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: xHE-AAC : The Death of OPUS? (Read 11889 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Re: xHE-AAC : The Death of OPUS?

Reply #25
I'd argue that AAC might actually be the most used lossy codec.  But MP3 certainlly has the biggest mindshare among the general population.

Re: xHE-AAC : The Death of OPUS?

Reply #26
Well, after a year of video and audio conferencing I would not be so sure.

Many VOIP and video conferencing solutions use Opus. And things like Discord, Teamspeak, the in game voice chats in games/launchers and I bet the Xbox/PSX voice chat, too. Then there are Browsers, Youtube, Whatsapp.

And if I understood correctly, the new Bluetooth LC3 codec borrowed quite a bit of Opus, too.



Re: xHE-AAC : The Death of OPUS?

Reply #27
could be as well, I'm just pretty sure MP3 is behind AAC (And I suspect Opus as well)

Re: xHE-AAC : The Death of OPUS?

Reply #28
I aggre with binaryhermit. AAC is probably the most used lossy codec atm.
Apple. Self explanatory.
All smartphones are recording videos using AAC. We are talking about billions of users.
All YouTube/Twitch streamers are using AAC.
YouTube is not using H264+AAC but it still has copy of them on the server.
I am using youtube-dl to download videos directly from YouTube and I am using H264+AAC for maximum compatibility.
All pirated movies are using H264+AAC.
Just for fun I checked a well known XXX website that is used by millions of people. Guess what, AAC again.

Re: xHE-AAC : The Death of OPUS?

Reply #29
Yes,
Add to it that  during several years  even most inexpensive cars come with bluetooth today so people can use Spotify/Tidal/Apple Music.
 So, considering popularity of AAC as described in previous post, MP3 isn't predominant format anymore.




Re: xHE-AAC : The Death of OPUS?

Reply #30
...
All YouTube/Twitch streamers are using AAC.
YouTube is not using H264+AAC but it still has copy of them on the server.
I am using youtube-dl to download videos directly from YouTube and I am using H264+AAC for maximum compatibility.
...
YouTube usually has Opus too, which from a frequency spectrum perspective, provides better audio (can you hear the difference though?). And if it's just the audio that I'm after, then I'll go with it.

But yes, from a compatibility side, then aac in m4a or mp4 is the better option.

Re: xHE-AAC : The Death of OPUS?

Reply #31
"predominant format" is a matter of definition really. Are we counting files/streams in use or are we counting files/streams being encoded or what?

* Even if everyone stopped encoding to MP3, it would be a giant format - there are tons of files that will live (at least among those of use who avoid transcoding lossies). Same with AAC, but to a lesser degree.
* There are tons lots of mp2/AC3 over broadcast. They aren't just gonna disappear even if they technically could.
* According to https://en.wikipedia.org/wiki/Technical_details_of_Netflix , Netflix delivers AAC but can also deliver Ogg Vorbis to Android devices. (Disregarding deprecated formats.)
* Sure YouTube encodes to Opus, but what devices accept what? By the way, I just checked a few old AND new videos, and it seems that YouTube now delivers Opus/AAC both at around 130, and two lower-bitrate Opus options. I wonder how present-day Opus at 130 compares with oooold AAC at 192. (Yes AAC at 192 should normally be transparent, but did YouTube's AAC encode efficiently or did it try to save CPU time?)
* What the covid-hell is Zoom using? (I'm pretty sure they haven't reinvented anything big, why should they?)
High Voltage socket-nose-avatar


Re: xHE-AAC : The Death of OPUS?

Reply #33
* Sure YouTube encodes to Opus, but what devices accept what?
https://caniuse.com/?search=opus
https://caniuse.com/?search=aac
Means that if your browser is Chrome/Firefox/Edge  audio is Opus

I wonder how present-day Opus at 130 compares with oooold AAC at 192.
Ohm, that's a pitty we don't have any test with Opus and AAC . But wait... we actually have, have we?  :-X
Anyway for 99.999...% Opus 130-160 is tranparent.  Because  96 kbps was already very hard https://listening-test.coresv.net/results.htm

oh and another one https://hydrogenaud.io/index.php?topic=120166.0

As you can see there is no point to even argue about quality for 128+ kbps area....  Ah yes, I know that some of us hear can do high bitrate but in reality it's transparent for big mass.

Re: xHE-AAC : The Death of OPUS?

Reply #34
Doesn't youtube use FAAC which is apparently really horrible?

Re: xHE-AAC : The Death of OPUS?

Reply #35
Doesn't youtube use FAAC which is apparently really horrible?

FhG, CBR, 128 kbit/s.

I uploaded 30 second clip (with lossless audio) on YouTube so I can compare it.
First picture is from YouTube.
Second picture is fhgaacenc.exe.
Pretty close.

Re: xHE-AAC : The Death of OPUS?

Reply #36
I uploaded 30 second clip (with lossless audio) on YouTube so I can compare it.
First picture is from YouTube.
Second picture is fhgaacenc.exe.
Pretty close.


I like your signature. 


Re: xHE-AAC : The Death of OPUS?

Reply #37
I wonder how present-day Opus at 130 compares with oooold AAC at 192.
Ohm, that's a pitty we don't have any test with Opus and AAC . But wait... we actually have, have we?  :-X

There was a word there.
And a context.
YouTube deprecated its 192 kbps AAC format. Was that a reasonable decision from the encoding has improved, we can reduce bitrate argument, or would that just be that they saw 192 as useless-in-the-first-place even without improving the AAC@128 / Opus@130 alternatives?
(That said, YouTube is big enough to negotiate license deals, so they might have had a "per encoded minute" and then for that reason dropped all AAC formats but one. At least it seems that there is just one non-Opus stream available for fresh uploads.)
High Voltage socket-nose-avatar

Re: xHE-AAC : The Death of OPUS?

Reply #38
I suspect that the future of lossy is "most nobody will even care what format their audio is in as long as they can play it".
Yes, because end users don't encode the files themselves, the services, protocol etc. do this.

Re: xHE-AAC : The Death of OPUS?

Reply #39

* According to https://en.wikipedia.org/wiki/Technical_details_of_Netflix , Netflix delivers AAC but can also deliver Ogg Vorbis to Android devices.

Netflix recently announced that they're rolling out xHE-AAC on Android and also taking advantage of it's ability to handle MPEG-D DRC metadata to  do "loudness management".

https://netflixtechblog.com/optimizing-the-aural-experience-on-android-devices-with-xhe-aac-c27714292a33?gi=244064d0be08

Back to the topic of other places that AAC audio is found, it's absolutely massive - in the UK alone we have the entirety of the BBC's online output, both "iPlayer" (11 TV Channels) and "Sounds" (~56 Local and National Radio stations (most of which have 4 different quality HLS streams (48kbps & 96kbps HE-AAC available globally plus 128kbps & 320kbps AAC-LC available in the UK only)) though I think some podcasts are still MP3. Also the other major broadcasters that provide streams or catch-up services: these days, certainly: Channel 4, Channel 5,  ITVs collection ,  Britbox,  Hulu, Amazon Prime (though I think they still offer MP3s for music), Disney+,  and even more Internationally.

But yes, as binary hermit said up-thread "MP3  still has the mindshare", it's the household name of {lossy) audio, and alas most normies have no reason to care that we can do better these days.

Re: xHE-AAC : The Death of OPUS?

Reply #40
AAC killed MP3 about 15 years ago.  Interesting that no one told Amazon, as they're still selling MP3's in 2021.

Re: xHE-AAC : The Death of OPUS?

Reply #41
Personally, I think the future is bright for Opus; however, not due to its own merits as an audio codec, but as a side product of being the most likely candidate for use in AV1. And with support from pretty much everyone - from hardware vendors to browser developers - AV1 itself is pretty much guaranteed to have widespread support, and it will drag Opus along.

Re: xHE-AAC : The Death of OPUS?

Reply #42
I mean, NPR declared mp3 dead in 2017 https://www.npr.org/sections/therecord/2017/05/11/527829909/the-mp3-is-officially-dead-according-to-its-creators

Meanwhile in their podcast department, the Planet Money podcast episode released on January 29 is 128 kbps CBR mp3 and the audio version of the Tiny Desk Concert released on January 21 is 256 kbps CBR mp3...

EDIT: I'm just saying that once a codec gets a certain level of penetration it's virtually impossible to "kill"

Re: xHE-AAC : The Death of OPUS?

Reply #43
I mean, NPR declared mp3 dead in 2017 https://www.npr.org/sections/therecord/2017/05/11/527829909/the-mp3-is-officially-dead-according-to-its-creators
The amount to which this never made sense boggles my mind. How in the world does Fraunhofer's patent licensing rights expiring translate to MP3 being dead? "Ah yes, you no longer need to pay Fraunhofer in order to manufacture hardware with MP3 support - therefore, MP3 is dead." I'm having trouble discerning if the person who wrote this article truly had no idea what they were talking about or if this was an intentional attempt of twisting the narrative for some reason.

Re: xHE-AAC : The Death of OPUS?

Reply #44
I mean, NPR declared mp3 dead in 2017 https://www.npr.org/sections/therecord/2017/05/11/527829909/the-mp3-is-officially-dead-according-to-its-creators
The amount to which this never made sense boggles my mind. How in the world does Fraunhofer's patent licensing rights expiring translate to MP3 being dead? "Ah yes, you no longer need to pay Fraunhofer in order to manufacture hardware with MP3 support - therefore, MP3 is dead." I'm having trouble discerning if the person who wrote this article truly had no idea what they were talking about or if this was an intentional attempt of twisting the narrative for some reason.

NPR to MP3: You're dead.
MP3: So is vinyl. Bwahahahahaha.

Re: xHE-AAC : The Death of OPUS?

Reply #45
I honestly almost wonder if Fraunhofer told them to frame it that way and they just did what they were told.

Re: xHE-AAC : The Death of OPUS?

Reply #46
Nor can tell a Lame encode at 192kbps from lossless.

I can't either. Also can't tell a 160k AAC/CoreAudio from lossless.

But my ears are 54 years old and my hearing drops off at 11.5 khz....

Re: xHE-AAC : The Death of OPUS?

Reply #47
I honestly almost wonder if Fraunhofer told them to frame it that way and they just did what they were told.
To be honest, that's definitely not too far fetched in my opinion. This portion of the article hints at that more than anything:
Quote
So is it the end of an era? We may still use MP3s, but when the people who spent the better part of a decade creating it say the jig is up, we should probably start paying attention.
"Well actually, MP3 can still be used, but I mean come on guys - Fraunhofer said the jig is up! Just... just stop, ok? I'm sure they'll offer paid licensing programs for some other format. So yeah. :) "

 

Re: xHE-AAC : The Death of OPUS?

Reply #48
Kinda off-topic but I wanted to share my opinion.

I am pretty sure they will release new codec/"container" within next couple of years because right now things are very complicated.
This codec/"container" will unify all existing codecs, extensions...

Let's start with MP3.
MP3 is very old codec and has some flaws. It was designed for music (stereo) at mid-high bitrates.
It doesn't support multichannel. They tried to fix it with "MP3 Surround" but it failed.
It doesn't support lossless. They tried to fix it with "mp3HD" but it failed.
It sucks at low bitrates. They tried to fix it with "mp3PRO" but it failed.

Let's start with AAC.
It supports multichannel. (Yay!)
Lossless is not really supported. Have you ever heard about SLS? Neither did I. To make it even more confusing, ALAC is also using m4a container.
It sucks at low bitrates. They tried to fix this not once (HE-AACv1), not twice (HE-AACv2) but three times (xHE-AAC).

If device says that it can play m4a files, how are they going to explaing to users why their files are not working?
This kinda reminds me of USB renaming scandal. (Is this USB 3.0? Well yes, but actually no.)
Things are more than complicated for average user.

Re: xHE-AAC : The Death of OPUS?

Reply #49
Quote
It sucks at low bitrates
Name any other lossy codec that performs better at those low bitrates.

Quote
It doesn't support lossless
Why does it matter if a lossy codec doesn't have a lossless mode?

 
SimplePortal 1.0.0 RC1 © 2008-2021