Please create an account to participate in the Slashdot moderation system

 



Forgot your password?
typodupeerror
×
Music

Amazon Music Rolls Out Lossless Streaming Tier (theverge.com) 48

Amazon Music HD is a new tier of Amazon's music service that offers lossless versions of audio files for streaming or downloading at a price that aggressively undercuts Tidal, the main competition for this kind of audio. "Amazon will charge $14.99 a month for the HD tier, or $12.99 if you're an Amazon Prime customer," reports The Verge. "Tidal's Hi-Fi plan costs $19.99 monthly." From the report: Amazon says it has a catalog of over 50 million songs that it calls "High Definition," which is the term it's applying to songs with CD-quality bit depth of 16 bits and a 44.1kHz sample rate. It also has "millions" (read: less than 10 million, more than one million) of songs it's calling "Ultra HD," which translates to 24-bit with sample rates that range from 44.1kHz up to 192kHz. Amazon Music HD will deliver them all in the lossless FLAC file format, instead of the MQA format that Tidal uses.

Amazon's VP of Music, Steve Boom, tells me that Amazon chose the HD and UltraHD terminology because it found it was more comprehensible to a mass audience than the current terminology for audio quality. And "mass audience" is exactly what Amazon is going for; it doesn't want Amazon Music HD to be a niche player like Tidal and other lossless music platforms like HDtracks or Qobuz. Boom says that "It's a pretty big deal that one of the big three global streaming services is doing this -- we're the first one."
In response to today's news, Rock legend Neil Young said (with no hyperbole whatsoever): "Earth will be changed forever when Amazon introduces high quality streaming to the masses. This will be the biggest thing to happen in music since the introduction of digital audio 40 years ago."
This discussion has been archived. No new comments can be posted.

Amazon Music Rolls Out Lossless Streaming Tier

Comments Filter:
  • Neil Young (Score:5, Funny)

    by PeeAitchPee ( 712652 ) on Tuesday September 17, 2019 @04:39PM (#59205636)
    No one in Neil's generation can remotely hear this type of high end. That subscription might as well come with a set of High Definition Ethernet cables so your 1s and 0s have more integers and fewer decimals.
    • It's actually easy to hear - play something with a lot of massed strings or lots of cymbals and listen for the "watery" sound in the 6-12 kHz range. Makes most lossy compression algorithms stand right out!
    • Compression artifacts can show up in lower frequencies too, though.

      But, this service is crying out for a good independent A/B test. And throw vinyl in the mix also!

    • It's not just about high frequencies. Lossy compression also corrupts the signal overall. Listen to cymbals or other spectrally busy musical content with MP3 for example - cymbals sound messy and awful.
      • MP3 is a very old codec by now. You'd make a better argument if you can say the same thing about AAC at reasonably high encoding rates. Just about any lossy codec will exhibit artifacts if you compress too aggressively.

        That being said, these lossy formats were invented when 8K streaming video and terrabyte-sized hard drives were not a thing. So, if people want lossless, why not?

        • Re:Neil Young (Score:5, Insightful)

          by JoelWink ( 1846354 ) on Tuesday September 17, 2019 @06:00PM (#59205978)

          Here's a great site for randomized A/B testing of AAC 320 kbps vs FLAC. Be sure to use a computer with a decent sound card and some quality headphones. Warning: you might find out that your ears aren't as "golden" as you think:

          http://abx.digitalfeed.net/ [digitalfeed.net]

          Some additional tests are here:

          http://abx.digitalfeed.net/lis... [digitalfeed.net]

          • by sl3xd ( 111641 )

            Randomness is so fun... I can't hear the difference at all, and picked "A" every time.

            I scored 100% and 80% on two of the samples, and 64% overall.

          • AAC 320 kbps is still overkill. For the vast majority of people, LAME MP3 at -V3 should be transparent, with -V4 being close to transparent. Those are approximately 195 and 165 kbps respectively. This is going off the old info on Hydrogen Audio's LAME page [hydrogenaud.io] which I believe hasn't been updated in years; it's entirely possible that LAME's gotten even better at encoding since then.

            Generally speaking, AAC has fewer artifacts than MP3 at equivalent bitrates on the lower side, though it's somewhat encoder-depe
    • I’ve always wondered if Neil Young is just abnormal in the sense that he has ears or neural wiring necessary for this kind of format to matter. Basically he’s Magic Johnson to the rest of us schlubs playing a pick-up game. I’m not saying this is the case, but maybe his aural senses are really on a different level and he thinks we’re the crazy ones just like you and I would think n a person mad if they couldn’t taste the difference between fresh, purified spring water and some m
      • True story -- Neil loves Lionel trains, and used to grow weed on his layout. I know this because someone I know got high with him at his house years ago!
    • by tlhIngan ( 30335 )

      No one in Neil's generation can remotely hear this type of high end. That subscription might as well come with a set of High Definition Ethernet cables so your 1s and 0s have more integers and fewer decimals.

      It never was about high end. The vast majority of music will make absolutely no difference. Why? Because at one point in their lives, they existed as lower quality file!

      The provenance of a good chunk of "high res" music is questionable at best - you'll often find the "studio masters" will be 44.1/49kHz

  • Neat! (Score:5, Insightful)

    by DogDude ( 805747 ) on Tuesday September 17, 2019 @04:48PM (#59205674)
    I listen to my music in hi-def, it costs $0 every month, I don't need an internet connection, and I don't have to give Amazon all of my personal information. I buy used CD's for a buck or two each and then rip 'em.
    • What's your success rate getting error-free rips from used CD's?
      • by DogDude ( 805747 )
        Pretty close to 100%. I get probably 10ish a month, and I don't remember the last bad rip I got. I don't do anything fancy. Just Winamp.
        • Out of the last couple-- four or five ripped successfully, one was ejected by my cd drive (La Cetra/Podger), and one caused iTunes to abort with an error (4 seasons/Podger).

        • What source do you use for ID3 tags, I guess that is working reliably too?
        • 10 CDs a month, say 2 dollars each, that's 20 bucks a month for 100 to 200 songs.
          I listen to 30 to 60 songs a day, each day, and pay 4.99 EUR a month through Spotify.

          Now, about Amazon's offer... I have checked one of my favorite bands (In The Woods...) on both Spotify and Amazon Music. Amazon has 4 "popular" albums with no apparent way to show all of them (In The Woods, Omnio, Strange In Stereo and A Return to the Isle of Men). Spotify has 8 albums (Heart of The Ages, Omnio, Strange in Stereo, Three Times S

    • Wouldn’t it be more fair to say that it costs you whatever you pay for the CDs minus anything you recoup if you sell them on yourself? I don’t pay for streaming either, but I still probably spend $20 on music per month if you were to average it out.
    • I listen in high def for the price of five of your CDs but I can listen to as many as I want and even in the highly unlikely scenario that I donâ(TM)t have internet access I can download the tracks I want to keep on my phone.
      • by DogDude ( 805747 )
        I've had my music collection for more than 25 years. I imagine I'll have it for another 50 years or so. I don't have to rent it. I like that.
        • by AntEater ( 16627 )

          No rent. It doesn't get removed from my library when the label contract expires. Nobody tracks what I listen to, when I listen to it or which device I use to listen with. No limits to how many devices I use or how old the operating system may be. Best of all, it works without any internet connection at all.

          I'm not against streaming services, if it works for you, but there is value in "owning" your collection that Spotify, Amazon and others don't provide. I may use them more like someone would traditio

    • by AmiMoJo ( 196126 )

      Used CDs can sometimes be better than even lossless streaming too. When CDs were relatively new they were properly mastered.

      By the 90s mastering engineers had realized that they could push the levels far higher than vinyl ever allowed and we were well into the loudness war. But if you can find original releases from the 80s they often sound superb.

      You can often get a great sounding CD for a buck, rather than paying out for the latest re-master that may or may not be distorted, compressed crap.

  • ""High Definition," which is the term it's applying to songs with CD-quality bit depth of 16 bits and a 44.1kHz sample rate"
    ""Ultra HD," which translates to 24-bit with sample rates that range from 44.1kHz up to 192kHz"

    44.1kHz or 48kHz sample rate at 16-bits or more per sample should be called "Standard Definition".
    96kHz and above at 16-bits or more per sample are "High Definition".
    768kHz at 32-bits or more per sample is "Ultra Definition".

    Of course this only applies to uncompressed or losslessly compressed

    • Re: (Score:2, Informative)

      There is a published standard of what High Res is [abt.com] - and it is greater than CD. You are correct, this is Amazon basically bastardizing what high res is. What they call high res, is actually Hi-Fi (not master) in Tidal-world.
      • Something that is an "Internal Use Only" marketing standard from Sony Corporation (the malicious malware distribution company) does not a "published standard" make.

    • Interesting. So you can hear sounds above 24 kHz? I'd like to see you ABX that.
    • Joint channel encoding is not a dorky format, it's a lossless transform that can significantly improve compression. Even FLAC uses it.

      And a sampling rate of 768 kHz is way overkill, no living animal in this planet can hear frequencies that high.

  • Odd this should come out. Almost a month ago to the day, the New York Times put up a long story about an interview with Neil Young [nytimes.com]. In short, he's grouchy, crotchety, hates pretty everyone, but most especially hates compressed music. A quote from the article:

    When you hear real music, you get lost in it, he added, “because it sounds like God.” Spotify doesn’t sound like God. No one thinks that. It sounds like a rotating electric fan that someone bought at a hardware store.

    And now he's go

  • This is an impossible claim for any digital format that can exist in reality, regardless of the frequency that it is sampled at.

    The losses may very well be inaudible, but until you are sampling at such a frequency that the effective wavelength is less than a Plank length, it's not truly lossless.

    • "Lossless" in this case refers only to the compression method used on an uncompressed digital source to produce a compressed audio file. You are talking about digitizing the music in the first place, which is not referred to as "lossless" in any way, as far as I know (I am not an expert).
    • "Lossless compression" means that the digital datastream that is fed into the compressor is exactly the same as digital datastream that is produced as the output of the decompressor. Most PC archiving programs (ZIP, BZIP, COMPRESS, RAR, etc) are Lossless compressors. There are also a few specific lossless audio stream compressors (though standard compressors would work just fine too).

      This contrasts to the non-lossless varieties of compressors, such as all the rest (MP3, AAC, etc) which do NOT produce outp

    • by mvdwege ( 243851 )

      Shannon and Nyquist [wikipedia.org] say you're wrong.

      • by lurcher ( 88082 )

        Yep, the original poster seems to be assuming that his original source has a infinite signal to noise ratio and frequency bandwidth. It hasn't and doesn't, so as you say, its possibly to convert to digital and back without losses.

        • by mark-t ( 151149 )

          The original source is analog... there is no "bandwidth". Sampling at twice the highest single frequency in the source will capture all of the nodes of the signal, but not necessarily all of the amplitude.

          The fact that these losses might be inaudible doesn't mean they aren't losses.

          • by mvdwege ( 243851 )

            Dear fucking God, are you stupid or what?

          • The whole world runs in analog and yes analog has bandwidth. Go read Nyquist and Shannon on Wikipedia.

            • by mark-t ( 151149 )

              There is no magic silver bullet that can capture 100% of an analog signal in digital form. It doesn't matter how fast you sample, or how many times faster than the highest frequency of interest you sample, it is naive to assume that the frequencies which might interest someone are the only frequencies that exist.

              If you sample at 192khz, there probably won't be any audible losses, but that doesn't mean there aren't losses at all. There are way more frequencies higher than what a 192khz sampling could re

  • Circa early 2012 Neil Young told us that 24-bit 192kHz downloads were "uncompromised studio quality" and that we needed this playing in our homes. Xiph's Monty Montgomery disagreed with Neil Young & Steve Jobs. Monty laid out good reasons for his disagreement [xiph.org].

    Neil Young recently told us:

    Earth will be changed forever when Amazon introduces high quality streaming to the masses. This will be the biggest thing to happen in music since the introduction of digital audio 40 years ago.

    Not only because of the obvious hyperbole (making the statement hard to take seriously on its face) but recent history would suggest skepticism is wise.

  • Anything less than 24 bit samples at 96 kHz should not be called high resolution audio, and I noticed they aren't calling it that, so buyer beware.
  • THAT would have been infinitely more interesting than whatever this shit is. yawn.

If all the world's economists were laid end to end, we wouldn't reach a conclusion. -- William Baumol

Working...