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: Opus 1.2 is out! (Read 120788 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Re: Opus 1.2 is out!

Reply #75
Well, AAC technologies aren't exactly smooth either between one another.
The thing is: AAC-LC and HE-AAC codes individual stereo channels, or higher order channels if necessary.
HE-AACv2 uses parametric stereo, which is a like spectral band replication, only instead of frequency bands, it works on the other channel to "fake" a second channel.
xHE-AAC goes even further and adds USAC using linear prediction.

So technically you'd need an individual line for either of those, and they'd criss-cross over each other. I'm not sure where this becomes too much and the complexity a little bit overwhelming. After all, this is just a diagram based on estimates.

Re: Opus 1.2 is out!

Reply #76
I think a better graph would be a graph that nearly looks the same, but with the 'narrowband, wideband etc.' sections removed and the color of the line determines the bandwidth.
The narrowband/wideband/fullband lines aren't "sections" they're simply vague indications of the quality of uncompressed audio of that bandwidth. So you can have superwideband audio with equal or worse quality than the "uncompressed wideband" line. Also, keep in mind that this figure is meant to give a rough idea of the quality. One shouldn't read too much out of it, especially considering that we're mixing in speech, music, mono stereo, and 5 different bandwidths. There's a reason the original one was drawn by hand on a piece of paper.

Re: Opus 1.2 is out!

Reply #77
Also, keep in mind that this figure is meant to give a rough idea of the quality. One shouldn't read too much out of it, especially considering that we're mixing in speech, music, mono stereo, and 5 different bandwidths.
My thought is the same.

Re: Opus 1.2 is out!

Reply #78
It's quite possible to come up with a graph that tries to be more exact rather than heuristic. igorc's graph from 2014 would be more of a starting point for such an endeavor. It'd need to be updated for Opus 1.2, and it would benefit from more listening test evidence.

Because some codecs' performance depends so much on the difference, one really would need at least two separate graphs - one for fullband stereo music and one for speech.

Re: Opus 1.2 is out!

Reply #79
Um, it will be great to have a graph indicating quality (mean opinion scores MOS) of different codecs at least in range of 8-128 kbps.  Not sure if it's feasible for bitrates higher than 128 kbps.
And create a wiki.hydrogenaudio page for that purpose.


Re: Opus 1.2 is out!

Reply #80
What if instead drawing fictional lines we would organize public listening test? Previous tests were done at 64 and 96 kbps. So possibilities are 128k or 32-48k test.

All 3 codecs Vorbis/AAC/Opus would be on par at 128 kbps  with an excellent scores about ~4.7-4.8 approaching transparency. It might be interesting but in pretty limited cases if You ask me.

So we end with ~32-48 kbps. HE-AACv2,v1 and Opus are the only serious competitors at those rates (Vorbis and MP3 won't be competitive at such low rates. still can be included for a full picture).

Opus 1.2.1 already does very good job and it's to suppose that 1.3 will be even better at those rates.
It's a possibility for low rate test in 2018. What do you think?

Re: Opus 1.2 is out!

Reply #81
Can anyone advise on updating opus 1.2.1 on Ubuntu 16.04 LTS?

The repos have libopus 1.1.2 and I had installed the "opus-tools" package previously

I have opus-1.2.1.tar.gz and compiled and installed.  Great

I did ./configure, then make, then sudo checkinstall, and finally worked out I needed sudo ldconfig

opusenc --version says "opusenc opus-tools 0.1.9 (using libopus 1.2.1)"

I am having problems with opus-tools-0.1.10.tar.gz. It wants "ogg >=1.3" which I appear to have (libogg0 1.3.2-1).  And do I need to update opus tools anyway? It's a little confusing since at http://opus-codec.org/downloads/ opus-tools is listed under development whilst libopus is described as stable.  So what is the stable version of opus-tools?

And I have my new package "opus" now, but I still need "libopus0" installed?

Re: Opus 1.2 is out!

Reply #82
Disregard the above.  Needed libogg-dev.  Silly me :)

Re: Opus 1.2 is out!

Reply #83
What if instead drawing fictional lines we would organize public listening test? Previous tests were done at 64 and 96 kbps. So possibilities are 128k or 32-48k test.

All 3 codecs Vorbis/AAC/Opus would be on par at 128 kbps  with an excellent scores about ~4.7-4.8 approaching transparency. It might be interesting but in pretty limited cases if You ask me.

So we end with ~32-48 kbps. HE-AACv2,v1 and Opus are the only serious competitors at those rates (Vorbis and MP3 won't be competitive at such low rates. still can be included for a full picture).

Opus 1.2.1 already does very good job and it's to suppose that 1.3 will be even better at those rates.
It's a possibility for low rate test in 2018. What do you think?

Hear, hear!

Re: Opus 1.2 is out!

Reply #84
  • Support for all of the fixes in draft-ietf-codec-opus-update-06 (the mono downmix and the folding fixes need --enable-update-draft)

Thanks. How was this solved? How can I use it in LameXP or any other GUI? How much does it decrease quality over no-fix, and at what bitrate does this version achieve transparency with the fix?

MP3 is the worst lossy codec ever released, so move to Opus ;

This is just wrong. VQF, WMA, Qdesign... were far worse, and when you consider how old mp3 is, the fact it can compete with aac and ogg as long as you use VRB LAME encoding it is amazing actually.