Opus 1.3 Released - One Of The Leading Lossy Open-Source Audio Codecs

[LightBit]

FLAC?

Free Lossless Audio Codec

[anarki2]

Unfortunately, still noone cares. Why can't it be added to Bluetooth, for example? Why, just why?

[Weasel]

Taking away inaudible data is the whole point of lossy compression, so this does not in any way tell us anything about Opus vs. AAC quality wise. If anything, it just shows Opus is doing it's job.

That is not the job of lossy compression. The job of lossy compression is to compress at a given bitrate on average.

How it achieves that determines its quality. All of them drop inaudible information, but they're not required to. If they can keep all data, then they should keep it. Thing is, they don't do that in equal ways. If a codec provides the same quality/information at low frequencies (I did a difference test and boosted signals massively, didn't notice much on spectrum compared to Opus), but provides more high freqs it means it's simply objectively better. I don't care if it's inaudible, it's still extra information.

Obviously AAC encoders already think they do a maxed job on low frequencies to even bother to encode high frequencies.



As an aside, what you said is just plain ridiculous. For example, MP3 cuts way more frequencies than Opus, does that mean MP3 is the best lossy compressor and does its job the best? Even more, older MP3 encoders cut WAY more than current MP3 encoders. These days they tend to cut as much as Opus but not more, while previously shitty encoders (remember Xing?) cut at 16k even with 256kbps!

Vorbis is ok but it suffers from transient problems where it needs very high bitrate to encode them in decent quality.

[caligula]

Unfortunately, still noone cares. Why can't it be added to Bluetooth, for example? Why, just why?

The market is run by big players who want their patented proprietary codec included. The home theater receivers and BT audio stuff already includes tons of lossy/lossless patented formats. The lossless formats are all crappier than mkv/flac, but who cares. At least you need to pay for them. Yay.

[jacob]

So YouTube actually converts AAC to OPUS? 'Cause Google recommends AAC for uploading: https://support.google.com/youtube/answer/1722171?hl=en

If you upload in AAC then yes, they do convert it to OPUS for WebM/VP9 playback.

And Netflix seems to use mp4a with VP9: https://www.quora.com/In-what-format...reaming-movies

OK, point taken. I didn't know that. I wonder what made them choose this.

[microcode]

Vorbis is better for replaygain support though… :-/

I was looking at that, it's arguably better for ReplayGain. If you are going from ReplayGain'd FLACs, it'll set the album level as the default program level for the file, and if you have a track gain, it'll include that as well (though your player needs to support reading the track gain).

What I've noticed is that this has the effect of automatically making the (album) ReplayGain work in everything, including random audio players on the web.

[microcode]

That is not the job of lossy compression. The job of lossy compression is to compress at a given bitrate on average.

How it achieves that determines its quality. All of them drop inaudible information, but they're not required to. If they can keep all data, then they should keep it. Thing is, they don't do that in equal ways. If a codec provides the same quality/information at low frequencies (I did a difference test and boosted signals massively, didn't notice much on spectrum compared to Opus), but provides more high freqs it means it's simply objectively better. I don't care if it's inaudible, it's still extra information.

Obviously AAC encoders already think they do a maxed job on low frequencies to even bother to encode high frequencies.



As an aside, what you said is just plain ridiculous. For example, MP3 cuts way more frequencies than Opus, does that mean MP3 is the best lossy compressor and does its job the best? Even more, older MP3 encoders cut WAY more than current MP3 encoders. These days they tend to cut as much as Opus but not more, while previously shitty encoders (remember Xing?) cut at 16k even with 256kbps!

Vorbis is ok but it suffers from transient problems where it needs very high bitrate to encode them in decent quality.

Please leave the explaining to people who have done their research. Nobody is judging you for not sharing your opinion, and falling short of the truth just dooms everyone who reads and believes you to be just as wrong as you.

[Weasel]

Please leave the explaining to people who have done their research. Nobody is judging you for not sharing your opinion, and falling short of the truth just dooms everyone who reads and believes you to be just as wrong as you.

Please leave the explaining to people who have done their research. Nobody is judging you for not sharing your opinion, and falling short of the truth just dooms everyone who reads and believes you to be just as wrong as you.

[microcode]

Please leave the explaining to people who have done their research. Nobody is judging you for not sharing your opinion, and falling short of the truth just dooms everyone who reads and believes you to be just as wrong as you.

Okay bud, lets break this down.

That is not the job of lossy compression. The job of lossy compression is to compress at a given bitrate on average.

But it in fact is the job of lossy audio compression to discard inaudible data, in particular, unpredictable inaudible data; lossy compressors don't mind coincidentally encoding inaudible data if those data are produced by the predictor at no additional cost. Discarding unpredictable inaudible data is the process by which a lossy audio compressor "compress[es]", and rate control is how it does so "at a given bitrate on average."

How it achieves that determines its quality. All of them drop inaudible information, but they're not required to. If they can keep all data, then they should keep it. Thing is, they don't do that in equal ways. If a codec provides the same quality/information at low frequencies (I did a difference test and boosted signals massively, didn't notice much on spectrum compared to Opus), but provides more high freqs it means it's simply objectively better. I don't care if it's inaudible, it's still extra information.

Obviously AAC encoders already think they do a maxed job on low frequencies to even bother to encode high frequencies.

Not really sure what you're getting at here.

As an aside, what you said is just plain ridiculous. For example, MP3 cuts way more frequencies than Opus, does that mean MP3 is the best lossy compressor and does its job the best? Even more, older MP3 encoders cut WAY more than current MP3 encoders.

Almost all MP3 encoders were capable of encoding wideband audio, and that was already the default width for standard music bitrates by the time anyone was really using it. MP3 is missing proper tools for encoding frequencies above 16kHz, so MP3s with any frequency content above 16kHz are basically a novelty, even if LAME manages to start to do something useful in that range at outrageous (180kbps+!) rates. To me, it seems like basically all MP3 encoders did wideband encodes then, and do little more than wideband encodes now.

These days they tend to cut as much as Opus but not more, while previously shitty encoders (remember Xing?) cut at 16k even with 256kbps!

At about 32kbps, Opus will enable fullband encoding; so virtually any music-rate Opus encoding will have more bandwidth than virtually any standard rate MP3. At a given bitrate, chances are that MP3 will be wideband (or somewhere between WB and FB), and Opus will be fullband. That is to say "These days", "[MP3 encoders]" don't "tend to cut as much as Opus but not more", but rather they tend to cut more than Opus at almost any given bitrate. The reason for this is simple: MP3 makes encoding frequencies above 16kHz very expensive, because it barely even designed with this in mind; if some of the people at MPEG had their way, they would have specified the decode sample rate at 32kHz and washed their hands of the whole ordeal.

Vorbis is ok but it suffers from transient problems where it needs very high bitrate to encode them in decent quality.

Yeah, Vorbis's rate control is not perfect, I agree.

Sorry, that was a bit snooty of me, and I'm no expert, but it just seems weird to me to snuff out somebody's good first instinct "compression is about inaudible data" with a hard denial, and then go on to explain why it's a good instinct.

[SciK]

That is not the job of lossy compression. The job of lossy compression is to compress at a given bitrate on average.

Well, an equally valid use-case is to target a certain quality (see e.g. x264’s CRF mode, vorbis’s and mozjpeg’s quality parameter, etc.).

How it achieves that determines its quality.

And you think that the higher-quality one is the one that spends its bits on useless information? Interesting.

All of them drop inaudible information, but they're not required to. If they can keep all data, then they should keep it.

But why? If they could achieve transparency at a lower bitrate, why even bother with the higher bitrate?

Thing is, they don't do that in equal ways. If a codec provides the same quality/information at low frequencies (I did a difference test and boosted signals massively, didn't notice much on spectrum compared to Opus), but provides more high freqs it means it's simply objectively better.

Spectrograms can be useful, but they only tell part of the story. Did yours show the phase as well or just the intensity? How is its temporal resolution?

Have you listened to the files? Because ultimately, that’s what you compress them for, right?