See If You Can Hear The "High Definition" Audio Jay Z Is Selling

See If You Can Hear the 'High Definition' Audio Jay Z Is Selling

Jay Z's newly acquired and relaunched streaming music service Tidal claims only one difference from Spotify and the rest: superior audio quality. Is it worth switching? Here's a simple test that will help you decide whether it's worth forking over $US20 a month for the best of the best quality.

Sixteen of the world's richest pop stars fed us a lot of idealistic bullshit about the future of music yesterday. But if you take Tidal's A/B test and you can't tell the difference between the low-quality and high-quality tracks, just stick to the crappy Mp3s you ripped in 2002.

In the test, you're presented with five songs, each in both low and high quality. You'll be able to switch between them as they play, and even listen again if you would like. Whether or not you can distinguish doesn't just have to do with the quality of your hearing. The environment you're in, the gear you're using, and whether or not you've trained your ears for close listening all play a part as well.

I first wrote abut this test in the fall, but it's worth revisiting in the wake of Jay Z's hype parade. Remember: No matter how high you score, you were probably listening really carefully. Chances are that if you're on a streaming music service you're listening to tunes in noisy environments like public transportation or the train. Hell, you're probably listening in the background at work as Gchat notifications pipe up ever other minute. Can you hear those bass flourishes over the sound of your screaming child?

Tidal's $US20/month service is only worth it to horrible nerds (like me!). Don't t pay the Jay Z premium if you're not sure you're one of them. [Tidal]

Can You Hear the Difference in High Quality Music? Test Yourself

Here's a simple test that lets you toggle between an extremely high-quality file music file and a low-quality music file. Can you tell the difference?

We've been hearing for years that MP3 encoding ruins music with its lossy compression. But can you really tell the difference? New streaming music subscription service "Tidal" is amongst the first to offer lossless audio, and in an effort to get more people to sign up, it's set up this simple A/B test. Fire it up, and the test will play you a song and allow you to toggle between two versions of five possible songs. This is a blind A/B test, meaning you won't know which is high quality and which isn't, which is one of the surest ways to eliminate your biases in choosing one or another version.

A number of things can affect whether or not you can tell the difference between lossless and lossy music, including the health of your hearing, training to know what to listen for, as well as the quality of your gear.

I got 4/5 right on my good mid-range headphones, but it's harder than you would think! And if I hadn't be scrunching my face and listening super carefully, who knows if I would have noticed. If you score very low, maybe you're not the type of person who needs to be investing in lossless files? Which great, that's a lot cheaper for you.

Let us know what you get in the comments below. [Tidal]


Comments

    5/5 on a MacBook Air with midrange Klipsch in ear headphones. Not that hard but without the side-by-side comparison I doubt I'd notice the difference in real world listening. Or care.

    I got 3 out of 5 on decent studio monitors – which is what you might get by chance.

    I think the reason why people say 'music is so low quality these days' isn't because of compression, but because people are listening on cheap earbuds/computer speakers etc. Which, incidentally, is fine because they’re quite happy with it.

    But if you want your music to sound better, don't invest in expensive lossless audio systems/files – just buy a decent pair of headphones / speakers and listen to MP3s.

    Even with music that sometimes suffers more from compression (like orchestral music), you’d struggle to tell the difference between a Spotify stream at 320kbps and a lossless format.

    I think consumer lossless audio is mainly a sales pitch along the lines of digital camera manufacturers ‘more pixels is better’. Or Monster cables.

      Rubbish - it's almost entirely because of dynamic compression (as opposed to lossy compression by converting to MP3/AAC/OGG etc).

      Give me a well mastered album converted to 128kbps MP3 over a hyper-compressed FLAC album any day. This is the same reason why many armchairists suddenly conclude vinyl is better than anything (because they normally don't overcook the mastering for vinyl for whatever reason).

        My fault, I probably wasn’t that clear. I think we’re making two sides of the same point.

        You make a good point that quality has a lot to do with what happens before encoding (mastering).

        My point was about making the music that exists sound better. You can't go back and remaster a track yourself, but you can buy better speakers/headphones which will increase the sound quality. Switching to lossless won't increase sound quality (or at least not noticeably).

        Last edited 22/11/14 9:43 pm

        Sorry pal, but you don't know what you're talking about. Compresssion is completely, utterly, totally lossless. It only affects the dynamic range and always has the opposite effect to that which you're implying. Without compression, whatever is loudest overwhelms the subtle elements of a mix. Compression allows everything to sit in the mix much, much better, letting you to hear everything that's in the recording. OK, that includes noise inherent in the recording process (hiss, hum, etc.) but that's easy enough to minimise/eliminate.

        Of course, it can be over-used sometimes but if you ask any recording artist which version of their material sounds better, they will always pick the well-compressed version over the uncompressed one. That's where problems can occur - the more you apply, the better it sounds - so you need to know when enough is enough. The reason sheeple think vinyl sounds better is mostly down to the amount of compression that has to be applied to it so that it can be carved into the master.

          I honestly can't tell if you're trolling.

    This test is impossible to do because of the player :l I'm getting inconsistencies on every listen.

    I got 5/5 listening carefully and 4/5 when all I clicked was A. Is this test perhaps skewed towards the suggestion everyone would benefit from this new music service? Has anyone got 0 or 1 out of 5?

    I got 4/5 on my laptop speakers. Can tell the difference even on low end equipment

    Weird. If I pick all the ones I think are uncompressed, then I get 4/5. I suspect I don't really know what I'm looking for.

    Correction: if I pick all the ones I think are compressed, I get most "right."

    Okay, I made a semi-reliable cheat sheet:
    1: choose the duller-sounding cymbals (go figure)
    2: choose the duller-sounding clappers (go figure again)
    3: the most obvious of all - choose the wider soundstage
    4: choose the sharper cymbals (concentrate on right channel)
    5: choose the sharper voice (past halfway mark)

    I'm not sure this is a great example of compressed vs uncompressed music, as generally you'd think the main aim was to preserve the real-life nature of the recording (say with symphonies, chamber music, concertos). The Killers don't have much soundstage to begin with, and I'm certain their music has a compressed dynamic range anyway as their CDs have a wall-of-sound effect. As for the second track, it's definitely mixed from separate microphones so soundstage is irrelevant. Daft Punk, although the most synthesised, is ironically the champion of this as it consistently has a better soundstage uncompressed.

    Also, this test is clearly not dependent on your ears' discerning ability as I was able to get 4/5 consistently with practice, whereas my first and second goes were 1/5 and 0/5 respectively.

      Good point about preserving real-life nature of music.

      If they had used jazz or orchestral tracks (with wide dynamic ranges), the difference between the two lossy compression levels might have been more obvious.

      When I was choosing a streaming service I used John Taverner’s The Protecting Veil to A/B various services. There was quite a bit of difference in sound quality, the winner being Spotify’s 328kbps stream. Rdio did a strange thing where they seemed to compress the track (reducing the dynamic range) above and beyond what was in the original track, which made it sound worse than Spotify’s.

      FYI - The Protecting Veil if anyone is interested http://www.youtube.com/watch?v=5PQF3PjxS2c

      Last edited 22/11/14 9:42 pm

    Welp, I got 5/5 on my first try. I guess you really can tell the difference with reasonably high-quality sound equipment (Fiio e18 + Heir 4.Ai+'s)

      Yeah, a good DAC makes all the difference. I am using my laptop audio out and I can't even hear close to the 44.1kHz 16bit noise floor because of the hum.

        The fiio is great, partly because it also works natively with Android OTG. You can pick one up for about 130 dollars if you know where and when to look

          Wow, that's awesome. *adds to OzBargain watch lists*

            Keep an eye on massdrop.com, they often have good audiophile gear at a good price if you're willing to wait.

              Thanks for the tip.
              Fiio E17, here I come.

                if you want to support australian small business, and still get a decent price, try either noisymotel.com.au or headphones.com.au

                Both ship express anywhere in Australia and on most orders shipping is free.

                +1 on the Fiio gear - I have had several Fiio headphone amplifiers and they all sound awesome ;)

    This is a nonsense test, and anyone who has studied introductory statistics will be able to tell you so.
    To demonstrate why, I'll use an analogy:
    Imagine you were trying to guess the result of 5 coin flips. Your chance of getting any individual flip right is 50%. If you're doing 5 flips, and you're guessing (because that's really the only option you get), then the distribution of probability of getting n correct is:
    0 correct: 3.125%
    1 correct: 15.625%
    2 correct: 31.25%
    3 correct: 31.25%
    4 correct: 15.625%
    5 correct: 3.125%

    Thus, even if you're guessing you are most likely to get 2 or 3 correct. In reality, it is only with all 5 correct that you can even have a reasonable claim that you could possibly tell the difference, and even then there is a 3% likelihood of getting that outcome even if you were just guessing.

    The other interesting point is that if you get 0, that is equally good evidence that you can tell the difference as getting 5. It is evidence that you think the lo-def sounds more high-def than the high-def, but it is still evidence of being able to tell the difference (at least to the same extent as getting 5 right is evidence).

    If you wanted to do a real A/B test, then you would run each sample multiple times; ideally many times, you wouldn't have the glitch when switching between A and B which will interfere with you listening for differences, and you would test multiple levels of compression because it isn't like encoders run at just one bitrate.

    IMHO, this is a stupid test which is trying to capitalise on the fact that 50% will by chance alone would get 3 or more correct but leverages that to make people thing they have highly sensitive ears and need hi-def music. Bah, bad-science nonsense.

    (end rant)

    (Extend rant)
    Just for good measure, here's a link to What is a blind ABX test ? on HydrogenAudio so people can see how proper testing is done.

    Last edited 23/11/14 12:17 am

      Yeah, I got 1/5 and 0/5 for my first two goes. Maybe I prefer compression artefacts.

        It seems that there might have been some equalisation differences between the two, beyond what you would expect just from compression.
        (Discussed a bit hear at head-fi),

    I got 4 out of 5 with a pair of Bose QC15s - but, if anything, this little exercise educated me on the miniscule difference between lossy and lossless. This site may have been counterproductive for the company - I'm now happier with lossy :)

      To be fair, they even say on the website that it's designed more to test your equipment than your ability to tell the difference. And with higher-quality headphones, DACs and Amps, you're gonna notice the glaring contrast between lossy and lossless a LOT more.

        If anything, compression at 320v kbps was meant to be almost transparent with high end equipment. Higher end DACs should actually make it more difficult to tell the difference.

    i have the chord hugo+th900(high tier gear), basically turns takes music and upsamples it several thousand times to make it sound "better", so i only got 3/5, really difficult to tell, tried it again with on-board sound, dead giveaway, 5/5.

    Browser not supported. Obviously this isn't meant to be used with your phone.

    3 out of 5 with a $99 bluetooth soundbar from Officeworks....
    http://www.officeworks.com.au/shop/officeworks/philips-soundbar-speaker-htl2160-pmhtl2160

    Some of them were really difficult, too. All it highlighted for me was that my current music library is dreadfully compressed (and that I can't be bothered replacing 5000 songs with high fidelity music!)

    Last edited 23/11/14 1:35 pm

    4/5 on a set of Sony headphones.
    Can't find the Model Number of them, though. :\

    3/5.
    I found it really hard to tell the difference with the more 'electronic' sounding genres. The easiest musical styles to distinguish for lossy/lossless are jazz (with lots of brass instruments) and classical.

    Do we know what the compression is on the compressed tracks? If it's 128kb/s (FM Radio quality) then most people should be able to tell the difference but if it's 320 then I believe it would take a trained ear and/or good equipment. DACs won't make much difference because they'll work just as well for each sample.

      The compression was apparently AAC 320 (although the tracks were out of sync and equalised differently according to some posts on reddit)

    I got 5/5, but the differences were minute and not worth stressing over unless you get off labeling yourself as a audiophile.

Join the discussion!

Trending Stories Right Now