Ars Technica reviews Pono Player

[mjb]

“Neil Young wants us to believe that higher-res audio files played through his banana-colored Toblerone will improve our music-loving lives,” Sam Machkovech writes for Ars Technica. “I’m here to say that he and his team are kinda full of crap"....

"No amount of testing made 192kHz/24-bit FLAC audio sound noticeably better than high-quality MP3s.”

ARS review here: http://arstechnica.com/gadgets/2015/...-of-snake-oil/

[Hdale85]

Yeah the device it self is nothing particularly special. You're buying into the ecosystem more or less.

[madmac]

“Neil Young wants us to believe that higher-res audio files played through his banana-colored Toblerone will improve our music-loving lives,” Sam Machkovech writes for Ars Technica. “I’m here to say that he and his team are kinda full of crap"....

"No amount of testing made 192kHz/24-bit FLAC audio sound noticeably better than high-quality MP3s.”

ARS review here: http://arstechnica.com/gadgets/2015/...-of-snake-oil/

"No amount of testing made the Hi Rez version sound better than high quality MP3's" ?? SERIOUSLY?? !!! What part of MP3 audio has ever been 'high quality'!!??

Ok.......That's impossible!!!.......unless they are hearing impaired or listening to the content on junk equipment.....That is BS!! I have NO experience with the Pono player but I have piles of experience listening to Hi Res digital audio and I can tell you that time and time again, I hear a MAJOR difference listening to Hi Res audio. Last night, I spun Chris Botti's live in Boston and I was blown away at how sweet the audio was. That disc is recorded in 96/24 bit sound and the difference to my ears (over CD quality audio) via my audio gear is immediately apparent!

Hell.....even DTS sounds better to me than Dolby digital on my system!.

Poppycock !:roll:

[mjb]

I'm sure I can hear a difference between 16 and 24 bit, despite lots of mathematical evidence to the contrary. Between 96K and 192K, I'm not so sure - despite the difference representing a shit load of data. I'm generally happy with CD quality - but less is unacceptable for serious listening. 24/96 is probably a realistic "top-end" sweet spot for HD.

I'm surprised at the ARS article though, their reviews are usually well researched, very thorough, and very accurate. But this reminds me of the Consumer Reports "Bendgate" controversy.

[Hdale85]

High bitrate MP3's properly done can sound pretty good. And what they are saying is that you gain more sound quality from good mastering rather than lossless files. Having all these "hi-res" files converted from the same masters doesn't make them any better than what's already been out there.

[madmac]

High bitrate MP3's properly done can sound pretty good. And what they are saying is that you gain more sound quality from good mastering rather than lossless files. Having all these "hi-res" files converted from the same masters doesn't make them any better than what's already been out there.

I'll agree with the good mastering comment. And like Mike mentioned above, a well mastered CD is perfectly fine with me. Most Hi Def digital recordings are very well mastered and that is because they care enough to do this and present it with Hi bit rates and sampling to sound their absolute best........And in most cases, they do! I even have a few HDCD's that sound just awesome! MP3's, even at their highest quality levels are too compressed and the music loses too much texture, especially in the higher frequencies.

[Ovation]

Mastering far outweighs the delivery format in determining a good quality listening experience. So too does a proper level matching of the various sources. It has been irrefutably established by hundreds of experiments that something that is louder--and it doesn't have to be by a significant amount--is invariably perceived as "better" as an initial impression. This is one of the main factors that drives the "loudness wars" in the mastering of popular music today (resulting in very little dynamic range). There is an analogous effect with video, where colours that "pop" are initially perceived as more impressive, even if the colours are clearly "off", than a more natural colour tone/palette. In each case, over time, greater dynamic range for audio and more natural colours for video, for those who bother to experience them, displaces the initial preference for "loud" and "colour pop", respectively. Sadly, few people bother to try to adjust to the more naturalistic audio and video settings and marketing departments know this--hence the continued lack of dynamic range in music and the "vivid" settings on video displays.

I have tried level-matched comparisons of various formats (I have a friend with an impressive home recording studio setup that allows such comparisons with ease and a high degree of accuracy). I've tried sighted and blind comparisons. The results are always the same--audio with wide dynamic range (with as little compression at the mastering stage as possible) sounds better than severe dynamic compression. This is true regardless of, and across, formats. For example, a high bit-rate MP3 (or its iTunes equivalent), with a broad dynamic range will sound better than a 24bit/192khz source with a limited dynamic range. Thus, depending on the mastering of the source material, it is quite possible for the audible difference between an MP3 and a 24/192 format to be indistinguishable in a comparison.

Much of the confusion lies in the use of the word "compression" to label two distinct procedures, only one of which has a constant effect on sound quality. There is dynamic range compression (at the mastering stage) and digital file compression (at the format storage stage). The first one is most responsible for degraded sound quality across all formats. Severely limited dynamic range (dynamic range compression) remains a major flaw, regardless of the final delivery format (lossy, lossless, hi-res, PCM, DSD, etc.) If the master is compressed in that fashion, it makes little difference how you deliver the audio. On the other hand, digital file compression has a far less effect on audio quality, though there are compression limits that eventually compromise audio quality rather dramatically. High bit-rate MP3 or the Apple equivalent (AAC) delivers (assuming a properly mastered source with broad dynamic range) an excellent degree of audio quality. If we drop down to 128 or lower Kbs, then we start to experience a noticeable degree of high frequency "smearing" (the kind of "metallic" distortion I notice in my wife's car when she tunes into satellite radio--cymbals and metal string acoustic guitars sound especially bad). This is often exacerbated by the dynamic range compression at the mastering level so we suffer the twin effects of too much compression. But while I would not go so far as to say it is impossible to tell high bit rate AAC from a lossless file (with very careful and attentive listening on my gear, with an appropriate attenuation of ambient noise, I can tell the difference), I would wager that fewer than one percent of listeners could successfully tell the difference in a level matched comparison. Certainly not if one is doing anything other than devoting one's full attention to listening to the audio in question.

In the end, my motivation for purchasing hi-res releases of music has revolved around two basic ideas. One--it is usually the only way to get discrete MCH audio presentations (without video) and I always favour a MCH version if available. Two--Hi-res releases are often given a better treatment at the mastering stage than non hi-res, so even if the format has relatively little to do with the audio quality (at least compared to the mastering stage itself), I'm more likely to get a better mastering effort from a hi-res release.

Just my 2 cents.

[wkhanna]

Your 2 cents is worth a million bucks, IMHO.

[madmac]

Mastering far outweighs the delivery format in determining a good quality listening experience. So too does a proper level matching of the various sources. It has been irrefutably established by hundreds of experiments that something that is louder--and it doesn't have to be by a significant amount--is invariably perceived as "better" as an initial impression. This is one of the main factors that drives the "loudness wars" in the mastering of popular music today (resulting in very little dynamic range). There is an analogous effect with video, where colours that "pop" are initially perceived as more impressive, even if the colours are clearly "off", than a more natural colour tone/palette. In each case, over time, greater dynamic range for audio and more natural colours for video, for those who bother to experience them, displaces the initial preference for "loud" and "colour pop", respectively. Sadly, few people bother to try to adjust to the more naturalistic audio and video settings and marketing departments know this--hence the continued lack of dynamic range in music and the "vivid" settings on video displays.

I have tried level-matched comparisons of various formats (I have a friend with an impressive home recording studio setup that allows such comparisons with ease and a high degree of accuracy). I've tried sighted and blind comparisons. The results are always the same--audio with wide dynamic range (with as little compression at the mastering stage as possible) sounds better than severe dynamic compression. This is true regardless of, and across, formats. For example, a high bit-rate MP3 (or its iTunes equivalent), with a broad dynamic range will sound better than a 24bit/192khz source with a limited dynamic range. Thus, depending on the mastering of the source material, it is quite possible for the audible difference between an MP3 and a 24/192 format to be indistinguishable in a comparison.

Much of the confusion lies in the use of the word "compression" to label two distinct procedures, only one of which has a constant effect on sound quality. There is dynamic range compression (at the mastering stage) and digital file compression (at the format storage stage). The first one is most responsible for degraded sound quality across all formats. Severely limited dynamic range (dynamic range compression) remains a major flaw, regardless of the final delivery format (lossy, lossless, hi-res, PCM, DSD, etc.) If the master is compressed in that fashion, it makes little difference how you deliver the audio. On the other hand, digital file compression has a far less effect on audio quality, though there are compression limits that eventually compromise audio quality rather dramatically. High bit-rate MP3 or the Apple equivalent (AAC) delivers (assuming a properly mastered source with broad dynamic range) an excellent degree of audio quality. If we drop down to 128 or lower Kbs, then we start to experience a noticeable degree of high frequency "smearing" (the kind of "metallic" distortion I notice in my wife's car when she tunes into satellite radio--cymbals and metal string acoustic guitars sound especially bad). This is often exacerbated by the dynamic range compression at the mastering level so we suffer the twin effects of too much compression. But while I would not go so far as to say it is impossible to tell high bit rate AAC from a lossless file (with very careful and attentive listening on my gear, with an appropriate attenuation of ambient noise, I can tell the difference), I would wager that fewer than one percent of listeners could successfully tell the difference in a level matched comparison. Certainly not if one is doing anything other than devoting one's full attention to listening to the audio in question.

In the end, my motivation for purchasing hi-res releases of music has revolved around two basic ideas. One--it is usually the only way to get discrete MCH audio presentations (without video) and I always favour a MCH version if available. Two--Hi-res releases are often given a better treatment at the mastering stage than non hi-res, so even if the format has relatively little to do with the audio quality (at least compared to the mastering stage itself), I'm more likely to get a better mastering effort from a hi-res release.

Just my 2 cents.

Very well said Ovation. However, I Always notice an immediate difference in audio quality when listening to a Hi Rez audio source. There's got to be a reason for that no? I find the audio warmer, sweeter and more textured with more depth. It may however, have more to do with the mastering than the Hi bit rate it is pressed with. However, I will happily take both whenever I can get it.

[Ovation]

The way to be sure you can always hear a difference is to listen to the exact same recording in the different formats. With my friend's studio, I was able to do just that. We recorded my friend (an excellent musician) at 24/192 and then mastered at different compression rates. We then made different formats of the different rates (example: hi bit rate AAC, 16/44.1, 24/48, 24/96, 24/192). With the very narrow dynamic range version (6db), no discernible difference between AAC lossy at 256kbps and 24/192--even with very revealing headphones. With medium and wide dynamic ranges, the different formats became easier to tell apart, though AAC always sounded quite good (certainly more than acceptable for non-focused critical listening). Our conclusion was that hi-res, well-mastered, had its place for critical listening on revealing gear, but was a waste on very highly compressed masters. Since, sadly, most popular music since at least the mid-90s is plagued by narrow dynamic range, unless one is sure a hi-res reissue has been remastered with much wider dynamic range, in most cases the AAC version is sufficient for 2 channel mixes. (MCH mixes, of course, usually require the purchase of a hi-res version, even if it may well be plagued by too much dynamic compression for maximum audio quality).

[mjb]

Much of the confusion lies in the use of the word "compression" to label two distinct procedures, only one of which has a constant effect on sound quality. There is dynamic range compression (at the mastering stage) and digital file compression (at the format storage stage). The first one is most responsible for degraded sound quality across all formats. Severely limited dynamic range (dynamic range compression) remains a major flaw, regardless of the final delivery format (lossy, lossless, hi-res, PCM, DSD, etc.)

Quoted for emphasis, I believe there is a lot of truth to this.

With the very narrow dynamic range version (6db), no discernible difference between AAC lossy at 256kbps and 24/192--even with very revealing headphones.

Well, the detail has already been thrown away, you can't polish a t**d as the saying goes!

The real enemy is "dynamic range compression", not the "delivery format" per se, and most often we (consumers/audiophiles/listeners) have little or no control over DRC (Dynamic range compression, not The Democratic Republic of the Congo).

Pono is a way to get access to to the original uncompressed master tapes, well mastered or not, but we at least have a chance of hearing a difference. The delivery (digital file) format is largely irrelevant, as long as the dynamic range is preserved. In order to reproduce this dynamic range, and therefore to hear any benefit, we're going to need a 'low noise floor' system.

I also prefer AAC if I have to go lossy. From wikipedia: Designed to be the successor of the MP3 format, AAC generally achieves better sound quality than MP3 at similar bit rates.

[madmac]

I guess my question now would be........Why do studios record music with narrow dynamic range and what do they perceive as the benefit to doing this? If it's worse, than why not make it better and press it with a wider dynamic range?

I still contend that a 24 bit source recording presented in 24 bit HAS to sound better, simply because it contains utterly millions of pieces of more audio information per second! How could it NOT sound better? One argument would be (And I believe the creators of CD said this) is that the human ear is simply not good enough to discern the difference??!!!

[Ovation]

I guess my question now would be........Why do studios record music with narrow dynamic range and what do they perceive as the benefit to doing this? If it's worse, than why not make it better and press it with a wider dynamic range?

The benefit is twofold, though neither involves superior sound quality. The first is a marketing benefit. Years ago, when people still bought their music exclusively in stores (no downloads yet), store managers would fill a carousel CD player with titles they wanted to push and let them play in the background. Sometimes, bands (and their sound engineers) would play around with dynamic range compression in order to get a particular "sound". Those with very little dynamic range played more loudly than those with wide dynamic ranges. Marketing people noticed this effect in music stores and also noted that people paid more attention to the louder tracks. Thus the loudness wars began. The way to make your disc stand out was to make it louder (more compressed). The competition jumped into the fray, and so on. Moreover, as I noted above, numerous studies demonstrated that "louder" is overwhelmingly perceived as "better" as an initial impression of any audio track (well, there is an upper limit, but we're not talking "ear-bleedingly loud"). And "initial impression" is what marketing is all about. So marketing tells management "louder (more compressed) sells", management puts pressure on bands (tells them it sells better is a good incentive as well--and many musicians are not the keenest arbiters of high quality sound recordings, as they are much more interested in "live music"). And so on.

The other benefit, if you can call it that, is for mobile listening. Listening in the car, on earbuds on the go (trains, subways, buses, etc.) and so on, a more compressed dynamic range avoids moments where the music is drowned out by ambient noise. When music became portable on a personal level (radio was already compressing its signal for the same effect), releases followed suit.

And let's face it--most people consider music a background filler for some other activity. The number of people in society as a whole who actually set aside time to sit and listen to music, without doing anything else, is vanishingly small. Hanging around fora like this may create a false impression otherwise, but a quick check with anyone's friends, family and acquaintances will quickly dispel this false impression.

I still contend that a 24 bit source recording presented in 24 bit HAS to sound better, simply because it contains utterly millions of pieces of more audio information per second! How could it NOT sound better? One argument would be (And I believe the creators of CD said this) is that the human ear is simply not good enough to discern the difference??!!!

The human ear argument is actually quite powerful. Virtually no one in their 30s and beyond can hear the entire 20-20000hz spectrum. Certainly not since the early 20th century, owing to our increasingly noise-filled environment (and declining hearing as we age). Additionally, the algorithms employed in high bit-rate lossy audio compression schemes today are quite sophisticated. There remain measurable differences, but their audibility varies widely (and far less, overall, than most people are wont to admit).

I have music recordings for which I consider the hi-res format in which I own them essential to fully enjoy them. But they are a small subset of my entire music collection. Owing to the vagaries of things over which I have no control (in chief, mastering decisions in the studio), much of what I own cannot be audibly enhanced by a hi-res format (not without a remastering to go along with it). And, in the end, while I always try to get the best quality recording of any music I want to own, I temper my purchases based on whether I believe the hi-res release will be audibly beneficial. If not, I same some money and buy other music.

Lastly, I make my listening choices based on the music itself first, and the quality of the recording later. I have some vintage 1930s 78s, the sound quality of which is, frankly, a touch below mediocre, but I love the music and there's no other way to hear it. The music matters most.

[madmac]

I'm pretty picky about my sound. The recording has to really good for me to sit down and critically listen to it on my system, regardless of how good the material is. If it sounds bad, I'm not going to listen to it. Background listening while cooking or doing the dishes is another matter........I just put anything average sounding on at a lower volume and let it fill the air with ambient sound. I won't do this with good recordings as I feel like I'm wasting it and missing out on not sitting down and listening to it.........kinda' like missing part of a good movie cause you had to go to the bathroom!

But you are right Ovation.........we are probably representative of 1% of the population!

[JonMarsh]

Before all this arm waving and indignation, look at the testing methodology. Earbuds and cheap headphones. That would be like evaluating a $1,000 CD player against the TotalDAC-D1 Dual using a set of small cheap logitech computer speakers. Yes, I can hear the difference... heck, I did that comparison with a friend's CD player against my Berkeley Alpha DAC, on a hundred dollar set of Logitech speakers.

The other potentially valid point is that $400 is not going to buy you an amazing clock and DAC, even if it is designed by Charles Hansen (who worked for me when he was a physics student at CU- good friends for a long time). It is going to buy you the ability to playback High Res files after a fashion, but again, if we fall back to the analogy of comparing a $400 table top DAC to a Metric Halo or Berkely Alpha DAC, we all know that that $400 DAC is not going to provide optimum playback of even CD material, much less more demanding high res, where jitter counts even more. The limitation in playback between a Pono and a current iPhone is more likely to be the headphones for most source material. (BTW, I do have

Let's get real, here, folks... of course, the dolts at Ars, who ARE good at some times of technology analysis, haven't a clue about high res audio. Neither does Pogue at the NYT. OK, that probably comes across as harsh and arrogant, but let's look at their actual experience and background and what they've done in the past.

It reminds me of a time that Gizmodo or one of the other self appointed tech sites sent someone out to interview Michael Fremer, with the intent of doing a hit piece, and the guy was blown away not just be Fremer, but by how good his system sounded- he'd never heard anything remotely like that. So the guy wrote what he saw and heard, and well and truly pissed off his editor, who was looking for and expecting a very different type of piece. Spin in media? You betcha! Unbiased investigative journalism? No such thing.

People often don't know what they don't know... and until it's rubbed in their nose/face, they keep preaching to the world like they've seen it all.

BTW, I tried to buy a Pono player online from their website, but when you click on checkout, it takes you to a page requiring a salesforce member log on. what the?

Ok, that's enough of my curmudgeonly rant for this fine Thursday morning...

[madmac]

Well said Jonmarsh !! HA !!

[wkhanna]

..... Neither does Pogue at the NYT....

Funny that I was just reading THIS minutes before I came here.....

[PewterTA]

This is one of the best threads on the internet about all this...just for the simple fact that in such few posts, everyone hits the nail on the head. I wish more people would read this thread an understand what the differences between hi-res, compression, and what they are actually hearing really is!

Great post everyone!

I would have to add that most of the reviews and what people think on things, they are all biased, they like what they like. Me and the rest of us in the Pittsburgh region who hang out have our biases towards things. The ONLY thing I think is different with us (and why we all enjoy hanging out) is while we have biases, we are all free thinkers and will admit (begrudgingly) that if something we think is wrong, that we can accept that and even CHANGE our opinions on things! We aren't stuck in our little niche can will never admit if something is better than what we have or have heard. We are all in it for the love of Audio. So if there's something new, great, lets experience it and enjoy the revelation. But we also know when we see someone is a snakeoil salesman just trying to convince people of whatever they have.

It's funny how the bigger our group has gotten we all have to admit how much more alike all of our systems have started to become. There's differences, but they all have a very similar feel and sound to them. Maybe all of our cycles are syncing or something......???!!????

[Hdale85]

That's exactly right Dan, everything was very open and fun in our get togethers lol. I'd love to start making it to them again as I certainly do miss them.

[JonMarsh]

Now perhaps I should start a new thread, but I thought it easier just to add on comments about Stereophile Reviews the Pono music player in the first thread, and keep things consolidated.

The reviewed performance is about what I'd expect... I think Ayre did the right thing at this price point, and focused on one performance aspects that's often problematical on other players- the output impedance of the headphone drive circuit is commendably low, at 3 ohms across the band. This is better that most other portable players, including most of the Astel&Kerns models. This probably does more for the sonic presentation than any other factor for most users. (BTW, iPhone 5 and 6 are about 4.5 ohms, which is also much better than most portable players)

One of the promotional statements from Pono is "This Portable audio player uses circuitry taken straight from Ayre's own top-of-the-line products, costing tens of thousands of dollars, for unparalleled sound quality and unrivaled listening pleasure". I really wish they'd inserted the word "some" before circuitry...

On the digital side, pluses are that the digital filter is the minimum phase type used by Ayre in their table top players, with it's minimal ringing apodizing filter.

On the minus side, with 44.1 kHz source material, the slow roll off results in poor rejection of early imaging products. To quote, "As a result, the image of the 19.1 kHz tone at 25 kHz is suppressed by only 6 dB or so, (44,1000 - 2 x 19,100), and an alias tone at 5.9 kHz (44,100 - 2x 19,100) appears in the audio band, though this lies at a probably inconsequential -74dB (0.02%).

In going from 16bit data to 24 bit data, the noise floor drops by 7-8 dB, putting it at about 17 bits of real resolution. For a battery powered unit with lowered maximum output level compared with the CD spec, this is pretty good performance, but it's not going to make owners of NAD M51's or Auralic Vega's lose any sleep at night (as they have real resolution based on working noise floor and digital linearity of about 21 bits). The classic test with 16 bit data of an undeterred 1 kHz sine wave at -90dBFS, which should resolve to the three level signal (zero, and the plus and minus values, as it's toggling the LSB's) is obscured by high frequency noise.

Let's put this in perspective- there has been a Playback Designs MPD-5 DAC which is supposed to support high res PCM and DSD, a table top unit, which is both a disk player and processor, and retails for $17,000, also obscures the -90 dBFS 1 kHz LSB steps, but with DSD or 24 bit data the noise floor is only 3 dB lower; what's more the left channel was enough noise than the right that with high resolution data, the left channel was noise than the right with CD data.

Also, in portable use, the ambient noise may often be fairly high, and headphone quality is a big factor- forget about ear buds and noise canceling headphones with digital processors, you really need some higher end cans to get the most out of even what the Pono player can do.

I think the Pono player is a good argument for an uncompressed music player for portable use, but I'm less sanguine about it's high resolution capabilities, especially considering the storage limitations of a portable player- 64GB internal, and 128GB total with a high capacity micro SD card added in. Considering 176.4 sampled albums easily run to 4-5 GB, plan on only carrying a few of your favorites around in AIFF, though FLAC will help reduce that by 30-40%, depending. I would really be happy to have something with a bit more juice and capacity, and would happily give up the small size- even something the size of a thick (think old iPod thick) device with the footprint of an iPhone 6+ would be fine with me.

BTW, if you want to be pleasantly surprised or shocked, try some uncompressed AIFF or get a FLAC player for iOS and try it on an iPhone 5 or 6 with a typical set of good Audio-Technica headphones like the Audio-Technica ATH-M50x (about $450). You may already have the high resolution music player you need in your pocket already...

[JonMarsh]

Another minor info tidbit, interesting to me and perhaps a little arcane for everyone else- in a recent blog I found out that Ayre did use the diamond driver stage for the headphone amp, so with increasing level, Charlie claims the impedance of the headphone stage actually goes down. Now, for my money, as portable players go, that may be the most significant advantage sonically, is the low overall output impedance and headphone drive available.

The VX-5 and AX-5 use the diamond output stage, and from my own experience testing it for a pre-driver stage in a pro amp design back in the late 70's, it does work as advertised- for a consumer level circuit and with modern transistors, it should work very well.

[wkhanna]

I would love to hear what you think of the circuit topography of the new DAC from Schiit.
I have a thread on it HERE

[JeanM]

Good posts above, you would think it would be basic/common knowledge about the "only as good as the source" saying

[wkhanna]

Audiophiledom is not immune from the oxymoronic phrase "Common Knowledge"