The Pono Player and Promises Fulfilled
Ed Note: This is going to be a subjective review...or as close as I know how to get. There is a ton of information out there on the technical details of the Pono Player, I'm not going to repeat them here as I really want to talk at length about my listening experience with it. I'll refer you to Michaels Lavorgna's AudioStream Pono review for a comprehensive look at the Pono Player, and John Atkinson's Stereophile Pono Player review, which also includes measurements.)
I've decided to take this approach because the experience I've had with the player demands it. Also because so much of Pono's marketing promise is based on the subjective experience of the player and high-resolution files, and because so much of the criticism leveled at Pono has to do with the audibility (or not) of improvements in performance relative to other players. This is my personal subjective experience with the Pono Player.
By the time the Pono Player arrived I was already sick of it. What a ridiculous curfuffle; it was as if the audio press had decided to play an excruciatingly convoluted game of ideological rugby with the little yellow Toblerone bar. This was not a product introduction, it was an open invitation for objectivists and subjectivists alike to tune up their well-worn talking points and go to war. Let's just say that by the time I was sliding open the top of the bamboo presentation case my desire to add my voice to the fray was closing in on zero. My expectation bias meter was barely bouncing off the bottom end peg labeled "Meh."
Well...that's not entirely true, I did have a couple of expectations that may have biased my first listening sessions.
I'm well aware of Charlie Hansen and Ayre Acoustics and am of the general opinion they make very fine DACs. I do have an older version of the QB9 USB DAC here, and I did use it for a number of years as my reference DAC. It has a very smooth and musical character in my experience. We'll chalk that up as a positive bias.
I was also well aware of the promises that Neil Young was making about the Pono ecosystem as a whole and the Pono Player in particular, and my feeling is he way over-promised the benefits of high-performance gear and high-resolution file formats. Apart from flying bands to play live in my own living room, I don't see how Pono would be able to deliver on the stunning improvements their PR campaign was expressing. The Pono hype was way too strong for my liking. We'll chalk that one up as a negative expectation. In my head there was no way the Pono Player could live up to Neil Young's hype.
Oh well...I went through the motions. Charged up the player; connected it to a computer and just dumped a bunch of test tracks and albums for listening on it (which was fairly easy, but a little slow in the transfer); got out my NAD VISO HP50s; and started listening as I wrote one of the "Headphone Measurements Explained" articles.
I had started with a few very familiar test tracks and quickly determined they sounded pretty good and I should just move on to some casual listening as I wrote. I pulled up an old favorite, Lyle Mays' (keyboard player on many of Pat Metheny's albums) "Street Dreams." Three or four tracks into it I realized that my eyes were closed and my fingers had been hovering over the keyboard unmoved for the last ten minutes...I had been sucked into an old favorite and was way inside this particular album again for the first time in a long time.
I shook it off, and tried another old favorite; this time Carmen McRae's "Carmen Sings Monk". First cut on the album is a live recording of "Get it Straight." There's an instrumental break in the middle of this short 3:55 tune, and a little bass solo that starts at 2:12 that has a spine tingling run of notes starting at 2:28 in which you can hear audience members gasp at the crescendo of notes and intensity. I was swept right along until, as the song wound down, I opened my eyes to see, yet again, my fingers hovering motionless above the keyboard.
"Damit, I'm not going to get anything done with this Pono gadget playing in my ears."
And then it hit me, the Pono Player was delivering exactly what it promised: a deep connection with the music. It pissed me off. I'm a wizened and experienced reviewer with large dollops of skepticism ready to be brought out at a moments notice; I'm not s'posed to be swept off my feet some three-sided marketing gimmick. What the hell is going on!?
I decided to drop Pono Player designer Charlie Hansen a line.
Inside the Pono Player
I told him of my experience with the distractingly musical Pono Player, and asked him what he thought were the technical characteristics that lead to that experience. Here are some excerpted comments from Hansen:
a) Brickwall filtering creates massive time smear. b) The human ear/brain is already known to be exquisitely sensitive to time smear. c) DBT and AB/X are really only sensitive to differences in frequency response. Using these tools for anything to do with music is like pounding a nail with a screwdriver. Ain't gonna work.
Specifically, one of the massive benefits of a higher sampling rate is not extended bandwidth. Instead, it allows for gentler filters to be used. In the case of the Ayre QA-9 A/D converter, the anti-aliasing filters have zero ringing or time smear for double and quad sample rates. (Only one cycle of ringing for single rates -- something has to give somewhere...)
When Ayre designed the PonoPlayer's audio circuitry, we held back nothing. We gave it everything that could fit within the constraints of the budget, physical space, and battery life. Every single secret we discovered went into the PonoPlayer. The digital filter is taken directly from our own products.
I tend to be a black-and-white kind of guy. I don't buy into this mythical "perfect amount of feedback - not too much, but not too little". If feedback is good, then use 1,000,000 dB of it. If it is not good, don't use any.
EVERYTHING from DAC to jacks is DC coupled. No coupling caps anywhere.
Everything is TRULY balanced from the DAC chip all the way to the output jacks. There is no virtual ground needed, as we have true +/- rails from the switching power supply. The raw rails go to SUPER low noise regulators, of which there are a TON.
The audio circuitry has their own dedicated +/- regulators. All of the digital circuitry runs off of positive voltage only, but three or four separate dedicated regulators there -- one for the audio master clocks, another for the digital side of the DAC chip and a third for the rest of the digital circuitry.
NOBODY builds portable players that are fully-discrete, fully-balanced, and zero-feedback. This all makes a huge difference.
To sum up, Hansen claims the sonic performance of the Pono Player is due to:
- Gentle digital filters with no pre-ringing.
- Discrete components.
- Zero negative feedback.
- No capacitors in the audio path.
- Audio circuitry with dedicated +/- supply regulators.
- Balanced headphone drive.
I like the idea of all the above items, but somehow the answer wasn't satisfying. What was it about the sound that I found so appealing? I decided some blind testing was in order.
I've put together a fairly simple system for blind testing sources. One of the most important things you need to do with blind testing is to set all the output levels exactly the same. To do so, I listen to a variety of my test tracks and set a good level on one of the players. I then play a 500Hz test tone on that player and measure the output voltage. I then set the same voltage on the two other players using the same test tone track. I use a passive three-way switch to route the outputs from the three sources to a pair of headphones.
The trick is I use a short headphone extension cable before the 3.5mm to RCA cable that goes into the switch. (All cables were relatively inexpensive, but identical.) The cables are long enough, and tangle enough, that it's impossible to visually tell which cable is going to what source on one end, and which switch input on the other. By simply grabbing the bundle of wires with hands to either side of the coupling connectors I can pull all three cables apart, and then, without looking, I scramble the cables in each hand. When I look back to reconnect the cables I have absolutely no idea which is which. I can then switch between the three sources blind.
The one flaw in this system is that the players take differing amounts of time to loop back to the start of the track, so I would have to take the headphones off when the end of the track came.
In my experience there are two ways to go about blind testing: objective evaluation and subjective experiencing. In objective evaluation you try to listen carefully for differences in the signal you are hearing. With subjective experiencing you don't evaluate the music but rather evaluate your experience while you are listening for enjoyment. The problem with objective evaluation is that you may end up knowing what sounds different between each player, but you'll usually not have a clear idea which you'd actually prefer. And the problem with subjective experiencing is you might know which you prefer, but you are likely not going to know why. It's kind of like the Heisenberg uncertainty principle: you can know where something is or how fast it's moving, but you can't know both simultaneously.
Most of my blind testing with the Pono was of the objective type, I wanted to know if there were any identifiable characteristics that were different than the other players (and sources) I had. (Astell&Kern AK100, AK100II, and AK240; Fiio X5; Samsung Galaxy Note II; Apple iPad latest generation; Auralic Vega and TaurusII; and Simaudio Moon Neo 430HA.) Most tests were done with Audio Technica ATH-MSR7 (reviewed here), but I also used the Philips X2; NAD VISO HP50; Sennheiser HD800; Westone ES5; and Jerry Harvey JH13FP.
I did about six hours of blind testing over the course of three days testing the Pono against a variety of other sources, and sometimes with various headphones. Sessions usually proceeded like this: First attempt with new devices would start me getting familiar with the new sounds. I would often misidentify the Pono in the first test. Second test would be about 50/50 getting the Pono right. Usually by the third test I would be able to rightly identify the Pono Player and by the fourth test could identify all three players and would generally get it right four out of five times thereafter.
By the time the third day of listening was done I hated that little switch box. When I said above that once I got used to the sound of the three pieces of gear in the test I could identify them correctly four out of five times I need to add that it was usually only after I had sat with my forehead on the table for 5-10 minutes flicking the switch between sources. I could do it, but it's not easy. The point is that the differences are subtlebut, in my opinion, important. More on this later.
Pono Player Vs. High-end Portables
The first test I undertook was the Pono Player against a couple of high-end portables: Astell&Kern AK240 ($2499) and AK100II ($899). Obviously these players are much more expensive than the Pono Player, but Pono's claim of a superior sounding product warrants comparisons with the best available. In this part of the review I'm going to limit myself to comments about sound quality and will address the relative value of features later (these players are extraordinarily feature rich).
Once able to reliably identify blind these three players, I found the Ak100II hit harder in bass, and the treble level was about same but sounded slightly dirtier and less refined than the Pono Player. The AK240 was more dynamic and hit harder in bass, and the treble was slightly more emphatic than the Pono.
Though the Pono Player might sound slightly rolled-off compared to the AK240, I also felt it's treble had superior resolving power and was less, what I can only describe as, digital sounding. High frequency transients with the AK240 seemed to have a slight additional "ting" sound to them making them seem more artificial compared with the more organically smooth but resolving sound of the Pono.
In this test the AK240 was the easiest to pick out with it's bass and treble emphasis fairly obvious. The Ak100II was a bit more difficult as the treble levels seemed similar and the slightly dirtier sound compared to the Pono a bit harder to identify. The increase in bass punch was usually the tell-tail sign.
Pono Player Vs. Consumer Handheld Devices
This test pitted the Pono Player against a Galaxy Note 4 and an Apple iPad Air. This was actually the hardest test to get a reliable grip on identifying the various players. Primarily, I think, because the tonal balance and dynamic authority was very similar between the three. Where in the previous test it seemed the high-end players had good power supplies and output amplifiers giving them stronger bass response and, in the case off the AK240, strong treble articulation, in this case the two handheld devices had about the same amount of bass response and dynamic puch as the Pono, and the treble level was similar as well.
Because differences were so small I had to rely more on a subjective approachjust relaxing and sensing how the music was affecting me. Usually it would take me 5-15 minutes of listening and slowly switching back and forth between sources before I could determine which was the Pono Player. It would happen all of a sudden in a moment of cognitive groking when I just felt "that's the musical one" and would stab at the Pono's pause button to see if I was right. The music would stop in my headphones and I knew I had got it, but there was always surprise and relief when it did because there was no objective characteristic identified. Though I could reliably tell which was the Pono, I did not get to the point in this test where I identified the difference between the iPad and Note 4. It's my feeling that your average contemporary hand-held device is pretty darn good these days.
Pono Player Vs. High-End Gear
My last test was to compare the Pono Player with some high-end gear. In this case against the Auralic Vega DAC ($3499) and Taurus II ($1899) headphone amp, and the Simaudio Moon Neo 430HA ($4300 w/DAC). In this case it was very difficult to tell the difference between the devices by the treble responseall were articulate and natural sounding. Even though it wasn't the identifying characteristic, by the end of this test I did feel that the Pono upper-treble was slightly lower in level than the other two.
The place where the difference was obvious was in the dynamic impact of the bass and mid-range. Where the Pono sounded sweet but slightly loose, the two other devices were tight, punchy, dynamic, and simply a more faithful reproduction of the original. I did not attempt to peer into the differences between the Auralic and Simaudio products as I will leave that for a future review of the Moon NEO 340HA.
Blind Testing with Mutli-Balanced Armature Driver In-Ear Monitors (BAIEMs)
As I mentioned previously, most of the above testing was done with an Audio Technica ATH-MSR7 headphone with a nominal 37 Ohm impedance. I next did a little blind testing with two custom in-ear monitors, the Jerry Harvey JH13FP and Westone ES5.
Now, this is going to get a little complicated for folks who are just coming in here to read a Pono review and aren't audio enthusiasts who have some familiarity with these kinds of issues, so I'll try to make this simple and quick. What I'm talking about here is in-ear monitors that have multiple balanced armature drivers. They work kind of like speakers with woofer, mid-range, and tweeter drivers and a crossover between to put low, mid-range, and high-frequency components of the incoming signal on the appropriate drivers. The problem with these types of headphone is that they tend to have very low impedances and will have very strong swings in impedance due to the reactance of the various driver and cross-over elements. In the plots below, the green line is the impedance plot and the scale in Ohms is to the left.
As you can see their impedance swings all over the place. The K3003 has a rather simple impedance plot that goes from about 8 Ohms to 16 Ohms; the Shure SE846 is more complex and ranges from 16 Ohms to 5 Ohms; the Westone ES5 is a little higher overall but swings wildly between 20 Ohms and 50 Ohms, and the Jerry Harvey JH13 also sees significant swinging between 10 Ohms and 43 Ohms. The reason I want to show you these plots is that they can react with the output impedance of the amplifier driving the headphone effectively changing the EQ of the headphone. The general rule of thumb is that the amplifier should have less than 1/10th the impedance of the headphone in order to keep the interaction between the amp and driver below an audible level. This ratio, of 10:1 in this case, is called the damping factor.
I measured the output impedance of all the devices used in blind testing. They were: Pono Player, 3.27 Ohms; AK240, 3 Ohms; AK100II, 3.47 Ohms; Fiio X5, 0.52 Ohms; iPad Air, 0.8 Ohms; Galaxy Note 4, 0.9 Ohms; Auralic Taurus, 0.9 Ohms; and Simaudio Moon Neo 340HA, 0.35 Ohms.
It's my general feeling that a headphone amplifiers for general portable use should have an output impedance of under one Ohm. This is easy to achieve if you use negative feedback, but since Hansen doesn't want any negative feedback there's a limit to how low you he could make the output impedance of the Pono Player.
The point I'm working my way toward here is that if a 10:1 or better damping factor is desirable, and if in the case of BAIEMs we're looking at a type of headphone that commonly has an impedance swing below 10 Ohms, then you'll really want the output impedance of the amplifier well below 1 Ohm in order not to have any tonal changes. Obviously the Pono and the Astell&Kern players have too high an output impedance, and will suffer some tonal changes when using multi-balanced armature earphones. How much tonal change? I've prepared the following two plots that calculate tonal changes for four BAIEMs with the Pono Players 3.27 Ohm output impedance and the iPads 0.8 Ohm output impedance.
Worst case changes for the Pono (upper plot) are the Shure SE846 that has a wide dip that produces a roughly 3dB reduction in level centered in the mid-treble. The Jerry Harvey JH13 the shows a rise in the same region of about 2dB. The same headphones when driven by the iPad's 0.8 Ohm output impedance (lower plot) exhibit only 1dB or less of frequency response change for the same BAIEMs.
When I did blind tests with the JH13 and ES5 I was readily able to tell when I was listening to a high output impedance device or a low output impedance device.
It's worth mentioning at this point that the BAIEMs we're talking about here can typically range in price from $800-$2000. With that type of investment one should take care to find a player with a very low output impedance, or it may be worthwhile to consider a separate low output impedance headphone amplifier after the player for quality listening.
Pono and Difficult Headphones
It's worth noting there are a number of low efficiency headphones out there. For example the Mr. Speakers Alpha Dog and Mad Dog; HiFiMAN HE-6, and Beyerdynamic T1 (among numerous others) all require more than 200mVrms to achieve 90dBspl at the ear. When I tested the Pono with headphones that are difficult to drive I felt the sound tended to slide from musical to a bit mushy. It's simple really, the Pono makes all it's power from a single 3.7V batteryit's not a desktop unit with unlimited juice available from a wall socket. In a portable design something's got to give, and in the case of the Pono Player, once headphones become hard to drive it begins to loose it's sonic composure.
In the case of tough headphones, it is possible to use the balanced drive ability of the Pono Player that will effectively double its driving power (more about balanced drive shortly) but the fact remains that at some point the Pono Player (and indeed most portable digital audio players) will need help from an outboard portable headphone amplifier.
Here is a way to check whether a particular headphone will mate well with your Pono Player. First go to the InnerFidelity Headphone Measurements Datasheet Download page. Find the headphone you're interested and click on it to bring up a headphone measurement .pdf. It will look something like this.
First look at the impedance plotit's the purple plot circled in red in the second graph down on the left. If this plot is above 30 Ohms you are good to go. In the plot above, you can see that this headphone is actually below 30 Ohms. In that case you want to make sure it doesn't wander up and down more than about 5 or 10 Ohms. Like with the discussion of multi-balanced armature IEMs above, at low impedance levels the EQ of the headphones will be changed in a way similar to the shape of this curve.
For the headphone above, there is a peak in the impedance curve at about 25Hz. Because it's only slightly below 30 Ohms, we can assume there will be only a slight boost to the 25Hz area as these headphones interact with the output impedance of the Pono Player. (This is true of any player with a similar 3-4 Ohm output impedance.) If the impedance plot above had the same shape but was generally aligned at 10 Ohms, the effect of the bump would be more. In the case of the Denon D5000 above, the slight boost in the bass at 25Hz will probably be helpful as in the frequency response plot above it you can see that the bass starts to roll-off with these headphones at about 35Hz. This impedance bump interacting with the output impedance of the Pono will have the effect of providing a little extra extension in the bass.
To reiterate: If the impedance plot is above 30 Ohm there's pretty much nothing to worry about. If it's below 30 Ohms you'll have to do a little interpreting to determine whether it will be problematic or not.
The second area you'll need to look at is the voltage efficiency of the headphone. Also circled in red in the lower right corner in the above data sheet is the RMS voltage needed to achieve 90dBspl at the ear. A low number means the headphone needs only a small amount of voltage to play loud.
I would say that any headphone under 0.100Vrms would be comfortably driven by the single ended headphone output (non-balanced headphone jack) of the Pono Player. If the number is between 0.100 and 0.200Vrms it will likely be worth while considering rewiring the headphone for balanced operation. If the number is between 0.300 and 0.500Vrms you will almost certainly need to run the Pono in balanced mode to get satisfactory listening levels and performance from the Pono Player. If the number is above 0.500Vrms the Pono is likely not the way to go at all. You might be able to get some useful low-level listening volumes with balanced mode, but you'll almost certainly get much better performance with a dedicated headphone amplifier.
That's a lot of information to digest...do you really need it? Well, the truth is that 95% of all headphones will work just fine with the Pono Player. The place to be careful is with balanced armature IEMs and with expensive headphones that might get recommended by headphone enthusiasts, every thing else is likely just fine.
I did run the Sennheiser HD 800 and Audeze LCD-3 from the Pono Player in balanced mode. In addition to getting more volume range from these headphones, I also heard improvements in performance very much along the lines of what I've come to expect from balanced drive: a greater sense of ease; deeper blacks between the notes; and better articulation in clarity. I would say the improvement is subtle but clearly evident.
I will shortly write a "Headphone 101" article on balanced drive. It's a subject that I would write about eventually, but with so many Pono Player users freshly being introduced to the topic I think it would be good to cover it soon. In the mean time, you can read one of my old articles on it at HeadRoom's website.
Wrapping up this First Bit
I started this page stating my initial subjective impression that the Pono Player was a distractingly musical device. I then went on to do a bunch of blind testing to see what the details of performance were, where the Pono player was strong and weak in various areas relative to some other sources; and which headphones were most suited to use with the Pono Player.
I found the Pono had an very smooth and artifact-free treble response and a sweet sounding bass and mid-range that, at times, fell a little short of the dynamic authority of some other devices. I also found it a more musical player then the consumer hand-held devices I tested.
The real question, the important question, the question of whether the Pono Player meets the brand promise or not is:
Is the Pono Player PONO (righteous in service to you and the art of music)?
Turn the page for my thoughts....