Parsing Pono Speak
I was interviewed on Leo LaPorte’s “Triangulation” podcast show back in September. We talked about a variety of things including the Kickstarter campaign for the “Music and Audio” guide, high-resolution audio|music, and Pono. It was not a live interview and I’m not sure when they’re going to run it but I think Leo understood the importance of provenance.
On Episode 218, Leo interviewed Phil Baker, “the product guy”. He’s had a lifetime of experience helping companies…including Polaroid and Apple…bring products to the marketplace. Currently, he is the VP of Product Development & Operations at PonoMusic. Phil is the guy that helped Neil realize his dream of a state-of-the-art, high-resolution, portable music player. He engaged the services of an electrical engineer, an industrial designer, and Charles Hansen to pull together the Pono player. As I’ve stated previously, the $399 Pono player does a very good job of reproducing music. I have problems with the shape, the screen, and the interface but OK…the unit can deliver very good sound to a set of headphones. Are there better options? Yes.
And Phil Baker is the real deal. This guy knows his business and makes a very credible case for the hardware. However, Leo and I talked about the lack of real high-resolution music on the PonoMusic website. I told him the same thing that I’ve been saying elsewhere, 99% of the albums/tracks that Pono offers on its website are standard-resolution…not high-res. Neil knows that and so does Phil. That’s why they have the “Pono Promise”, which is supposed to give comfort to PonoMusic customers about the content that they purchase. It states that if a higher resolution version of a previously purchased album ever becomes available, Pono will upgrade your copy free of charge. Sounds like a great thing until your realize that the “higher resolution” versions don’t sound any better than the ripped CDs…except in rare cases. And the rate at which the upgrades are coming from the labels is a trickle…maybe 50 per month.
At about 1:04:00 into the interview, there is the following exchange:
LL: It’s one of the issues, of course…I’m not going to put you on the spot with this, but a lot of Pono music is just upsampled CDs, it’s not really…
PB: No, we don’t do it.
LL: You don’t do that.
PB: We don’t upsample. Nope, no Pono music is upsampled.
LL: Some people have claimed…but you’ll say right now that this is not true.
PB: That’s right. If we find something that was upsampled, it goes.
So who’s right? The “some people”…including myself…who assert that the vast majority of albums/tracks on the PonoMusic site are CD-res or Phil Baker who emphatically defended Pono and told Leo that none of the music is upsampled? We both are.
I agree with Phil, none of the music on the PonoMusic site is “upsampled”. They present their 2 million ripped CDs at the original sample rate…44.1 kHz. They DO NOT upsample the content they receive from their supplier in the UK (the people that do the ripping and tagging). It might actually help the fidelity if they did upsample the tracks…but they don’t. They present them at straight CD specs.
Leo simply asked the wrong question. If he had asked Phil what percentage of the music from Pono is CD-res vs. Hi-Res, Phil might have had some trouble answering. There are only 6000 or so new transfers from analog tape available on the PonoMusic site. They’re not high-resolution either…but at lease they exist in a high-resolution container.
So goes the Pono dance. The “world’s largest hi-res music store” isn’t the world’s largest and it doesn’t offer hi-res music. The devil is in the details.
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
3 Days and counting down for the “Music and Audio” A User Guide To Better Sound”. If you’ve been waiting for the right time to join the over 706 backers, now is the time.
Hi Mark,
I really appreciate your daily newsletter and thanks for it.
This up sampling story. As part of my system I use a Cambridge Stream Magic which up samples everything to 24 384. The sound does change a little but who knows if it’s better? I previously ran Squeezebox via Behringer up sampling DAC and switching back and forth there were changes. Again who knows?
Currently the Cambridge is only used to stream to my TacT/Lyngdorf amp and room correction and the DAC in the TacT handles the conversion. It may sound dumb but digital into the Room Correction then converts to analogue to the amp and back to PWM. In the words of Peter Lyngdorf the Class D amp is just a high power DAC…..?
Speakers are either my Double Tannoy Autographs with mid, treble and super tweeter ribbons or Altec A7 setup both of which are very revealing.
Regards
Chris
South Africa
Mark wrote: “They DO NOT upsample the content they receive from their supplier in the UK (the people that do the ripping and tagging). It might actually help the fidelity if they did upsample the tracks…”
How?
The reconstruction filters would have a more gentle slope and therefore the sound could potentially improve…a little.
Recent behavioral studies on human subjects proved that humans can discern timing alterations on a 5 microsecond time scale, indicating that that digital sampling rates used in consumer audio are insufficient for fully preserving transparency.
Jay,
I’m not familiar with the research that says 5 microseconds can be discerned. Could you explain or provide a reference? What is the experimental condition?Are the auditory differences played to a human subject with a fixed microsecond difference between their two ears? Are the tones a single frequency?
If I assume this was the type experiment, I wonder if this approach appropriately mimics the frequency domain of digital music? In other words, assuming we could discern 5 microseconds difference in audio delay, does that mean our hearing could detect the impact resulting from “undersampling” at 96kHz? I don’t know the answer, but while these two frequency/timing measurements are related, I’m not sure audio delay timing and the frequency domain in digital samping is a equal comparison, other than they can both be expressed in the same units of “per sec”.
And this theoretical argument ignores the fact that John Siau mentioned: a DACs with increasing sampling frequency often degrades other performance specs (was it noise? I can’t remember the specifics, someone may have to jump in here).
Todd, check ‘Dr. Kunchur 5 microseconds’ out. There was also some J.D.Johnston against these results, but with little to no success. As to John Siau, then he just had replied to this all on a forum, but completely missed the essence of the subject.
Here only two aspects :
1} a sampling rate of >1MHz is required to prevent ‘sound blur’ ;
2} a sampling frequency of >>192kHz is needed to preclude ‘high-frequencies attenuation’ .
If you have two sharp peaks of sound pressure separated by less than the sampling period, the two will become blurred together: the temporal density of digital samples is then simply not enough to represent the two peaks distinctly and nothing you do with the bits can change this. Unless a different interpretation of minimal temporal separation is taken, it is completely fallacious to assert that a CD can resolve less than 5 microseconds when its individual samples are separated by periods of 23 microseconds. Hence, even 192 kHz is hardly enough…
Jay – if “you have two sharp peaks of sound pressure separated by less than the sampling period”, you have a signal that includes frequencies well above the Nyquist Frequency for the sampling period. Those sharp peaks will be eliminated by the low pass filter before being converted to digital.
Thanks Mark…I dare not tread here.
Although I am not the author of those lines & still being unsure whether to send him such an objection, aren’t you implying that when a recording has been made at 44.1 kHz sample rate, it consequently lacks any sharp sounds? However though, thank you for your not refusing CDs to contain sounds as such.
Jay, a 44.1 kHz PCM sampled audio track will have the entire audible frequency range of human hearing. What is a sharp sound?
Dunno. One Ph.D. simply wrote ‘…sharp peaks of sound…’ and some Mark U for whatever reason just clung to this expression with an irrelevant comment. The only thing I can add to this is that after Upsampling the sounds like piano and other sharp ones become not to say sharper but rather clearer. Sampling theorem, of course, self-regulates, but it doesn’t consider how humans perceive entire soundstage.
I take Mark’s comment as being completely relevant. What is it that you believe makes a sound clearer? Higher frequencies? Better timing clocks?
That’s a good question since the same perfectly applies to video as well. In audio, the effect is like ‘2D’ dough becoming 3D puff pastry in the oven. What’s stranger is the more precise soundstage that’s also there. (And if one could hear frequencies higher than 22.05kHz, then it would no other than just vanishingly add to the ‘picture’ with principally no change in whole perception.) So, they say it’s due to the human ability to differentiate time periods on the order of one microsecond, and indeed, may seem like sample rate higher than 1 MHz doesn’t improve sound further. Needs to be rechecked.
The whole problem of audio & video is ‘temporal aliasing’. It can be eliminated by means of very heavy oversampling.
O.K. The leading sound accompaniment on this song is with no doubt sharp :
That’s a good question since the same perfectly applies to video as well. In audio, the effect is like ‘2D’ dough becoming 3D puff pastry in the oven. What’s stranger is the more precise soundstage that’s also there. (And if one could hear frequencies higher than 22.05kHz, then it would no other than just vanishingly add to the ‘picture’ with principally no change in whole perception.) So, they say it’s due to the human ability to differentiate time periods on the order of one microsecond, and indeed, may seem like sample rate higher than 1 MHz doesn’t improve sound further. Needs to be rechecked.
I don’t follow the “sharp” concept.
I have never purchased from Pono but I have quite few downloads of CD resolution from Qobuz, which I presume gets the same CD rips as a re sold to Pono. The stuff I have purchased are of jazz recordings that I have been able to get for very reasonable prices on par with buying a used CD from Amazon with shipping costs included. Most sound wonderful. I can live with that as I know what I am buying. Now, if I am buying a CD rip in a 24/96 container and paying the added cost, often twise as much, that is another matter and one I would not he happy about. Provenance is key!
It’s simply the value proposition. You can do better at less expense…but it is convenient.
What would you say the percentage of the 6000 hi rez recordings on the Pono (or HDTracks, etc) site are just direct 24/96 or 24/192 transfers of the master analog tapes, vs the products that have had the sound from the masters touched by someone to offer a true improvement in sound quality. IE stuff such as the Steven Wilson remastered Jethro Tull Thick As A Brick & Aqualung offerings.
I was under the impression that the labels were making an effort to improve the he rez releases in some manner other than just a direct transfer of the masters to a big digital bucket?
Steve Wilson’s remixes and any remasterings done can change the sound but cannot improve the base level fidelity. Very few are remastered.
Mark, I understand the rationale you use in defining hi-res music, but by that definition doesn’t it also mean that HDTracks suffers from the same problem you say Pono does — i.e., they are not offering hi-res music, just analog transfers that can’t be called hi-res?
I ask because I’ve noticed on your RealHD-Audio website that one of the primary advertisers is HDtracks, and quite frequently the album featured in the ad is an older album from the ’60s or ’70s that was originally recorded on analog tape and the headline in the ad says the album is “Available Now in High-Resolution.”
Do you object to HDtracks advertising like that on your site and calling those analog transfers high-resolution? You do, after all, profit from those ads.
Also, I assume that most all new music from contemporary artists released each week is recorded digitally, often in high-res. Both HDtracks and Pono add those hi-res recordings to their online stores each week, so even under your own definition of hi-res how can you make a blanket statement that Pono “doesn’t offer hi-res music.” Aren’t you flat wrong in saying that?
I ask these questions because I can’t quite understand your animus towards Pono. If you don’t like Pono because they call their analog transfers of older albums hi-res, shouldn’t you be equally harsh with HDtracks and other online hi-res music outlets that do the same?
I would think that the public attention that Pono helps raise towards better-sounding music would also help you in your commercial endeavors to record and sell hi-res recordings, and that broadening public interest in better sound is in everyone’s interest. A rising tide lifts all boats.
-Charles C.
I’ve been very critical of HDTracks since they launched in 2008. In fact, it was their “Ultimate High Resolution” claims that started a lot of this stuff. Both David and Norman Chesky are friends but that doesn’t mean that I’m going to ignore their false advertising. At least they don’t have 2 million “so-called” HD Tracks that are really CD spec. Yes, they do pay for a banner ad and they supported the KS campaign. If you search my site you’ll see I write about them often. The difference is the blatant charade that Neil Young has pulled on the market. It’s great to raise awareness but it’s important to be honest. Phil Baker could have corrected Leo Laporte, he didn’t.
The content on Pono (and all other sites that license their content from the majors) should rightly be called “High-Resolution transfers”…but calling them Hi-Res sells more.
Mark, you can’t criticize Phil Baker for failing to “correct” Leo Laporte; Phil Baker honestly answered Leo’s question regarding whether Pono upsampled any music (as you yourself noted).
I’m sure that if Leo had asked Phil about the ratio of CD resolution songs and albums to high-res versions, he would have been honest about it. Everyone who visits the Pono Music Store knows that the vast majority of music is currently CD resolution; the resolution of each song and album is very clearly labeled. And Phil Baker also would have noted how Pono, Neil Young and others have been urging the record labels to provide higher resolution versions of their older albums, so that the ratio of hi-res to CD-res recordings in the store will continue to grow — including as artists continue to release brand-new albums that are natively recorded in hi-res digital (which, as I noted in my original post, is why your statement that Pono “doesn’t offer hi-res music” is simply incorrect.)
Btw, I’m fine with the term “high-resolution transfers” for those original analog recordings that are transferred to digital higher-than CD resolution; I find that most such hi-res versions sound better than the CD resolution versions.
-Charles C.
I don’t think I criticized Phil…I think he’s the real deal with regards to product development. Leo asked the wrong question. The advertising that PonoMusic does…like the postcard I picked up at the California Audio Show in San Francisco…is misleading. They state, “PonoMusic bring you the largest collection of Hi-Res music anywhere with over 2 million tracks!”, which is clearly untrue. HDTracks has a larger number of “high-resolution transfers” and only a few CD transfers.
You may not think you criticized Phil Baker, but the way you wrote your initial response to my comment certainly seems critical: “It’s great to raise awareness but it’s important to be honest. Phil Baker could have corrected Leo Laporte, he didn’t.”
I take you at your word you didn’t mean to criticize Phil; as someone who does a lot of writing myself, I often discover that what I initially wrote conveyed a somewhat different meaning than what I intended.
Lastly, I can’t speak to the postcard ads or other advertising that Pono distributes, but on their website at least they describe their music store as follows: “…our music store that brings you musical art in its original, highest quality form.” I’ve had to deal with folks responsible for advertising in the past, and have found they frequently have a tendency to over-promise. If Pono’s postcard ad states what you say, I agree it should be more clearly written for the sake of accuracy.
Btw, thanks for taking the time to reply to these posts. Not everyone is so diligent.
Charles, I wrote about the CAS Pono Post Card and actually scanned it because of the disconnect between their site and this marketing piece.
To be fair Mark, you didn’t correct Leo either. I just listened to the interview he did with you today and when Leo asked about the Pono Library being upsampled CDs you simply stated that only 5000+ are HiRes or HiRes Transfers.
Now, of course you guys were having a conversation and maybe you missed him saying that, just pointing it out.
OK, I missed that…but I did categorically state the 99% of the stuff comes from CDs.
Pono doesn’t have “2 million “so-called” HD Tracks”
They very clearly state with text and visuals that the 44.1 tracks are 44.1 quality. They are not upsampled.
They DO say however, buy from us and WHEN it get’s upgraded then you will get that free, which we certainly can’t say that about iTunes or Amazon. I don’t think there is any other place you can download even CD quality versions of these tracks, so at least at Pono you can get the Lossless 44.1 versions and then when they are upgraded great, now they can be a complete copy of the master in case the 44.1 wasn’t.
Tom, they are cagey about the information they provide. The postcard I mentioned previously states, “PonoMusic bring you the largest collection of Hi-Res music anywhere with over 2 million tracks!” No you could argue that this is actually true…but it sure seems to imply that they have 2 million “hi-res” music tracks. And elsewhere on the card it says, “Pono High Resolution Music Store”. It would be more correct to say, “Pono Standard Resolution Audio CD-Audio Rips with some Hi-Res Transfers.
Mark,
Been a follower of both you and Pono for some time now.
I split my time between the USA and Singapore, and I appreciate the Pono Music store for what it is – a place I can get my music in an uncompressed form. CD Media is dying, and the last major CD store in Singapore shut down over 6 months ago. Plus, the hassle of ripping to FLAC format is worth a few dollars.
While I have purchased from other sites, and find sites like murfie.com very useful for getting CD-quality FLAC format, I do purchase a fair amount from Pono, and I have experienced the Pono Promise first hand. I’ve even seen a couple of cases where the Pono store has stopped selling a 192/24 or 96/24 when the provenance showed that it was originally recorded/mastered in 44.1/24, and the 44.1/24 version is sold moving forward.
I would also agree with you on some of the recordings that are “true” 96/24 or 192/24 – in my purchases, I’m still finding a lot that look like a lo-pass filter was applied at 22KHz, and only a very very tiny amount of signal above that line. Can’t say for certain why that is – mics? mastering? There is signal above that mark, but it’s so small that any “ambience” it creates would be negligible.
It is a complicated issue, and I applaud your efforts to show that it’s time to move away from analog recording and DRC turning recordings into mush.
But I also think you are attacking the wrong people when you go after the guys at Pono – they are trying to move things in the right direction, and build a company that can fix the issues wrong with modern recording and mastering. Pono does a service to the community by providing the best available recordings, without compression. Yes, there’s still a lot of 44.1/16 content. Yes, there’s a lot of marginal content in larger bit buckets. But there are also gems like the Tom Petty Live Anthology. Great example of proper mastering, with cymbals and sibilance up in the high ranges unhindered on most (not all) tracks. I agree that’s one among millions, but the community at Pono helps the cream rise to the top, and sinks the refuse to the bottom. (There was a big push to eliminate the 96/24 master of Blood Sugar Sex Magic by the RHCP because of the remastering there.)
In a perfect world, I’d like to see you and the guys at Pono WORKING TOGETHER, not sniping at each other. What if Neil Young recorded something with you? Or pulled some strings with other recording artists to do a “Waldrep Sound” signature series, with Tom Petty, Pearl Jam, etc. You work your recording magic, and Pono sells the exclusive results? No one would question the provenance, and it would give you the stage you deserve to showcase what true digital hi-res should sound like??
I think customers and music enthusiasts deserve to know what PonoMusic is selling. I have not beef with Neil…in fact, I’m a fan. I’ve worked with him on a DVD project or two. I actually tried to work with him through his former CEO John Hamm, but was told they weren’t interested.
Mark, I just wan to say kudos to you for speaking up about this. I also listened to your Home Theater Geeks podcast and I was overjoyed with your comments. I’ve written about it extensively myself and we are on the same page.
However, what IF Neil and Co. developed a “Mastered for Pono” initiative similar to what Apple is doing with iTunes. But this time, really crack down on Loudness War mastering – disallow brickwall limited masters, standardize a loudness (DR8 or higher on the TT or perhaps come up with their own metric), ensure that a “Mastered for Pono” high-res track is guaranteed to be recorded and engineered at the base bit-depth/sample rate advertised, i.e. a MfP album at 24/96 means it was recorded, mixed, and mastered at 24/96, etc.
This is something I think Pono can do to not only distinguish themselves from say HDTracks but give more credence to the idea that high-res can be a release vehicle for better recorded music. Right?
I wrote a series of posts recently called “10 things the labels and site could do to get back on track with high-res music” or something to that effect. Simple stuff…like being open and honest about what you’re selling. They aren’t listening. They are struggling and looking to be the largest…not the best high-res site.
Mark, I read the 10 steps article and agree with all of them. However, none of them give any of the distributors financial incentives to change.
That’s why I think a Mastered for Pono initiative makes more sense. They can charge a premium for it.
The financial incentive is that there will be a real business in selling better copies of older recordings. Otherwise, the gravy train evaporates.
LOL. Obviously you’ve never taken a marketing course… (just kidding, just kidding).
I think you have to speak dollars AND sense (get it?) if you want say, the Chesky brothers to listen. I want HDT to succeed, but based on solid ground.
I really think we need a bunch of small labels, artists, and engineers to band together and create an ecosystem for high-res. You’re part of it Mark, but there needs to be more. Sorta of a collation for high-res we real standards. I think that would resonate among audiophiles.
I just an audio engineer, musician, and composer.
Now, what exactly must be done for video to look excellent, the four smart steps:
1} set to a fixed but sufficient number of pixels ;
2} forget compression ;
3} use 64 bits per pixel ;
4} record with as much frames per second as physically possible to introduce temporal separation that will present natural 3D .
Hi Mark – I thought this was interesting — https://youtu.be/ARLbfS84rug
“Smart transmitting of 3D audio up to unrestricted NHK 22.2 sound, a problem Swissaudec solved with ECMA-407: NHK 22.2 sound is built with 24 channels, 22 speaker channels and 2 subwoofer channels, consuming 28’800kb/s bandwidth in PCM. With ECMA-407 Swissaudec provides the solution to compress this amout of data up to 600 times – holding the world record in audio compression! By internally multiplexing additional 2kb/s, they extend any HD 5.1 or 7.1 carrier to transmit UHD audio up to NHK 22.2”
I wasn’t aware of this…thanks!
1. Is it true, that a tape transfer to 24bit – 96 or 192kHz is an exact (complete) copy of the tape that was transferred?
2. Would a tape transfer to 16bit – 44,1kHz (CD quality) be missing some (audible) information that was on the tape?
If 1.) is correct, it would make sense to buy those files because one would have an exact copy of the (master) tape.
But, if CD quality offers all the information that was on the tape which was used for the transfer, then there is no need to purchase the Hi-Res downloads from PONO or HDTracks etc.
This is a tough question because not all master tapes are equal. No. 1 is definitely true for all analog master tapes. And No.2 is true for 90% of master tapes. I spoke to my studio neighbor and he just finished a project direct to 2-track 2″ analog tape. Said it sounded amazing. Too bad we didn’t have that option years ago.
Buying from HD tracks or Pono for a good digital file of an album is fine…but just know that you’re not getting hi-res music.