4 ms·
PonoPlayer: Neil Young’s new streaming device sounds no better than an iPhone
- tzs 12y ago> But here’s the catch: According to people who study human physiology, 44.1-kHz recordings provide enough upper-end treble to exceed the capabilities of our ears. Anything higher and you’re spending money for sounds your dog can listen to but you can’t. You can't hear above 20 kHz in the sense that if someone plays a 22 kHz tone and asks if you hear a tone, you won't be able to answer correctly more accurately than if you just guessed. However, that doesn't mean that it makes no difference if music above 20 kHz is present. See the paper "Inaudible high-frequency sounds affect brain activity: Hypersonic effect" [1], available for free online here [2]. They exposed listeners to gamelan music, which is rich in content above 22 KHz, sometimes with the ultrasonic sound included, sometimes with it removed, and looked at brain responses via EEG and PET. They clearly found statistical differences in brain response when the ultrasonic components were included. They also did long double blind listening tests where the participants were asked to subjectively rate the sound, and the music that included the ultrasonics was rated higher. [1] T. Oohashi, E. Nishina, M. Honda, Y. Yonekura, Y. Fuwamoto, N. Kawai, T. Maekawa, S. Nakamura, H. Fukuyama, and H. Shibasaki. Inaudible high-frequency sounds affect brain activity: Hypersonic effect. Journal of Neurophysiology, 83(6):3548–3558, 2000 [2] http://jn.physiology.org/content/83/6/3548.full http://jn.physiology.org/content/83/6/3548.full
- tmllyd 12y agoFlawed methodology and their results have since failed to be reproduced [1]. There's a run-down of the major issues with it in post #156 here [2] (all verifiable by reading the research paper). There are also various studies which essentially show the opposite results ([3], [4]). [1] http://www.aes.org/e-lib/browse.cfm?elib=12375 http://www.aes.org/e-lib/browse.cfm?elib=12375 [2] http://www.head-fi.org/t/266217/from-cd-to-sacd-how-much-of-difference/150 http://www.head-fi.org/t/266217/from-cd-to-sacd-how-much-of-... [3] http://www.aes.org/e-lib/browse.cfm?elib=14195 http://www.aes.org/e-lib/browse.cfm?elib=14195 [4] http://www.aes.org/e-lib/browse.cfm?elib=13185 http://www.aes.org/e-lib/browse.cfm?elib=13185
- tzs 12y agoThe run-down in post #156 at link is not very impressive. His first criticism is that the listeners knew the test was ABBA, with the listeners knowing that A and B are different. He thinks that is a flaw. It would be if the test were "can listeners tell the difference between A and B?". It's not when the test is "do listeners prefer A or B?". His second criticism is that there are slight differences in the spectra below 20 KHz. That's a legitimate criticism. Third, fourth, and fifth are ad hominem and similar bogus arguments. The sixth criticism is that if people do find they like the music with the high frequency components better it is because it is leading to more alpha waves and a relaxed state, which makes the people think they like the music more. He says that this isn't hearing and concludes that this somehow makes it irrelevant. This is ridiculous because it doesn't matter HOW it affects perception. If high frequencies somehow induce relaxation which makes us perceive the music more positively, presumably that would happen during a live concert. Hence, if we want the recorded music to be perceived as being as good as the live, we'd need those frequencies.
- stolio 12y agoI'm getting kind of tired of the pseudo-scientific smackdowns. I like my headphones just fine, and I couldn't tell the difference on them! So what? Those are consumer headphones, they're designed to sound _good_ not _accurate_. If you're testing two music players you might use a pair of headphones that's designed to give an accurate representation of the signal they're given. Is 192kHz a sham? Probably. But so is going lalalala nothing sounds better than an iPhone, you idiots are buying snake-oil
- quesera 12y ago> I'm getting kind of tired of the pseudo-scientific smackdowns. Were you expecting a scientific treatise on Slate? Try this instead: http://people.xiph.org/~xiphmont/demo/neil-young.html http://people.xiph.org/~xiphmont/demo/neil-young.html
- stolio 12y agoI just want people who pretend to be scientific to take themselves seriously enough that if somebody says "Your headphones suck" that they say "do they? Why?" That article is better :) 192 is silly pretty much always, 24 bits isn't very useful for playback, but a music player with better digital-to-analog converters and a better amplification stage than an iPhone? That should make a difference. An iPhone does ten million things, there should be a way to make a better music player that only does one. (edit: now that I think about it 24 bits could be useful for playback.)
- csbrooks 12y agoFrom the article: "A note from Neil tucked inside the Pono box assures you that the Pono makes your music sound better no matter what headphones you connect to it."
- jdietrich 12y agoThe PonoPlayer sounds no better than the iPhone to an untrained listener using mediocre headphones. The writer misses some important quantitative differences, because he knows very little about audio engineering. The most crucial and easily understandable difference IMO is the headphone amplifier. The iPhone has good quality digital-to-analog converters and generally sounds fine through easy to drive headphones, but it falters if you use large high-impedance headphones that need a lot of power. Hook up a pair of Sennheiser HD650s or AKG K701s to an iPhone and things go badly wrong - the bass disappears and transients start to distort, because the tiny headphone amp just doesn't have the power to cope. Yes, the audiophile world is full of bullshit, but that doesn't mean that all quality audio products are bullshit. Professional audio engineers with degrees in physics or engineering are not fools, and they do not do critical listening through $130 headphones and whatever converter and amp happens to be on their device. They use $2000 converters, $4000 monitors and $600 headphones because they are measurably better in all sorts of ways to cheaper equipment. If you're listening to music for pleasure, use whatever equipment you like. If you don't hear the difference that better equipment makes, then don't pay the extra. The only thing that matters is your enjoyment of the music. Conversely, just because you can't hear a difference, don't assume that it isn't there. Really understanding audio and being able to critically assess recordings and audio equipment takes a great deal of training and practice.
- tmllyd 12y agoYes, when discussing Pono it's important to differentiate between Pono the device and Pono the push towards higher sampling rates. The former may very well be better than other mobile devices in important ways as you mentioned. The latter is not backed up by evidence, but perched precariously on a conflation of the sound of low bit rate lossy compression and RedBook CD (which while on the edge, is adequate for delivery as far as the current evidence shows), and the naive misconception that faster and faster sampling equals better sound. There are solid technical reasons why 192 kHz sampling is too fast [1], even ignoring the evidence that shows that we don't benefit from the additional bandwidth (note the difference between modulator speed and sample rate!). There are scenarios when up-sampling an already digitised signal before certain kinds of processing is very beneficial (to reduce aliasing in the audible range), but that's an entirely separate discussion. [1] http://lavryengineering.com/pdfs/lavry-sampling-theory.pdf http://lavryengineering.com/pdfs/lavry-sampling-theory.pdf