IETF Opus codec now ready for testing, That's CELT 0.11 Options

[jensend]

Is it worth increasing framesize at the default bitrate of 96kbps or will this have a negligible impact upon quality?
I am not sure whether 96kbps is considered to be a fairly low bitrate or not given that opus can go as low as 6kbps.

96kbps is definitely not a low bitrate in most opus contexts. The bitrate savings of the >20ms frames is about 0.4kbps, which won't be noticeable above 16kbps. The longer frames also save transport protocol/container overhead if you're streaming over the Internet or whatever; that can be a much bigger deal, but is irrelevant for opusenc.

Presumably what is considered a low bitrate for music would probably not be considered a low bitrate for speech?

Actually, in this instance it doesn't matter; stereo vs mono doesn't matter either. It's just a question of how 0.4kbps compares to your target bitrate.

Also, is there a Debian package of the Xiph ABX program Squishyball available to download anywhere? I would like to do some ABX tests with opus but, as I use Linux, foobar is not an option.

I don't see any Debian package, but it compiles from source quite easily and cleanly in my experience. You could also use the java version of ABC/HR from rarewares.

[Speckmade]

In this context, what is regarded as a low bitrate?

Bitrates, where overhead from transmission protocols gets relevant.

Is it worth increasing framesize at the default bitrate of 96kbps or will this have a negligible impact upon quality?

So: no.

[CoRoNe]

I've encoded 6_Channel_ID.wav with opus-tools-0.1.4-win32 but I've found nothing using BASSOPUS.dll which can properly play it back. And same goes with the release version of the DLL. However decoding back to WAV creates a working file.

I have no idea why they release it as bassopus.dll, because you have to rename it to bass_opus.dll for it to work.

But it seems to be a complete channels mayhem with 5.1 files or is it only me?

Strange indeed. Although the opus-tools should have multichannel support (don't know who doneish is), they spit out stereo only.

[kode54]

For it to even play streams normally in BASS, you need to load it with BASS_PluginLoad.

[NullC]

Although the opus-tools should have multichannel support (don't know who doneish is), they spit out stereo only.

Huh? Opus-tools has complete multichannel support, and multichannel files round trip fine. If you force it to crazy low bitrates it will downmix for you rather than produce crap, but it warns you when it does this.

This post has been edited by NullC: Aug 22 2012, 03:51

[mudlord]

There is no "perfect" antivirus. They are all "snake oil" you would trust until they fail. They all didn't find Flame early, rated it at most "suspicious" for years; and even worse are the over-optimistic false positive heuristic results which detect usual Windows kernel features as malware (e.g. Avira kept users from logging in not only once).

But well ... this is an Audio forum.

Excuse me, but Flame is a state sanctioned rootkit. Of course it won't be detected! Did you know some mobs pay AV companies to NOT detect thier crap?

[Atak_Snajpera]

I must admit that Opus at 64kbps with some magic tuning may finally reach mp3 CBR 128 (joint-stereo) in terms of quality.
Already Opus 64kbps sounds less annoying than mp3 CBR 128 (stereo) for me!

[LithosZA]

Document is in AUTH48 State

[CoRoNe]

Huh? Opus-tools has complete multichannel support, and multichannel files round trip fine. If you force it to crazy low bitrates it will downmix for you rather than produce crap, but it warns you when it does this.

CODE

Notice: Surround bitrate less than 32kbit/sec/channel, downmixing.

Ah, I see. I'm using Foobar, so I didn't know.

But it seems to be a complete channels mayhem with 5.1 files or is it only me?

I can now reproduce the issue with bass_opus.dll indeed. It seems the opus-plugin for Foobar doesn't have multichannel support at all.

[Garf]

It seems the opus-plugin for Foobar doesn't have multichannel support at all.

Of course it does. Please post in more detail what you're doing (wrong :-)).

[CoRoNe]

Ugh...darn dsp filters. Wouldn't accept multichannel. Sorry, my bad!

[CoRoNe]

I've encoded 6_Channel_ID.wav with opus-tools-0.1.4-win32 but I've found nothing using BASSOPUS.dll which can properly play it back.

BASSOPUS 2.4.0.1 should fix that problem.

[zerowalker]

Document is in AUTH48 State

What does that mean;O?
Is it soon approved?

[db1989]

From Google:
http://www.rfc-editor.org/pubprocess.html#auth48

[IgorC]

Have tried these two builds on transients samples at 64 kbps.
https://people.xiph.org/~greg/opus-tools_exp_tfsel5.zip (have called it as transients2)
https://people.xiph.org/~greg/opus-tools_exp_32024cb5.zip (transients1)


Velvet sample. T2 and T1 have the same bitrate and less or more the same quality.

CODE

ABC/HR for Java, Version 0.53a, 25 August 2012
Testname:

Tester: IgorC

1L = D:\Opus\opus_transients_TEST\Samples for transients2\01 Velvet\velvet_TRANSIENTS_1.wav
2R = D:\Opus\opus_transients_TEST\Samples for transients2\01 Velvet\velvet_TRANSIENTS_2.wav

Ratings on a scale from 1.0 to 5.0

---------------------------------------
General Comments:
---------------------------------------
1L File: D:\Opus\opus_transients_TEST\Samples for transients2\01 Velvet\velvet_TRANSIENTS_1.wav
1L Rating: 3.0
1L Comment:
---------------------------------------
2R File: D:\Opus\opus_transients_TEST\Samples for transients2\01 Velvet\velvet_TRANSIENTS_2.wav
2R Rating: 3.0
2R Comment:
---------------------------------------

ABX Results:

Fatboy. T2 - 92 kbps, T1 - 90 kbps.

CODE

ABC/HR for Java, Version 0.53a, 25 August 2012
Testname:

Tester: IgorC

1L = D:\Opus\opus_transients_TEST\Samples for transients2\02 Fatboy\fatboy_TRANSIENTS_2.wav
2R = D:\Opus\opus_transients_TEST\Samples for transients2\02 Fatboy\fatboy_TRANSIENTS_1.wav

Ratings on a scale from 1.0 to 5.0

---------------------------------------
General Comments:
---------------------------------------
1L File: D:\Opus\opus_transients_TEST\Samples for transients2\02 Fatboy\fatboy_TRANSIENTS_2.wav
1L Rating: 3.5
1L Comment:
---------------------------------------
2R File: D:\Opus\opus_transients_TEST\Samples for transients2\02 Fatboy\fatboy_TRANSIENTS_1.wav
2R Rating: 3.3
2R Comment: The strange clicking-like pulse during the first second.
---------------------------------------

ABX Results:
D:\Opus\opus_transients_TEST\Samples for transients2\02 Fatboy\fatboy_TRANSIENTS_2.wav vs D:\Opus\opus_transients_TEST\Samples for transients2\02 Fatboy\fatboy_TRANSIENTS_1.wav
5 out of 5, pval = 0.031


---- Detailed ABX results ----
D:\Opus\opus_transients_TEST\Samples for transients2\02 Fatboy\fatboy_TRANSIENTS_2.wav vs D:\Opus\opus_transients_TEST\Samples for transients2\02 Fatboy\fatboy_TRANSIENTS_1.wav
Playback Range: 00.000 to 29.206
8:20:04 PM p 1/1 pval = 0.5
8:20:06 PM p 2/2 pval = 0.25
8:20:08 PM p 3/3 pval = 0.125
8:20:14 PM p 4/4 pval = 0.062
8:20:16 PM p 5/5 pval = 0.031

EIG. T2 and T1 have the same bitrate.

CODE

ABC/HR for Java, Version 0.53a, 25 August 2012
Testname:

Tester: IgorC

1L = D:\Opus\opus_transients_TEST\Samples for transients2\03 EIG\eig_TRANSIENTS2.wav
2R = D:\Opus\opus_transients_TEST\Samples for transients2\03 EIG\eig_TRANSIENTS1.wav

Ratings on a scale from 1.0 to 5.0

---------------------------------------
General Comments:
---------------------------------------
1L File: D:\Opus\opus_transients_TEST\Samples for transients2\03 EIG\eig_TRANSIENTS2.wav
1L Rating: 3.0
1L Comment:
---------------------------------------
2R File: D:\Opus\opus_transients_TEST\Samples for transients2\03 EIG\eig_TRANSIENTS1.wav
2R Rating: 3.2
2R Comment: Less clicking.
---------------------------------------

ABX Results:
D:\Opus\opus_transients_TEST\Samples for transients2\03 EIG\eig_TRANSIENTS2.wav vs D:\Opus\opus_transients_TEST\Samples for transients2\03 EIG\eig_TRANSIENTS1.wav
5 out of 5, pval = 0.031


---- Detailed ABX results ----
D:\Opus\opus_transients_TEST\Samples for transients2\03 EIG\eig_TRANSIENTS2.wav vs D:\Opus\opus_transients_TEST\Samples for transients2\03 EIG\eig_TRANSIENTS1.wav
Playback Range: 00.000 to 15.000
8:58:18 PM p 1/1 pval = 0.5
8:58:23 PM p 2/2 pval = 0.25
8:58:30 PM p 3/3 pval = 0.125
8:58:35 PM p 4/4 pval = 0.062
8:58:40 PM p 5/5 pval = 0.031

T2 was better than T1 only on Fatboy sample while also increases bitrate for it a bit.

[Brazil2]

I've encoded 6_Channel_ID.wav with opus-tools-0.1.4-win32 but I've found nothing using BASSOPUS.dll which can properly play it back.

BASSOPUS 2.4.0.1 should fix that problem.

This version is now working fine, thanks for the info

[zerowalker]

I know i asked this before, but when is OPUS supposed to be released?
And what is it about the IEFT, is it approved or what?
Where can i check the status?

Thanks:)!

[LithosZA]

Where can i check the status?

http://tools.ietf.org/wg/codec/

[mamboman]

With the official release of opus imminent would be curious to know how we can anticipate future development.

Is future opus development likely to be mainly community driven (as Vorbis was with different tunings such as those of Aoyumi)?

Once opus is officially released will Xiph then concentrate their efforts upon another project (such as Ghost or some video codec work) or will official opus development continue to be priority for foreseeable future?

Also, being a bit speculative here but would be interesting to hear do folks believe that in time opus will be likely to supersede Vorbis as the open source lossy codec of choice for pre-recorded music?
I am sure Vorbis will be around for some time, but will Opus relegate it to being a legacy format as newer users favour opus?

[NullC]

There is a new opus 1.0.1 RC3 release candidate up, as well as a 0.1.5 opus-tools at http://www.opus-codec.org/downloads/ . These releases make minor build system changes and other small cleanups, other than some command-line options being changed a bit in opus-tools they are functionally identical to prior versions.

As always testing and trouble reports are appreciated.

[darkbyte]

I've encoded some C64 game music with Opus using the latest exp build with --bitrate 96 and they ended up as ~160kbps VBR files although the sources where mono. Is this a bug or just Opus likes SID music that much?

[LithosZA]

Weird, does CVBR do the same?

[Martel]

Did you use some low-pass with the C64 stuff? I know I tried some emulator once and it was producing way too much harmonics above 12kHz, it was practically unlistenable using monitoring headphones (unless you like your ears bleeding from the treble).

[jmvalin]

I've encoded some C64 game music with Opus using the latest exp build with --bitrate 96 and they ended up as ~160kbps VBR files although the sources where mono. Is this a bug or just Opus likes SID music that much?

Are you talking about a one-channel file or a two-channel file where both channels are the same. If it's the former, then the behaviour is probably normal considering that your C64 music is likely highly tonal.

[LigH]

The SID synthesizer is rather simple, the spectrum of frequencies probably very distinct. Similar to the results for "Stranglehold".