24/192 Music Downloads and why they make no sense (2012) • Hacker News Board

Re: 24/192 Music Downloads and why they make no sense (2012)

#71

Post by teknico » Wed, Mar 06, 2019, 4:49 PM UTC

deevious wrote: post #6

This reminds me of Monty's A Digital Media Primer for Geeks[0] and Digital Show & Tell[1] - the delivery, the explanations and the way the experiments are set up is superb. [0] https://xiph.org/video/vid1.shtml [1] https://xiph.org/video/vid2.shtml

The article's author, Chris "Monty" Montgomery, is one of the authors of Ogg Vorbis [1] and Opus [2].

It puzzles me that many people don't yet know about Opus. Let me quote the FAQ [3]:

"Does Opus make all those other lossy codecs obsolete?

Yes.

From a technical point of view (loss, delay, bitrates, ...) Opus renders Speex obsolete and should also replace Vorbis and the common proprietary codecs too (e.g. AAC, MP3, ...)."

[1] https://xiph.org/vorbis/

[2] http://www.opus-codec.org/comparison/

[3] https://wiki.xiph.org/OpusFAQ#Does_Opus_make_all_those_other...

Re: 24/192 Music Downloads and why they make no sense (2012)

#72

Post by vortico » Wed, Mar 06, 2019, 4:49 PM UTC

sideshowb wrote: post #47

vortico wrote: post #14

Earlier quoted context omitted.

So you think our senses transcend what tools have the ability to measure? Maybe that was the case in 1970, but in 2000+, hearing (and vision) is completely understood scientifically and far surpassed by measuring apparatus at every frequency range. Saying otherwise is an appeal to what is called audio mysticism and is caused by placebo and confirmation bias, which was mentioned in the article.

Actually that's a defensible position when you're mixing and mastering audio. A spectrogram won't tell you which settings sound better, but your ears will. Maybe not relevant in the context of GP's post though. In the context of digital tools telling you two signals are identical, then I suspect they are, and if you want to prove to me your ears hear better then you're going to need a blind A/B test.

Right, this debate is not about determining whether two different audio clips A and B sound better. It's about determining whether two clips can be distinguished by our ears at all.

Re: 24/192 Music Downloads and why they make no sense (2012)

#73

Post by jscheel » Wed, Mar 06, 2019, 4:51 PM UTC

mruts wrote: post #4

Problem with music sounding bad doesn’t really have much to do with the distributed format: V0, V1, or 320 mp3s should sound pretty much the same compared to 16-bit flac. You can only the difference between mp3 and flac at shitty bitrates no one uses anymore (like 120). The reason why a lot of recent digital music sounds bad is because of the intentionally terrible mastering. Since everyone is listening of crappy ear…

So this is why streaming Google Music on my fairly nice sound system ends up sounding like total crap?

Re: 24/192 Music Downloads and why they make no sense (2012)

#74

Post by gpjanik » Wed, Mar 06, 2019, 4:52 PM UTC

I wonder about one thing. Sure you can't hear above/under certain frequencies, but these frequencies still resonate with parts of your body (that are not ears) and you might feel it in other ways than just hearing, and also their presence generates harmonics. Not sure if it is observable to a human, but just because you don't hear N hertz, doesn't mean you can't hear its harmonics/it doesn't affect your _perception_ of the rest of the signal at all. Cutting off some frequencies can create patterns that are not hearable per se, but might induce unwanted sensory feelings (*opinion, not a fact). I think using physics to break this down doesn't make much sense, and that the most practical debate solution would be to do a double-blind test on a statistical group of the so called audiophiles.

Re: 24/192 Music Downloads and why they make no sense (2012)

#75

Post by bch » Wed, Mar 06, 2019, 4:52 PM UTC

vortico wrote: post #66

bch wrote: post #59

Earlier quoted context omitted.

So, a high frequency DTMF[0] construction should render a pre-echo? What’s the lowest definition of “high” pairs that would satisfy showing this encoding breakage but be hearable by a majority of people[1]? [0] https://en.wikipedia.org/wiki/Dual-tone_multi-frequency_sign... [1] https://en.m.wikipedia.org/wiki/Presbycusis

No, pre-echos are unrelated to high-frequency sine waves. And DTMF aren't what I'd consider high-frequency. I'm talking about 4-20kHz.

Ahhh... I think I misunderstood your “dialup noises” as dial tone noises. You’re talking about the high pitched squelch-y initial negotiation of a modem.

Re: 24/192 Music Downloads and why they make no sense (2012)

#76

Post by PascLeRasc » Wed, Mar 06, 2019, 4:52 PM UTC

mikeash wrote: post #29

PascLeRasc wrote: post #9

I have a degree in electrical engineering, and I'm currently in a graduate course on computer music systems, so I hope that qualifies me enough to avoid the author's ad hominem attacks he seasons this stinkpiece with. I can't stand seeing frequency response charts and scientific measurements in articles about audio. Like my favorite audio reviewer says [1], I listen to music for enjoyment and I talk about audio in su…

If two wines were chemically identical but someone claimed that one of them tasted “warmer” than the other, would you just go with that?

Sure. I don't need to invalidate someone else's experience.

Re: 24/192 Music Downloads and why they make no sense (2012)

#77

Post by themodelplumber » Wed, Mar 06, 2019, 4:52 PM UTC

martin_a wrote: post #8

mruts wrote: post #4

Problem with music sounding bad doesn’t really have much to do with the distributed format: V0, V1, or 320 mp3s should sound pretty much the same compared to 16-bit flac. You can only the difference between mp3 and flac at shitty bitrates no one uses anymore (like 120). The reason why a lot of recent digital music sounds bad is because of the intentionally terrible mastering. Since everyone is listening of crappy ear…

But who is using MP3 players these days any more? I found myself to buy an iPod in... like... 2011 or so. Converted all the CDs I had to FLAC because losless was the way to go. Two or three years (let it be 5, doesn't matter) pass by, I got a better Smartphone, Spotify Premium and don't touch my 1xx GB of FLAC music anymore, because I don't want to carry around another device etc. I'm not sure but I think "owning" mu…

I hike a lot and hate using my phone's battery power for music. On top of needing that power for other things, it just feels wasteful. I bought a cheap MP3 player to try out in 2016 and have been hooked ever since. These devices are smaller and lighter than spare phone batteries or power banks.

In addition, I find that I use the MP3 player when I'm out running normal errands precisely because I've organized my music by hand and even edited tracks by hand in some cases. Examples would be things like rare covers that can only be found on YouTube, or favorite songs from niche internet music communities which were poorly mastered.

It's also a bit of a gear hobby now since there are so many MP3 players on the market. Prices are low and performance is great.

I have to agree about the iPod though, as I found the need for proprietary software, and really annoying software at that, made me use it less and less until my 32GB iTouch was mostly used as an ebook reader. I also prefer physical buttons for my mp3-listening while on the go.

Re: 24/192 Music Downloads and why they make no sense (2012)

#78

Post by discreditable » Wed, Mar 06, 2019, 4:54 PM UTC

badfrog wrote: post #24

Mediterraneo10 wrote: post #5

Thanks to this article, when I torrent music that is shared in 24/192 format, I always resample to 24/96: ffmpeg -i foo.flac -ar 96000 -acodec flac bar.flac

I'm curious why you torrent music still when streaming is so widely available and free/cheap? I did a lot of torrenting back in the 2000s, but thinking back on it I spent a ton of time finding things, organizing my file system, transcoding, editing metadata, etc. I do not miss that hassle at all now.

Some bands still refuse to be available on streaming (ex: Tool). Some will never be on streaming.

Re: 24/192 Music Downloads and why they make no sense (2012)

#79

Post by shmerl » Wed, Mar 06, 2019, 4:55 PM UTC

teknico wrote: post #71

deevious wrote: post #6

This reminds me of Monty's A Digital Media Primer for Geeks[0] and Digital Show & Tell[1] - the delivery, the explanations and the way the experiments are set up is superb. [0] https://xiph.org/video/vid1.shtml [1] https://xiph.org/video/vid2.shtml

The article's author, Chris "Monty" Montgomery, is one of the authors of Ogg Vorbis [1] and Opus [2]. It puzzles me that many people don't yet know about Opus. Let me quote the FAQ [3]: "Does Opus make all those other lossy codecs obsolete? Yes. From a technical point of view (loss, delay, bitrates, ...) Opus renders Speex obsolete and should also replace Vorbis and the common proprietary codecs too (e.g. AAC, MP3, .…

I use Opus for music playback for all my archived music. The reason it's not more widespread was opposition of the likes of Apple to free codecs. Today they are losing this, and Opus is making its way even to Apple's systems.

Re: 24/192 Music Downloads and why they make no sense (2012)

#80

Post by xoa » Wed, Mar 06, 2019, 4:55 PM UTC

pjc50 wrote: post #58

The choice of colour analogy is unfortunate, because there really are colours that are "out of gamut" and cannot be accurately reproduced on normal monitors. If you have the opportunity to and look at one of the IKB works in person you'll see what I mean. https://www.tate.org.uk/art/artworks/klein-ikb-79-t01513 https://contemporaryartetc.wordpress.com/2007/09/13/fact-of-...

I don't quite agree with you, taking into account it's an analogy designed to help illustrate the issue for general audiences. It's not as if we don't have ProPhoto RGB or other wide gamuts or don't understand the issues of rendition accuracy and resolution within the visual spectrum. There was never any debate that sRGB alone in particular was quite limited, or that dynamic range was an issue. It's just that it represents a ton more data and is technologically and commercially much, much harder. As tech has caught up displays have continued to chase human visual limits, starting with resolution, then frame rate, and finally major industry wide improvements to gamut and range with BT.2020/2100. I mean heck, it wasn't that long ago that we barely had color at all. I still remember well the first 8-bit system I ever got, or back when I regularly had to manually change between 16-color/256/16k to prioritize resolution or color because my system just didn't have enough VRAM to handle both at once. Audio did far better at matching human limits much, much longer ago.

But within the visual spectrum but not showing on the screen is still within the visual spectrum. The article examples refer to infrared and UV+ for contrast, and that's entirely correct. Monitors displaying either of those would make no difference (well, beaming ionizing EM at your face raises significant concerns audio doesn't at any level) at any point. They're simply beyond human eyes period. It's an accurate analogy. Failing to reproduce something within human limits would be what you're talking about, but that's a solved problem and not something 24/192 offers you anything with.