare this with the regular 1.1 alpha at various rates and point us to any interesting results you find (e.g. where it does obviously worse or better, or where it catches fire and/or eats household pets instead of babies). Obviously you shouldn't be encoding your music archives with this stuff yet.

This post has been edited by NullC: Feb 13 2013, 01:07
Go to the top of the page
+Quote Post
 
Start new topic
Replies
Gainless
post Feb 12 2013, 12:52
Post #2





Group: Members
Posts: 172
Joined: 28-October 11
Member No.: 94764



At 192 kbps it does slightly better at this sample:
Sycho Active

The artifact at the 2nd kick is still pretty clear, but seems to be a resampling issue (SoX gives the same audible results)

With Muse Breaks it still gives the same old glitches, though, I think even a few more.
Attached File  Muse_Breaks_Rmx__Sample_.flac ( 1.13MB ) Number of downloads: 109


This post has been edited by Gainless: Feb 12 2013, 13:36
Go to the top of the page
+Quote Post
NullC
post Feb 13 2013, 01:16
Post #3





Group: Developer
Posts: 200
Joined: 8-July 03
Member No.: 7653



QUOTE (Gainless @ Feb 12 2013, 03:52) *
The artifact at the 2nd kick is still pretty clear, but seems to be a resampling issue (SoX gives the same audible results)


If you're getting audible issues with good resampling, then the problem is clippingó codec can't help that. Try opusdec -gain -6 to attenuate it a bit.

What rate setting are you hearing artifacts at for muse?

This post has been edited by NullC: Feb 13 2013, 01:52
Go to the top of the page
+Quote Post
Gainless
post Feb 13 2013, 21:54
Post #4





Group: Members
Posts: 172
Joined: 28-October 11
Member No.: 94764



QUOTE (jmvalin @ Feb 12 2013, 23:23) *
OK, I was able to identify at least one bug. See if this new build is any better.

Nope, sorry, still the same artifacts at 192 kbps for both samples.

QUOTE (NullC @ Feb 13 2013, 01:16) *
If you're getting audible issues with good resampling, then the problem is clippingó codec can't help that. Try opusdec -gain -6 to attenuate it a bit.

Have tried it didn't change anything. I've also done another SoX conversion with a gain reduced version of the sample, so there would be no clipping at all (at least according to the RG peak) and found that I could still ABX it. Clipping doesn't seem to be the problem, but maybe some flaw in general? I'm not that familiar with resampling, so I just used "best" quality, without touching the Passband and Phase response.

QUOTE (NullC @ Feb 13 2013, 01:16) *
What rate setting are you hearing artifacts at for muse?

192 kbps VBR. The glitches appear at some kicks, when some kind of cymbal like (a bit like sand paper) sound is faded in.

This post has been edited by Gainless: Feb 13 2013, 21:57
Go to the top of the page
+Quote Post
jmvalin
post Feb 13 2013, 22:17
Post #5


Xiph.org Speex developer


Group: Developer
Posts: 480
Joined: 21-August 02
Member No.: 3134



QUOTE (Gainless @ Feb 13 2013, 15:54) *
192 kbps VBR. The glitches appear at some kicks, when some kind of cymbal like (a bit like sand paper) sound is faded in.


Can you try all three versions (1.0.x, 1.1-alpha and babyeater) and tell us how they compare on that sample? Also, does the artefact eventually goes away as the bitrate increases?
Go to the top of the page
+Quote Post
Gainless
post Feb 14 2013, 13:35
Post #6





Group: Members
Posts: 172
Joined: 28-October 11
Member No.: 94764



QUOTE (jmvalin @ Feb 13 2013, 22:17) *
QUOTE (Gainless @ Feb 13 2013, 15:54) *
192 kbps VBR. The glitches appear at some kicks, when some kind of cymbal like (a bit like sand paper) sound is faded in.


Can you try all three versions (1.0.x, 1.1-alpha and babyeater) and tell us how they compare on that sample? Also, does the artefact eventually goes away as the bitrate increases?

Sure, done a little test with all three versions now (1.0.1, 1.1 alpha, BE deluxe). 1.0.1 performed better than the other two, pretty close to transparency at 192 kbps and not ABXable at 320 kbps, while 1.1 and BE couldn't achieve transparency even at that high bitrate.
Go to the top of the page
+Quote Post
jmvalin
post Feb 14 2013, 20:00
Post #7


Xiph.org Speex developer


Group: Developer
Posts: 480
Joined: 21-August 02
Member No.: 3134



QUOTE (Gainless @ Feb 14 2013, 07:35) *
Sure, done a little test with all three versions now (1.0.1, 1.1 alpha, BE deluxe). 1.0.1 performed better than the other two, pretty close to transparency at 192 kbps and not ABXable at 320 kbps, while 1.1 and BE couldn't achieve transparency even at that high bitrate.


That information is very useful. Could you try comparing just 1.0.x to 1.1 alpha, but in CBR mode? That will tell us whether the problem is rate related or somewhere else (e.g. transient detector). Also, it's not always possible, but if you were able to pinpoint the most obvious artefact with ms accuracy, it would give me a better chance to investigate.
Go to the top of the page
+Quote Post
Gainless
post Feb 15 2013, 20:42
Post #8





Group: Members
Posts: 172
Joined: 28-October 11
Member No.: 94764



QUOTE (jmvalin @ Feb 14 2013, 20:00) *
That information is very useful. Could you try comparing just 1.0.x to 1.1 alpha, but in CBR mode? That will tell us whether the problem is rate related or somewhere else (e.g. transient detector).

Ok, after a bit a bit more testing I can't really tell for sure which of them wins at CBR. 1.1 was the one I could still ABX at 320 kbps, so maybe 1.0.1 is a bit better, but it was pretty hard with both above 256 kbps.
QUOTE (jmvalin @ Feb 14 2013, 20:00) *
Also, it's not always possible, but if you were able to pinpoint the most obvious artefact with ms accuracy, it would give me a better chance to investigate.

It's around second 7,490. Visually I can't see any differences between the WAV and the Opus encodes, so the exact place might differ a few ms.

This post has been edited by Gainless: Feb 15 2013, 20:44
Go to the top of the page
+Quote Post
jmvalin
post Feb 16 2013, 04:58
Post #9


Xiph.org Speex developer


Group: Developer
Posts: 480
Joined: 21-August 02
Member No.: 3134



QUOTE (Gainless @ Feb 15 2013, 14:42) *
It's around second 7,490. Visually I can't see any differences between the WAV and the Opus encodes, so the exact place might differ a few ms.


Thanks, this is actually very helpful. There's a transient around there, but it doesn't seem to be detected. Can you try encoding with --framesize 5 or --framesize 2.5 and see if this particular artefact is still present (it may cause other artefacts)?
Go to the top of the page
+Quote Post
Gainless
post Feb 16 2013, 11:56
Post #10





Group: Members
Posts: 172
Joined: 28-October 11
Member No.: 94764



QUOTE (jmvalin @ Feb 16 2013, 04:58) *
QUOTE (Gainless @ Feb 15 2013, 14:42) *
It's around second 7,490. Visually I can't see any differences between the WAV and the Opus encodes, so the exact place might differ a few ms.


Thanks, this is actually very helpful. There's a transient around there, but it doesn't seem to be detected. Can you try encoding with --framesize 5 or --framesize 2.5 and see if this particular artefact is still present (it may cause other artefacts)?

blink.gif Switching to 5 ms framesize and VBR made the artifact almost totally disappear, now it's quite subtle at 128 kb/s and barely ABXable 192 kb/s anymore with both versions, pretty amazing.
Go to the top of the page
+Quote Post

Posts in this topic
- NullC   Opus 1.1a BABYEATER build   Feb 11 2013, 21:51
- - C.R.Helmrich   QUOTE (NullC @ Feb 11 2013, 21:51) What t...   Feb 11 2013, 22:57
|- - jmvalin   QUOTE (C.R.Helmrich @ Feb 11 2013, 16:57)...   Feb 11 2013, 23:24
|- - NullC   QUOTE (jmvalin @ Feb 11 2013, 14:24) (unl...   Feb 12 2013, 00:46
- - IgorC   eig sample has the issue. _http://depositfiles.or...   Feb 12 2013, 09:45
- - Gainless   At 192 kbps it does slightly better at this sample...   Feb 12 2013, 12:52
|- - NullC   QUOTE (Gainless @ Feb 12 2013, 03:52) The...   Feb 13 2013, 01:16
|- - Gainless   QUOTE (jmvalin @ Feb 12 2013, 23:23) OK, ...   Feb 13 2013, 21:54
|- - jmvalin   QUOTE (Gainless @ Feb 13 2013, 15:54) 192...   Feb 13 2013, 22:17
|- - Gainless   QUOTE (jmvalin @ Feb 13 2013, 22:17) QUOT...   Feb 14 2013, 13:35
|- - jmvalin   QUOTE (Gainless @ Feb 14 2013, 07:35) Sur...   Feb 14 2013, 20:00
|- - Gainless   QUOTE (jmvalin @ Feb 14 2013, 20:00) That...   Feb 15 2013, 20:42
|- - jmvalin   QUOTE (Gainless @ Feb 15 2013, 14:42) It...   Feb 16 2013, 04:58
|- - Gainless   QUOTE (jmvalin @ Feb 16 2013, 04:58) QUOT...   Feb 16 2013, 11:56
|- - jmvalin   QUOTE (Gainless @ Feb 16 2013, 05:56) QUO...   Feb 16 2013, 19:42
- - jmvalin   OK, I was able to identify at least one bug. See i...   Feb 12 2013, 23:23
- - IgorC   It's getting better http://pastebin.com/KGLCy...   Feb 12 2013, 23:51
- - Gainless   That's nice to hear, would be great if it coul...   Feb 16 2013, 20:40
|- - jmvalin   QUOTE (Gainless @ Feb 16 2013, 14:40) Tha...   Feb 18 2013, 06:43
- - Gainless   Sound both pretty transparent, thanks for the effo...   Feb 18 2013, 22:12
|- - jmvalin   QUOTE (Gainless @ Feb 18 2013, 16:12) Sou...   Feb 18 2013, 23:53
- - Omicron   Hello! I have some files: original wav 16khz u...   Feb 21 2013, 13:01
- - RobertM   QUOTE (Omicron @ Feb 21 2013, 22:01) Hell...   Feb 25 2013, 11:21
- - RobertM   I'm not a listening expert. However, I was int...   Feb 25 2013, 12:13
|- - RobertM   The samples encoded with the latest Opus version a...   Feb 26 2013, 06:55
- - Omicron   QUOTE what kind of problems did you hear? It soun...   Feb 26 2013, 23:10
- - RobertM   I think I've noticed a minor issue in the enco...   Feb 28 2013, 09:36
- - Gainless   Another sample making issues: DWK Sample A sweepin...   Apr 8 2013, 14:14
- - jensend   I can't reproduce your problem on mainline bui...   Apr 8 2013, 17:16
|- - jmvalin   QUOTE (jensend @ Apr 8 2013, 12:16) jmval...   Apr 8 2013, 18:20
||- - jensend   QUOTE (jmvalin @ Apr 8 2013, 11:20) Actua...   Apr 8 2013, 22:41
||- - jmvalin   QUOTE (jensend @ Apr 8 2013, 17:41) QUOTE...   Apr 9 2013, 03:21
|- - Gainless   QUOTE (jensend @ Apr 8 2013, 18:16) I can...   Apr 8 2013, 18:39
- - db1989   The above post has been moved from the associated ...   Apr 8 2013, 17:32
- - IgorC   The version 1.1a is the last testing build that wo...   Apr 8 2013, 19:05


Reply to this topicStart new topic
1 User(s) are reading this topic (1 Guests and 0 Anonymous Users)
0 Members:

 



RSS Lo-Fi Version Time is now: 17th September 2014 - 23:20