It's the Masters, Damit!!!

My morning started like most days: cup of coffee on the back porch, surfing news feeds, reading email, seeing what's up on Facebook...when I stumbled across this post by Headphoneguru (Frank Iacone) pointing out a new short film put together by Harman to extoll the virtues of hi-rez audio and warn consumers of the evils of lossy compression, ear-buds, and laptop speakers. It's a nice, warm, fuzzy film about good sound quality...but I find it somewhat misleading. Here's the film.

I'm totally cool with Harman's idea that the consuming public would strongly benefit from a better understanding of how crappy the average music listening experience is today...but it pisses me right off when this laudable effort is corrupted by an unbalanced, and sometimes plain false, portrayal of the problem.

For example, at the 12:00 mark in the film, the narrative talks about the problems with lossy MP3 compression, but the images show a signal that is more indicative of dynamic range compression (DRC). Here's a still.


Oh, there's no doubt that heavy handed DRC is evil...but it's got nothing to do with lossy MP3 compression and streaming service low bit-rates. Technical accuracy in this area is something that Harman should be able to achieve. But it's not the real problem with the film; the real problem with the film is that the real problem with music reproduction today somewhat complicated, and the film doesn't strike the appropriate balance needed to educate the public meaningfully. Well, I suppose that's not in Harman's marketing department job description, but it seems an awful waste of a teaching moment to me. Oh, and Harman's not the only one. Remember this graphic from the Pono launch?


Fortunately, they seem to have stopped using that graphic, but have replaced it with this one.


I'm sorry, but I find all this patently absurd. Perceptual encoding to reduce data file size is an amazing process and works very well. All the images on this page use .jpg perceptual encoding to significantly reduce the size of a bit-perfect image file, and it works quite nicely. Perceptual encoding may reduce the data to 1/10th the original, but the perceived loss of quality is far from linear. This whole "bigger is better" message to the general public is very misleading...and I think it's going to come back and bite the music industry in the ass if they don't start delivering a clear, true, and balanced description of the real problem.

Real Problem #1: It's the Masters, Damit!
I probably won't need to tell InnerFidelity readers about the "Loudness Wars", heavy-handed DRC has horribly polluted our sources of music for the last 30 years. If record labels want to make big bucks from hi-rez streaming services, they're going to have to deliver a product that is clearly better sounding, and they're not going to do that just by pooping out the latest release with crappy DRC no matter how many bits it has. People (normal consumers, not audiophiles) just won't hear it, and the press is more than happy to make that point.

What people will hear—and they'll be able to hear it even with 320kbs MP3s—is freshly remastered releases that go back TO WHAT THE ORIGINAL ARTIST INTENDED!!! I believe this is happening. From this Rolling Stone article about Pono:

Pono's preservation of the fuller, analog sound already has the ear of the Big Three record labels: Warner Music Group, Universal Music Group and Sony Music. WMG—home to artists including Muse, the Black Keys, Common and Jill Scott—has converted its library of 8,000 album titles to high-resolution, 192kHz/24-bit sound. It was a process completed prior to the company's partnership with Young's Pono project last year, said Craig Kallman, chairman and chief executive of Atlantic Records.

From that little tid-bit and many others I've seen lately, it looks like the labels and the industry are on-board for re-mastering or re-transferring original recordings and it seems quite a bit of work has already been done. The only one left out in the dark is the consumer, who's being fed bullshit buzzwords about bit-rates and has no idea about the evils of DRC.

And once the consumer hears no meaningful improvement with hi-rez they're gonna be pissed.


To the Industry
I love that Pono, Harman, and Apple are focussing on improving audio quality and telling consumers about it. I heartily applaud your efforts. But please, please, please, for the sake of the Art of Music, balance the message being transmitted to consumers. Sure, tell them 48kbs internet radio and satellite streaming sucks...but also tell them that 320kbs MP3s are pretty good, and some streaming services already do that. Tell them that cheap ear-buds and laptop speakers suck, and some pretty good gear is available for not too much money. But please, tell them what you're doing about higher quality re-releases of back catalogs, tell them that music on hi-rez streaming services will be the better for it, and tell them about the evils of heavy-handed dynamic-range compression. I know it's a bit of egg on your face, but you're fixing it. Shout that from the rooftops! Bang your drum, this is a worthy endevor for everyone's music listening pleasures. Once the consumer understands your love of the music in your care, and knows that you're preparing it for the coming age of awesome streaming music, they'll return to buying in big numbers. Sure, it may be $30/mo hi-rez (16/44.1) streaming services, but that's good money and I reckon the public will buy it in a heart beat if they know they'll be hearing their music free of heavy-handed DRC. I know I will.

Harman's "Distortion of Sound" website.
Pono website and Kickstarter page.
Just a few of the many Stereophile articles on the evils of DRC here, here, here, and here.
Michael Lavorgna's thoughts on Qobuz hi-rez streaming service.
Disgruntled press comments about Pono and hi-rez here, here, here, and here.
A quick look at Apple's "Mastered for iTunes" and Apple's instruction .pdf for the program.

Really good YouTube explaining the "Loudness War" DRC problem.

anetode's picture

"And once the consumer hears no meaningful improvement with hi-rez they're gonna be pissed."

That is, unless they convince themselves that they do. Having to pay extra and priming by high-rez hype along with its bullshit infographics are powerful motivators.

lafaard's picture

It drives me mad that a respected "Rock" legend such as Neil Young supports that idea, because people will believe him.

TheAudioGuild's picture

This sort of grossly misleading propaganda needs to be called out for what it is.

Tyll Hertsens's picture

I don't think it's grossly misleading. I think it's oversimplified, and lacks a focus on really important things that are happening.

I think the dumb mistakes parts in the film can be chalked up to marketing people who weren't well lead by upper management and industry associations who haven't developed a clear and common message for consumers. They made a nice little film, but it could have been so much more.

There really is an incredible story just about to happen under our nose. I think we're about to head into a great era of music availability...but nobody's telling us about it clearly.

TheAudioGuild's picture

I think that when the recurring theme is that MP3 is absolute crap (with such gems as claiming it removes 90% of the music) it's being grossly misleading.

TheAudioGuild's picture

Ah, this seems to be the real reason behind Harman's "documentary."

Currawong's picture

...that they will damage the hobby and industry as a whole with this misleading advertising.

TheAudioGuild's picture

Here's some more grossly misleading propaganda this time with regard to DSD.

floweringtoilet's picture

I would not say Harmon's film is grossly misleading, but I don't know how else to categorize Pono's marketing materials. Actually, "grossly misleading" is kind.

iTroll's picture

Need to start livin' in the post-loudness-wars world, folks ...
A lotta times ... yeah, the recording has been loudness'd so it fails it the DR Meter computes a relative low score (e.g., compared to 80s era CD). But sometimes, the newer version has better EQ, or they really did find a more) original source/master ... not sure there are (yet) scientific tests that will reveal subjective qualities such as tonality, micro-dynamics, etc ... but these will exist concomitantly with low DR (loudness).
There are no easy answers ... you gotta take one recording at a time, may be conduct your own listening evals ... all this takes time patience and analness ... a lotta young-n-dumb headphiles are into this stuff ... me: been there, done that.
'Scuse me now ... I'm on the throne and my iPad is about outta juice ... I need a wipe ... anyone? Hertsens?

Mr.TAD91's picture

I agree with your conclusions of the Harman video. Mp3's at 320kps can sound pretty good on a basic source like a smartphone if they were recorded well to begin with. And as for DRC, it's downright scary. Much of my music collection is lossless, but I do have some high quality mp3s. I actually struggle with stuff that was not recorded well to begin with - on my most accurate/resolving rig. Sound is dry as toast and lifeless. The good stuff however, is sublime - That's how it should be.

I think the enjoyment of music boils down to the emotions conveyed by a song, the meaning of the lyrics and the overall response to the stimulus the user experiences. The average user couldn't care less about "sound quality" because they tell me as long as they hear it, they're happy. They don't understand nor care about high frequency detail, transparency, neutrality, nor the closeness to the studio recording, or the integrity of their source.

Most people these days simply aren't paying attention to sound quality. If a pair of 'popular' and stylish headphones hit the market, everyone wants a pair. Your average consumer wants coloured sound, not accurate sound. And this is where I feel like the average consumer is irrelevant. - It's only when they desire a better listening experience that they'll be saved from lo-fi --entering the pearly gates of mid-fi!

- Mr.TAD91

zobel's picture

You tell it like it is! You are on a mission!

In regards to the high res numbers game:
You can fool some of the people all of the time, all of the people some of the time, but not all of the people all of the time.

I love live music and playing it with others. I love recordings that take me there. I loved the Montana Folk Festival in Butte this past weekend, lots of GREAT music! Looking forward to Tommy Emmanuel in Missoula on the 23rd of July. Looking forward to our drum circles here in Hamilton, every 2nd and 4th Wednesday at the River Park on 9th street at 6:30! Join us if you are in the Bitterroot! Fun! Free! Open to all ages and skill levels. It ROCKS.

DaveBSC's picture

Since roughly 1997 or so, the industry has settled around DR5-DR6 as the average level for mainstream CD releases, and despite claims by folks like Bob Katz that since Soundcheck level matches recordings, thus making dynamically crushed recordings sound considerably worse (and no louder) than their dynamic counterparts, there's no logical reason to continue with the war, on and on it goes.

Mainstream "HD" releases on stores like HDTracks are extremely hit and miss in terms of dynamic range. Beck's latest in particular is an absolute disaster, and there remains serious questions about whether at least parts of some of the tracks started out as low bitrate MP3s before being upsampled and sold as HD (in addition to being crushed every bit as hard as the CD.)

The Loudness War is a cancer on music, plain and simple, and those who ignore it, or throw up their hands and say that there's nothing that can be done, or accept the excuses that "it sounds better on earbuds/laptop speakers" are just as much of a problem as the artists and engineers that continue to pump out DR5 and DR6 garbage year after year (to say nothing of the shockingly horrible DR4s and DR3s).

Buying these albums is basically the same as paying to see the latest Transformers movie, they are the lowest of the lowest common denominator. If you stop giving Michael Bay your money, he'll stop making Transformers movies. And if you stop buying dynamically ruined albums, and contact the bands on their Facebook pages or however you can reach them and tell them WHY you didn't buy their album, maybe something can be done to end this idiocy once and for all.

When artists bother to create a master for vinyl, it's usually dynamic, or at least a lot more dynamic than the CD release. They think people *want* CDs to be like that. You have to convince them otherwise, or "HD" music will be every bit as bad as CDs have been for the last 20 years.

castleofargh's picture

step one: hear nice song with good rythm on a car radio.
step 2: buy the album to listen to the song on your hifi system.
step 3: realize the song is over compressed overclipped utter crap.
step 4: hope piracy will actually kill the label.

can't wait to listen to the black eyed peas albums in 384khz.
radioactive(imagine dragons) is gonna be T3h 8OM8 in hires. after all the song won so many awards everywhere, it has to be just that good. "welcome to the new age... it's a revolution I suppose..."

Steve Guttenberg's picture
Why are recordings handicapped this way? I asked a mastering engineer friend, and he described a "feedback loop" where producers, artists and engineers make their way through the studio process to release an album, and how they later hear it in the real world influences the way they'll make their NEXT album. That's the way it worked in the 1950s, with LPs and singles, which were heard at HOME on a hi-fi and on the radio, so the engineers made recordings that sounded best at home and on the radio (radio stations compressed the dynamics so when you listened in a car the quiet parts were louder). That initially held true with CDs, but now that most people experience music away from home, in noisy environments, on lossy streaming services like Spotify, all of that informs the way producers, artists, and engineers approach their next recording. He believes that the way today's pop music is arranged, designed and performed is strongly influenced by the quest for extreme loudness and limited dynamic range required for music listened to on-the-go with tablets, ear buds and phones. It's one thing to hear the sound in the studio, but the music will mostly be "consumed" as background filler. It's rarely the prime focus anymore. Engineers aren't trying to make the most accurate or natural recordings, for a few nerds, no, they make recordings that sound some version of good for the masses of people who listen on crap Bluetooth speakers and ear buds, or in their cars. At home, most people have music on as background sound, so soft to loud dynamics would be an annoyance; maximally compressed music is a plus.
AstralStorm's picture

Their problem is they don't realize all those portable devices and laptops have a capability for enforcing equal loudness, be that by internal automatic gain control w/ DRC or loudness specification tags (preferrable).
Really. Even friend's $10 MP3 player has this thing, suprisingly some cell phone apps don't, where it should be trivial to add.
And Spotify might as well add this in their app or precompress the streams like radios do.

Precompressing the track is a terrible copout and makes for super annoying listening experience, where every other track I have to fiddle with volume control, because nothing is consistent. Both tracks being DR6, the crunched one will be much louder (and crappier), while the loud uncrunched one will be quieter. Especially noticeable in metal.
Well, until I tag it, then the compressed tracks sound a bit quieter than everything else.

By the way, even some radio stations overcompress *much* too much. Cars are not jet planes, that much compression kills the music, makes lyrics and even voice less legible without giving anything in return.
See, a bog standard running car + FM radio is easily 50-60 dB SNR. A crushed track can often be 20 dB SNR thanks to all the extra distortion.

lafaard's picture

So basically, today's music is mastered in such a way that it sounds better in noisier environments? This would fit most people's needs and makes perfect sense. I have to admit that a lot of detail in the music is lost when you're distracted and/or there is background noise.

xnor's picture

ReplayGain (same perceived loudness) and user-configurable DRC in every player.

For example, commercials used to be very loud and extremely compressed here, but guess what, since the broadcasters here use loudness normalization the advertisers noticed that their commercials sound wimpy and distorted.
So the result is not only comfortable volume (some commercials used to blast your ears), but less compression, overall much better sound quality.

drez's picture

This video is honestly one of the worst peices of communication. I came across a post from Linkin Park (lol!) on Google Plus and all the follow up replies show that laypeople only got the part about mp3 compression, and learnt nothing about DR compression which is the real evil here.

If the recording is bad, then no hi-res format, or playback equipment will save it. Funny thing I posted a thread about this not more than a week ago. Badly recorded and compressed music sounds exactly the same on a megabuck hifi as it does in vbr mp3 with freebie earphones. Funnily listening to HDTracks "24 bit" master of Linkin park ( with DR4 ) triggered this bit of nostalgia.

Good recordings sound better on just about any half decent equipment and storage format. They might sound even better in Hi-Res and with good playback system but this is secondary.

False promises about Hi-Res will not help Pono. It takes a very high end system to show advantages of Hi-Res formats. Will most people realistically be using good enough equipment to hear the format difference? I think not.

carvern's picture

Taking an old audio tape and encoding it at high-res bitrates, doesn't make it hi-res. Unfortunately the CEA has decided that you CAN, in fact, call it hi-res. I applaud remastering recordings to remove the dynamic range compression that was necessary for vinyl releases, and encoding them at higher bit rates will make it sound better, but do call these hi-res is just another way for the record labels to get consumers to pay up for something they don't need.

What we need are recordings actually recorded in hi-res; i.e. hi-res mixers, no dynamic range compression. That will be worth paying for; not $25 for a "hi-res" version of a 1960s Blue Note jazz recording.

Steve Guttenberg's picture
Well, but the vast majority of hi-res releases have always been sourced from analog masters. As for new music, very little is recorded in hi-res, and as for pure DSD (with no analog sources or PCM conversions) on rock or jazz? There's just a mere handful. Why? Because as I said before the engineers and bands aren't courting a few thousand audiophiles with awesome sounding recordings. It's not worth the extra expense to record, mix, and master in hi-res. That's reality. Most civilians are now used to getting music for free, or at best, a $10 monthly subscription. They're not going to pony up $20 for one hi-res album, but a few thousand audiophiles will. No more than that.
AstralStorm's picture

I'd like a good proof that "hi-res format" is actually noticeable.
To me, high res master means good noise floor and excellent miking on recording combined with a bit of musical touch.

There's no other high res required or possible, anything else is a bandaid. For instance, certain "hi-res" masters of analog recordings just used good noise suppression and a bit of equalization. That's a band-aid if there ever was one, though it can be useful for really old ones.

The digital part is trivial to achieve nowadays, we live in the DDD world. So, the only thing required is to master in 24-bit or 32-bit resolution. Not publish in this format, that's unnecessary. Sampling rate is even less relevant.

labjr's picture

I'm so tired of reading this opinion in every thread. Don't buy it if you can't hear the difference. All the music I like was recorded on analog tape. I'm not gonna start listening to different music for the recording quality. And everything I've compared with CD sounds better to me. I don't care why. So no need for you be informing everyone. I'm pretty sure we can judge for ourselves.

mproietti's picture

Excellent post! The biggest detriment to music is happening in the studio!! I have yet to be convinced that redbook CD format (16b/44.1Kh) is not sufficient to produce a stellar musical experience...

Bill Leebens's picture

As Art Dudley pointed out, if you watch the pitch-video for the Pono Kickstarter campaign, the majority of those gobsmacked celebs were given demos in Neil's Cadillac. Given the overall hypey-ness of that pitch, it's a little amusing to see the demos done in such a compromised environment.

Or is the message supposed to be that the improvement is so overwhelming that it can be appreciated even through two tin cans and some twine?

We're back to the Wild West that existed at the beginning of the CD era, when recordings from all manner of provenance were being hyped as digital. If yoy take a Redbook recording and upsample the shit out of it, as is often done, will it be as good as a natively hi-res PCM or DSD recording? I don't think so.

Why is this so difficult to convey? VHS to HDTV was a far easier pitch. Do we just trust our eyes more than our ears?

metaldood's picture

@Bill Leebens Music is listened in the background while doing other work. Very few (compared to watching TV) listen to music by just sitting in front of your audio system.

ulogin's picture

Is that your cat, and is that your PONO?

Willakan's picture

...but I'm extremely skeptical that many tracks will be remastered for Pono. Companies have been submitting "hi-res" masters to iTunes as part of the "Mastered for iTunes" program for a while now (so they've got the stuff ready to sell) and all the wording on the part of those involved has been remarkably airy-fairy.

You've got Neil Young talking about digging for the best masters in his interview with AudioStream, but then he also declares that Pono is essentially no different from HDTracks and the like with the exception of its target audience.

The only "solid" confirmations about hi-res material sourcing are extraordinarily carefully worded. WMG has "converted" its library...interesting choice of words. To the cynic, the impression is given that they simply changed their internal archival procedures (which would tally, actually, as AFAIK it's the norm to record and store at 24/96), and then jumped at the chance to sell their back catalog all over again.

At the same time, the CEA is working on a formal definition of "hi-res". Their definition actually contradicts itself in an effort to allow absolutely anything to be sold as such.

The music industry has always made its money in back catalog sales, and has always done so in an astonishingly cynical fashion. I don't think *I'm* being cynical when I say that "This time it's not different". Hell, they couldn't even be bothered to tweak their "mastered for vinyl" stuff for early CDs - here if they do nothing at least they won't sound worse...

ultrabike's picture

Seems like a lot of marketing mumble jumbo. From this link, it seems Clari-Fi brightens and tins out the sound:

Who knows what kind of randomness is going on with Crapi-Fi, but it seems it is making inroads into the car industry:

The claims above however, is that this is does not merely equalize, but "cures" some kind of compression... actually not sure which. Weird juju stuff...

TheAudioGuild's picture

Yeah, when I listened to the demos, I thought my ears were going to start bleeding. Sort of reminded me back when I was a teenager and always kept the bass and treble cranked.