« eBay Find of the Week | Main | Refreshments! »

Sunday, 16 September 2012


Feed You can follow this conversation by subscribing to the comment feed for this post.


* The rDAC's highest sampling rate via USB is still 96KHz according to the user guide. Is this what you're using? It's still considered HD?

* The graphic of the back panel shows an Antenna label amongst the input section. I see no documentation for this. Is it just an indicator light for antenna operation of the kw?

So, I create both music and photography. And I find it interesting that I've got a very digital music playing and listening setup, whereas I've got a very analog photography setup.

And I'm obsessed with quality, either way.

But... I don't really find that there's any difference that I can actually hear between uncompressed 44kHz audio and 96+kHz audio. I can hear a huge difference between compressed audio and uncompressed audio. And I know that I don't have golden ears, but my range of hearing goes pretty far up for a person in their thirties, actually.

There is, of course, often a change in the way they master for the higher-quality formats. I think I'd be happier with a version that's free of excessive 'loudness' from being mastered to sound loud than a higher-resolution version.

After reading your Open Mike, I think I realized the difference.

If I go through an exhibition of art photos, I can step up close to the ones that are taken on a 4x5 view camera or a medium format film camera or even one of the really good digital cameras, and I can see that... oh, yeah, there's no pixels. And then I can see something taken on somebody's inexpensive P&S and upsampled at a reasonable viewing size and see how the resolution trails off into blurry pixelization.

And.... sometimes, that's really unpleasant, like seeing a model's pores. But for a landscape, that can be really beautiful and reveal new layers to a composition.

However, there's no such equivalent for audio. Once you've got a reasonably flat sound system and a room free of distorting resonances and ambient noise, that's it.

You can't 'zoom in' aurally.

Interesting stuff. I have a question - if I was to skip the hifi part and listen to HD audio through my Mac's headphone port using decent (okay, semi decent) headphones, would I hear the difference (assuming I am capable of it)?

It's not that I want to skip the hifi part obviously - but I think getting a bank loan to do it might put a dampner on the 'ol marriage...

In the USB DAC world it seems pretty hard to beat the ODAC (here is a vendor: http://www.jdslabs.com/item.php?fetchitem=46 ).
Does Audio Formats: 16/44, 16/48, 16/96, 24/44, 24/48, 24/96.
149USD +shipping.
I have one and it does what it says on the tin as far as I can tell, just plug it in and go with Foobar 2000 (no 129 dollar music software for me...).
As far as the supposed benefits of audio files larger than redbook CD standard, well that tearing noise is the sound of cans of worms opening...
(But then I am firmly in the objectivist camp and and our noble editor is quite clearly a subjectivist. Yer pays yer money, yer takes yer choice as they.)

Happy listening - and debating - everyone!

It doesn't have to be so complicated to build a HD-capable system. A network 'stream' player obviates the need to fuss with software, DACs and digital interfaces. The music files reside on a computer, but you use the player itself or an iphone to pick songs and control playback, just like a normal cd-player. Better still get one of the all-in-one units from Naim or Linn to build a great one-box system. I bought a Naim Uniti after owning several component systems with a DAC, and have never looked back (and the build quality is really top notch too, not all high end manufacturers have cheapen out at the entry level). A thoroughly musical system that removed the need for (often misguided) upgrades. Admittedly, some of these units are a bit pricey, but it's no worse than a full-frame DSLR with a good lens, and better value IMHO.

Uh-oh...another quagmire! Your article interested me, so I looked around to learn about HD music. I came across a Ctien-like paper that seems to debunk HD music, showing that it's not possible to hear the difference between HD and CD. It even quoted a research paper in which a number of listeners were asked to listen to the same piece twice - once in the original HD and once converted to 16/48. The results? They picked HD or CD accurately 49.8% of the time - chance, in other words.

Here's the link: http://people.xiph.org/~xiphmont/demo/neil-young.html

It sounds like you can hear the difference. Can you help us understand why/how etc? And where to you think this writer gets it wrong? What do you hear differently if you A/B the same piece in the different compressions?


1. Yes, as I wrote in the piece, my system goes up to 24/96. And yes, that's HD.

2. Arcam offers a wireless version of the rDAC. The wired version is widely agreed to sound better.

"Once you've got a reasonably flat sound system and a room free of distorting resonances and ambient noise, that's it."

No, that statement is absolutely, unequivocally not true.

I don't know, as I've never tried that. I don't listen to headphones of any kind.


Check out opus
Which seems to be the next big thing
Don't know what services support it though

Mike, is this kinda like Sheffield analog recordings? great recordings of mediocre performances?
if so who cares?????
I love a great recording but I still listen to Rubenstein's Chopin, Chet Baker's trumpet and Voice etc...

It's not a quagmire at all. As I said in the post, you need a good stereo to be able to tell the difference. If you don't have one, maybe you won't.

You can't necessarily tell the difference between a 24MP FF file and a file from a 5MP compact camera by comparing online JPEGs, either. That doesn't mean the differences don't exist.

By the way, double-blind tests don't work in audio. First because everything--including such details as how long the subject has been listening and how much caffeine he or she has had--affects the sound, and thus affects the experiment; but mainly because it simply takes time to become clear about what you're hearing. I often can't identify flaws in reproduced music on first hearing that become just plain as day--I mean plain as day--after a few weeks of familiarity. (And that's not just me. Virtually any audiophile will tell you the same.) That they're not immediately apparent also doesn't mean the differences aren't there.

Haven't you ever met a woman you thought was perfect on the first date but who later turned out to have some, er, idiosyncrasies? Did the fact that you didn't apprehend her whole personality and character instantaneously mean there was nothing more there?

To name just one trivial example, salesmen in the old days of stereo stores would often just turn up the volume a little on the speakers they wanted you to buy. Even on A-B comparisons, people couldn't even tell that one was louder than the other. Give them more time and more careful listening, and repeated back-and-forths, and almost anyone could tell.

Just as with getting to know people, it takes time to really get to know a component, a system, or a recording.


Yes, I agree. I listen mainly to jazz, which means I listen to many old and bad recordings. I even got a preamp that has tone controls, which many audiophiles dismiss these days. I need 'em. (Sonny Clark's "Cool Struttin'" just needs a bass boost and a treble cut, and that's all there is to that.)

Just because I want to listen to HD doesn't mean that's the ONLY thing I'll listen to.


No, Mike, despite your audio knowledge you're not an "audiophile". Audiophiles don't really like music: they use it to meter their audio systems' performance. They listen to the strangest music conceivable because they like the way it sounds through their gear. They don't enjoy music: they analyse the sound.
You, on the other hand, like the late, great Hank Jones and the sessions he recorded with the no less great Charlie Haden. ('Steal Away' still consistently delivers chills down my spine.) You like music, so I'm sorry to announce you don't qualify as an audiophile.
And thank you for the useful piece of advice on HD. As for me, as an ex-audiophile, I still resist entering the computer audio fray, but after some critical listening I have concluded HD is the future. It's on a par with good, 180g vinyl, which is no mean feat. But for now I'm still enjoying the vinyl resurgence; computer audio may have to wait a little longer.

The truth is, non-audiophiles think I'm an audiophile, and audiophiles think I'm not. [g]


HD Tracks won't sell to me here in Australia.

I must confess, before I got to the end of the article I had developed the proverbial "thousand-yard stare". I still buy CDs and I've never downloaded a single track of anything. After 8-track, vinyl LPs and cassettes all faded away, I decided I would stick with CDs. Then I went soft-headed and bought into the SACD. As expected, it was not a wise decision. Back to the CD. "Good enough" is good enough for me.

Right, HD Tracks is USA only (basically). Like I can't buy that Cohen album from Qobuz in France...not offered to the USA.

I assume it's a bandwidth issue but I don't know.


Nothing wrong with that. I did the same for years.


Just curious, Mike.

Do you do "active listening" to the exclusion of everything else most of the time?

I do, mostly when I'm alone, but not when I'm cooking.

Guess that I'm actually lucky that my hearing is going west.
I suppose that I'll just have to be happy with the sound from my ancient Electo-Voice Patrician and Klipschorn speakers driven by elderly McIntosh amplifiers. The Infinity K-9s still sound okay to me, too.

Hi Mike,
if you really want to move on from vinyl and mp3 and their limited resolutions, your real answer lies in high bitrate multichannel music. There is nothing that any 2-channel format can do with music, that comapares with the ability of good multichannel productions.

Unfortunately that makes the particular dac you have promoted, and the particular music download website you have promoted, inadequate. Maybe try something like itrax for multichannel music downloads. Once your system has evolved to the appropriate level, of course. :)

*But how does the FLAC of the Cardas Sweep sound on your system? That's the true test.


The problem with audiophilia is that for every objective test you can design the audiophiles can counter that the test is somehow not perfect, and thus dismiss it.

But some things are true:

1) A digital signal can perfectly reproduce any frequency up to 1/2 of the sampling rate - that is, the Nyquist frequency.

2) For a CD recording (44 kHz) the Nyquist frequency is 22 kHz, meaning that all of the original frequencies at or below that rate can be reproduced.

3) The upper limit of human hearing is 20 kHz (actually quite lower as we age - search "mosquito ringtones").

Thus the CD sampling rate was not chosen arbitrarily based on the storage capacity of a CD or some other technical limitation, but rather because it would allow perfect reproductions across the range of the human ear. The extra 2 kHz headroom allowed for a roll-off low-pass filter to remove the higher frequencies that could cause aliasing problems in reproduction but which cannot be heard by the human ear.

So while I absolutely believe that there's a wide range of stereo qualities, I absolutely do not believe that there is any advantage in higher frequency sampling, simply because the science doesn't support it.

The audiophiles response is that while we can't hear the higher frequencies we can detect their loss in "harmonics" or "resonances" but those harmonics we could hear would be at lower frequencies - and thus would be accounted for by the lower frequency recordings.

The camera analogy assumes that obviously bigger is better, since 24MP is better than 4MP - but is 200MP better than 24? What about 2000? There comes a point where the sampling rate is truly high enough - not just in a "good enough" way but in a "the human senses are not capable of detecting the difference" way; CD audio has passed that threshold.

How long before we go back to MIDIs, and play them through Raspberry Pis attached to actual musical instruments?

Another vote for the ODAC. Fuss-free, it does what it says, and the specs are impeccable. So is the sound. I used internal PCI sound cards (M-Audio, EMU1212) and this is just a so much more sensible solution. Portable too.

There is a version you can get (O2 amp with ODAC) that combines a headphone amp with the above, which has strictly a line out signal, and cannot drive headphones.

Just a question, iMacs and I think all new Macs have a native optical out port built into there headphone jack, this is how I run my audio out to my amplifier, which allows me to also edit audio in 5.1 surround sound. Wouldn't this be able to output the highest quality audio possible or is it limited in some way.

I have to say, that the article Hawkwood links to is pretty darn good science, as such. In other words, the methods given, and the physiological and psychological principles involved are sound.
I won't comment on the conclusions, (I haven't fully vetted them to my own satisfaction) but they are "doing it right."


Thanks for clearing that up, it saved me a lot of research time.I wasn't sure if you were talking about some magical ability to transfer wonderful analog recordings into something better,media wise...Just a new Sheffield kinda thing unless my almost neighbor Bob Ludwig works with it. (he is about to have a release party at Transparent Audio's demo room for his contribution to the remastering of some classic Zappa albumsIn order to win admission, you needed to know more Zappa Trivia than I care to know... ) And for those "number people" I have always used the piano analogy, that Steinway, Yamaha and Kawai all use 88 keys, there must be other numbers that apply that no one has measured yet.... and to the guy with the Patrician! it only works in mono, and holy crap the IMD on a 30" woofer!!!!

Hey Mike, how about throwing in the Music Hall DAC 25.3 with the valve (tube) output stage? Then we can get really audiophile geeky! Not to mention controversial ...

As a regular supporter of T.O.P, both written and monetary (just ask Mike, he knows)
every once in a while the writings reveal something of which I know nothing nor do I understand.
HD music is just that.

As one who has been through a number of audio systems that I refined kept for a number of year then sold am realizing like many happenings in my loire, as probably all of you have found; why do you listen what is entitled "music?"

This is my concern, why as opposed to the how.

As a recovering cancer patient with numerous post treatment problems
and having had a rather busy life of existence have found that which did excite me, does no more

As one of the other respondents noted a digital audio system he owns as a opposed to a non-digital photographic existence. And like him can well understand the quality of viewing, in my case an historical glass plate 4" x 5" glass plate negative from 100 years prior and the exquisite detail of the process
compared to the pixel madness of today's sadly preferred digital imagery. Often wonder is this progress or is this us wanting more sooner, faster and with considerably less depth so to speak?

One of my passions of audio is listening to and appreciating pipe organs, both classical and cinema.
And being able to read a musical score and at one time able to manipulate a pipe organ be it
cinema (Casa Loma/Shea's Theatre/Maple Leaf Gardens) organ or any number of one two or three manual classical/church organs. These days too digital has replaced pipes, digital being what people think an organ should sound as because
the digital aspect is what has been recorded from various venues and rolled into the sound of digital organs. Hence a wind charged pipe organ to me is more "real" that a digital or manufactured sound if you wish.

So it is with audio. The sound we hear or listen to is what we perceive the sound to have been at time of origination. Not what our system delivers. Short of having a pipe organ in one's home (and have seen and played a few as such) the audio system is what we have, in lieu of.

In my own case, the cancer, the after effects, then selling my family home of 58 years, then losing my Mum with who I lived with after my father died and moving into a top floor flat of a 15 story structure
means my audio equipment has been stored. So now am wondering how to restart, if I so choose?

I tended to listen to old and ancient. In the climate controlled off-site storage location there must be easily 300-400 pre WWII 78 rpm recordings, virtually all cinema and dance band recordings as well as
1500 or more 33.3 rpm records all catalogued and neatly kept. For years I never had a television and to this day maybe watch a maximum of three hours a week. I have recorded all of my listening music on audio magnetic tape using a pair of Revox A77 machines with the large 15" reels. My form of continuous uninterrupted music. I'd come home from work, make a small evening meal, and sit down and listen to a reel a night. Then it was bed time and up the next day. Over the years a large amount of
my music was then placed on audio cassette and as it become difficult to find anybody qualified in open real technology realized things were changing.
Still have those audio cassettes as well.

Compact discs are the current method although have never recorded my own music to same. And must admit am overwhelmed by all the choices for music sourcing in this day and age.

Then too I believe we are often confused by aural noise/music/conversation.

So reviewed what have now, none of which is situ. Bryston 2B amplifier, and related pre-amp
as separate components, Two Revox A77 tape decks, both non-operational, a very heavy (300 pounds) 78 rpm direct drive turn table salvaged from a radio station, a pair of Paradigm 2 speakers as well as two then top of the line Panasonic auto-reverse cassette decks and an older Sony turntable for the 33.3 albums. And maybe 60 or so compact discs. Oh and a Magnum DynaLab FM stereo receiver.

Ironically most of my recent listening has been on my 15" Macbook Pro and specifically YouTube.
I'd like to have some of those recordings to be played without the computer but so far haven't figure out quite how to do so, or record same to something
more portable.

I could say it's not the money rather what do I still enjoy? No longer sure of that or anything else of late. As much as I prefer film as colour slides and yet find digital imagery to be like a fart in a windstorm, there aurally but gone in an instant if not retained in some form.

Maybe those who've commented previously including Mike may well have perhaps an idea or two.

Keep in mind am in Canada and many items that would be available elsewhere are not available here
and if they are, are often extremely high-priced.

We're not going to get into the ageless disputes* here, but there have been many cases in the history of audio where "science" said that "the human senses are not capable of detecting the difference," and yet later, obvious reasons for audible differences were discovered. Cf. early solid state vs. tubes, TIM distortion, jitter, etc. etc. There are MANY examples. If you have any doubt that SACD and DVD-A sound better than CDs, you just haven't done enough listening. Respectfully. In the '90s I had a guy lecture me with great feeling that lamp cord could not be improved as speaker wire because "electrically there can be no difference." Even though wire is indeed the snake oil of the audio world, and many fantabulous and preposterous claims are made about it, he was still wrong about that. By now we know many of the *scientific* reasons why.

However, I am NOT suggesting in any way that you can't enjoy music on CDs or MP3s on an iPod or 78s on a Victrola. They all have their charms and their interest. Personally, I wanted to play around with HD for a while, to demystify it for my own empirical old self, and hear it with my own ears. My previous system didn't allow me to do that. That's all I'm talking about here.


*Subcategory "Perfect Sound Forever" in this case....

I'd like to ask, how long does it take for you to upgrade your audio system? sounds like an interesting "weekend project"... *if* a couple weekends is all it'd take.

Daniel S.,
Well, I've been at it for a couple of months already. But there are a few complications that make the game more, er, intricate: first of all, I have quirky tastes and too much curiosity, not very much money (although what I spend seems like a whole lot to me), and very few local dealers. I don't have the luxury of buying many things new, and buying used presents its own challenges. And I have a very poor space for listening, which creates other difficulties. And so on....


This is encouraging. I thought that the world had completely abandoned high quality audio. I find myself listening to my very compressed music on my ipod. CD sounds wonderful by comparison.

I am by no means an audiophile, but I can really hear the difference. CD was not a bad format, but what we have now is a major step backwards. I always thought things would get better. I'm glad to hear that I did not know the whole story.

Another great source for HD music is Bowers & Wilkins' Society of Sound.


Also, you can play FLAC and Ogg Vorbis files in iTunes by following these instructions: Fortunately, you needn’t convert Ogg Vorbis and FLAC files before you can play them in iTunes (though you have to gently fiddle with the FLAC files). Download the free Xiph QuickTime Component, install it in the Components folder within the Library folder inside your user account (create that Components folder if one doesn’t exist), quit QuickTime Player Pro and iTunes if they’re running, and relaunch iTunes. You should now be able to add Ogg Vorbis files to iTunes, where they will play normally.

To make FLAC files compatible with iTunes, download and install Cubic Fruit Design’s free Fluke. It installs components necessary for FLAC files to play in QuickTime-friendly applications (including iTunes).



Ignorance was bliss!

There I was, all pleased as Punch, sitting here, listening to (and visually admiring) my new bamboo Audoioengine P4s and N22 (plugged directly into the computer's headphone out with plain old 1/8" audio cable).

And then you had to go and write this...

Must postthis:


I messed around with trying to create a system that was tied into and controlled by my Mac, but in the end, I bought myself the Linn Integrated digital streamer Majik DSI (now DSM)... They have a couple of cheaper versions, and they have a couple of really expensive models, Akurate and Klimax.

For me, these are the perfect solution, they all pass "the tune test", and they are all completely digital... they weigh like a feather, compared to traditional systems.

Highly recommended.

[Hi Mike,
if you really want to move on from vinyl and mp3 and their limited resolutions,...]

Arg, please tell me how an analog recording on vinyl can have a "limited resolution". A generalizing comparison of vinyl and MP3 makes no sense at all.

Well one thing that didn't help HD audio standards either is that for the biggest part people listen to music in places that are not ideal for listening to music in the first place. A room with furniture, drapes, etc. creates a listening environment that absorbs sound in all sorts of different ways...most stereos have balance, some have 3- to 10-band graphic equalizers (stereos of the '80s couldn't be sold without it)...etc. etc. but none of these are helping.

Then there are all kinds of noises, what good is a signal to noise level of the esoteric variety in a city apartment block were the air con is hissing, the fridge is humming, the noise of cars filters through the cracks and a low frequency rumble of a distant freeway creeps in through the concrete superstructure. And not to mention the constant hiss of the stereo itself...spinning CD's....

The only way you can listen to audio IMHO is by using a set of really, really, really good headphones and even then, the blood rushing through your ears is a nuisance....

So high-end audio, I wouldn't bother too much with it...since music usually sounds better when your stereo really made a dent in your budget. But most of that is (if your measure it) a figment of the imagination.

BTW Arg...couldn't agree with you more, I've owned a 16 bit 16 channel audio recorder (AW4416 by Yamaha)....and given the right circumstances, listening to a band play on 16 channels sure beats a 2 channel format for depth and spatial resolution.

Greets, Ed.

One of the benefits (dubious they may be) of my 62 year old ears is that I no longer have to fuss over frequency response and "hi-fi" issues like I did when I was younger. The tinnitus covers up any issues with listening to only CD quality music!

@Dave - the objectivist only needs to run through a little thought experiment in order to see the subjectivist's point of view. Imagine a violin duet, recorded in stereo, where the engineer and producer have chosen to mostly direct one violin through the left channel and the other violin mostly through the right channel. When a violin plays a note the fundamental frequency is typically less than 1kHz, but the timbre of the instrument is carried in higher frequency overtones, up to tens of kHz. In our thought experiment we could imagine that the left violin is producing a strong overtone at around 25kHz and the right violin an overtone at 30kHz. To someone present in the room they would hear neither overtone directly because they are beyond human hearing, but if the two overtones arrived at the person's ears with a similar intensity they would hear the difference being an audible tone of 5kHz. This would contribute to the overall timbre of the ensemble. But if the recording and playback was at CD quality (with strong stereo separation) this 5kHz tone would be missing. If the microphone, recording equipment, storage medium, playback equipment and speakers are capable of rendering these high frequencies then the overtone would be correctly reconstructed within the stereo image. But limitations at any stage in the process can lead to the loss of those overtones. Recordings made using room mikes, rather than close miking of each instrument, and recorded without strong channel separation don't tend to have this problem, but the effect is real and can cause recorded music to lose some its subtleties in some circumstances.

Out of curiosity, Mike: have you ever tried to check your own hearing capabilities?

I am an audiophile as well, but I suffer tinnitus in the right ear, and overall I'm afraid that I could not listen to frequences well over 20,000 Hz, even if my system allowed me to do it. So when I read to ultra-high definition systems like the one you mention here, I quickly wonder if the limiting step could be the listening apparatus of the owner, rather than the hifi system itself.

"The uncompressed types are WAV, AIFF, and FLAC"

Isn't FLAC actually a *compressed* format?

All three are lossless, though.

Having built a nice collection of CDs since the late 1980s I've reluctantly spent the past few years researching new playback options and I've found HD downloads such as those from HD Tracks are very tempting. I haven't pulled the trigger on going the HD download route just yet because of lingering questions about transfer rates for computer architecture and such. At any rate, a few months ago I found this:

"24/192 Music Downloads...and why they make no sense."


I offer this only as an interesting read and not as a statement in either direction regarding hi-resolution audio files. At present, I listen to LPs and CDs in my family room at home, CDs and FM radio in my garage, and CDs and lossless files in my car.

One thought that keeps coming to mind regardng the analog vs. digital debate is that of the processes used to create pre-recorded CDs back in the 1980s what with the standardized SPARS coding for disks denoting the type of master media, the process, and the final transfer media.


Most older albums, of course, originated from analog media so they received the AAD code whereas some albums recorded during the mid to late 1980s were fully digital in workflow with the DDD code. At that time it was obvious the analog original media imparted its own stamp on the music while fully digital recordings were left with a rather cold and clinical sound, as if even the air around the musicians in the recording studio had been evacuated. I've often thought the Redbook Standard can simply reduce music, or more appropriately the recorded sound, to its very basics and it is this lack of sonic character and nuance which turned so many listeners back to analog.

So what of nuance then? Well, perhaps the most engaging stereo setup I've listened to during my 35 years of enjoying music was built around a Sansui 28wpc receiver I bought for $20 from a thrift store. Pure magic with a Garrad turntable with the now-banned Shure cartridge and old as the hills Fisher speakers. I'm not sure I'm a valuable resource for opinions on audio gear!

"Well, perhaps the most engaging stereo setup I've listened to during my 35 years of enjoying music was built around a Sansui 28wpc receiver I bought for $20 from a thrift store. Pure magic with a Garrad turntable with the now-banned Shure cartridge and old as the hills Fisher speakers."

Let me take a totally wild guess--the Sansui AU-555A?


One note of caution to anyone contemplating software which alters iTunes to play lossless formats. Apple has entirely rewritten iTunes and last week at the iPhone event they announced that the latest version of iTunes will be available in October.

This presents two problems. First, its questionable whether software like PureMusic will be compatible with the new version of iTunes which is a "ground up" rewrite. Before purchasing it might be wise to ask Channel D the maker of PureMusic. If you do already own PureMusic and IF it isn't compatible with the new iTunes you'll need a strategy for protecting your investment as Apple's software update will try to push the new version to you over and over again if it senses you're running older software.

I hope I'm wrong but its worth mentioning in case.

Dear Mike, Evan, and Hawkwood (and perhaps others),

Errrm, no, the "science" does not support a 44KHz upper limit. That's an incorrect understanding of the science that is also present in the xiph.org article.

The Nyquist criteria tells us that we cannot reproduce any frequency above 1/2 the sample rate. It does NOT say that you can accurately reproduce all signals below that frequency; it merely constitutes an upper bound.

The criteria only strictly applies to a constant, unvarying signal. In the real world, a signal that remains unchanged for at least 20 waveforms is a sufficiently good approximation. Then you can get away with Nyquist. When the signal varies in time faster than that (and music usually does), increasingly high sampling rates are required to preserve both the transient and the phase information.

Human ears are sensitive to both of these well above the continuous-tone frequency response. It's hard to explain why without getting into discussions of Fourier space correlations and other evil stuff. Best simple analogy I can make is to visual resolution. We can't resolve more than 10 lp/mm at normal viewing distances, but vernier acuity (visual transient response) is about three times as good as that. In blind (ahem) comparisons people can see the difference between prints with soft 10 lp/mm detail (e.g., a sine wave) and very sharp 10 lp/mm detail (a square wave). They'll consistently pick out the latter print as "sharper" although they won't be able to tell you why.

Phase response in the ear is a similar kind of thing. Our brains correlate phase data; it's part of how we determine the direction a sound comes from. We also correlate phase vs frequency (phase-harmonic) relationships. Also part of directionality and a big part of the "cocktail party factor" that lets us pay attention to one signal in a sea of background noise. This requires detection down to a fraction of a wavelength. You can't hear that fraction directly but, like vernier acuity, it alters your perception of the sound. We're also good at audio transient response-- what we think of as crisp vs mushy sound. It's a big part of why it is so hard to make recorded sound really sound like live.

Whether, in practice, you can hear the differences in your sound system is another matter. As Mike has well-explained, running comparison audio tests is very tough. If I might hazard a guess, perhaps a tight loop of a soft (so it doesn't clip) snare drum rimshot might work as a limiting test? That's kind of as different from normal music as a resolution target is from a normal photograph, but it might constitute a good quick and dirty test of detectability for one's own environment?

pax / Ctein
-- Ctein's Online Gallery http://ctein.com
-- Digital Restorations http://photo-repair.com

Dear Mike,

Two other FLAC players worth noting:

Audacity ( http://audacity.sourceforge.net/ ) is free open-source software available for many different OS's. It's primarily meant as a recording and editing tool, but it makes a perfectly fine FLAC player and supports up to 192KHz files.

Adobe Audition isn't anywhere close to free, but it's bundled into many versions of Creative Suite, so some readers will find they already own it. I reviewed it here:


It is severe overkill as a mere FLAC player. But if you've already got it, works great. It's what I play my virtual vinyl FLACs with.

pax / Ctein

regarding S/PDIF — most current Macs have it built in (coaxial, inside the headphone jack), and it supports high bit-rates; for those who have receivers with S/PDIF inputs, assuming they trust the DAC in the receiver, by far the simplest approach is just to buy the appropriate optical cable; there are also minimalist S/PDIF DACs with built-in amps like the Nuforce Dia for those who don't need or want a receiver

the DAC i use isn't S/PDIF, however, it's a type you didn't mention; i use a PreSonus Firebox, which is one of many FireWire-based DACs which easily handle HD data; OS X has built-in support for FireWire audio, so devices like this are plug & play, and are better than USB at avoiding contention with other devices (such as hard drives)

regarding FLAC & iTunes — when i need to work with FLAC on my Mac, i just use Max to convert it to ALAC which supports high bit rates

Max is free: http://sbooth.org/Max/

despite being called "Apple Lossless", ALAC is an open-source format, and supported by several players, not just iTunes; it can be converted to other lossless formats should the need arise; i use iTunes because i get a solid database for my large collection, plus a way to send audio & video around my house (via Apple TV and/or Airport Express, along with the iPhone/iPad Remote application) and also a tool to move tracks onto my iPhone

Dear Mike,

Just bough Duende through Qobuz. Not sure my equipment allows me to hear the difference — I have Future Sonics Atrios earphones, and my macbook pro's built-in DAC (which I believe does 96KHz and 24bits, but is probably a joke compared to proper DACs). That album was on my list anyways.

I saw a "gift" option when ordering — if you want, I could send one copy your way.

"Let me take a totally wild guess--the Sansui AU-555A?



I wish. No, it was the lowly 1000X but what a "full" sound it had.

Thanks for sharing your experience with your gear and your thoughts on downloads. I'd like to set up a PC-based music server at home so I appreciate the information.

Mmmm...OK, I 'll give you the Arcam (Stereophile Class B October 2012) but my favorite, the Cambridge Audio DacMagic Plus http://www.audioadvisor.com/prodinfo.asp?number=CADACMAPL works well and sounds great with computer audio as well as the output from my LG flatscreen.
Also, for Windows users, consider JRiver Media Center instead of iTunes. It manages all your media very efficiently AND it upsamples!

Oh, Jeez...I take some of it back. I just read the ad copy for the DacMagic Plus at Audio Advisor and it quotes KR (of all people) who raves about it.

Dear Mike and all,

it is sometimes possible to legally download HD music from abroad, if there may be region restrictions. It can be done with a VPN connection, mostly the pay-to-use ones, since you want to have some good download speed for the big files.

In my case it works pretty well. Based in Europe, I can purchase online HD music anywhere in the world.

Hope this helps.

best, Andreas

Two observations:

First, for the Windows users, good old Winamp supports high sampling & bit rates and most HD audio formats natively.

Second, before you go out buying expensive DACs, it might be worth checking if you need one. Modern build-in soundcards generally conform to the Intel HD audio standard, and can output HD audio without external help. Most use the Realtek ALC892 chipset, which can do anything up to 24/192.

All you need is connect the line-out to your amplifier with the interconnect of your choice, and see if HD-audio is the thing for you.

This is worth repeating:

"[..] any test of auditory stimuli that wants to distinguish only in terms of the auditory stimuli must:

- Have a falsifiable nature (i.e. be able to distinguish between perception and an actual audible effect)
- Must isolate the subject from changes in other stimuli than audio
- Must be time-proximate
- Must have Controls
- Must have trained, comfortable listeners"

Source: this video clip.

Nicholas: your test case using the two violins seems to address recording technique, specifically mic placement and has nothing to do with the obj/subj contrast. FTR I did not make any value judgement about the two "camps", just identified them and indicated my allegiance. We're talking about a harmless hobby here, it's all good (although I suspect the subjectivists actually get a great deal more pride of ownership out of their purchases than the other camp!).
Now if this was a pharmaceutical research forum things would be entirely different - but it's not :)

I've always wanted to have a good hi-fi set-up, but think I will have to be happy with my mid-grade know-little slop job of a stereo system with extra-thick 18 gauge lamp cord for speaker wires. I did purchase a Flac HD album out of curiosity after reading this, the re-release of Gaye's What's Going On, and unfortunately either my speakers/amp/headphones/ears/wire is/are sub-par, or my slipstream server with slim player compresses the file on the fly (don't think so). Whatever the case, compared to the 192k variable bit rate mp3 from Emusic, I really can't tell the difference, except that the Emusic file is louder. Interesting article non-the-less, and I'd love to hear what a high-end system sounds like.

I dissent. I've been an "audiophile" ('though I hate the term) for more than 25 years and during that time I've assembled a revealing, but highly musical system. Over a year ago, I acquired a DAC capable of decoding HD and began avidly collecting HD recordings. A few months later, I subscribed to MOG, which streams in mp3 format at 320kbps. While I believed that I could hear significant differences between my HD downloads and the MOG streams, I thought MOG sounded pretty good. I decided to do some blind ABX testing (using Foobar 2000) to see how much I was missing listening at 320kbps compared to my HD downloads. I planned a series of comparisons of HD and converted redbook and 320kbps versions of a dozen tracks using a good set of headphones, which I know to be more revealing than my good living room system. I gave up about half way through the test because it quickly became apparent that I could do no better than chance in identifying the HD versions even compared to the 320kbps conversions. Results of other ABX testing I've seen more or less confirms these results.

I draw several conclusions from this experience: (1) Perception of differences in sound reproduction is extremely vulnerable to psychological bias. Before the testing, in normal listening I believed that the HD tracks sounded better. The testing showed me that, in fact, I couldn't hear any differences. Having completed the test, I don't believe I can hear differences anymore. (2) While it's possible that some people may be able to (perhaps through training) hear differences, any sound quality distinctions between HD and lower resolution recordings is extremely subtle. They are easily swamped by much more significant factors in the audio chain such as recording quality, speakers, room acoustics, amplifiers, etc. Therefore, increasing recording resolution should be a low priority. The benefits of HD downloading (if they exist at all) are, in my opinion, completely overwhelmed by the enormous advantages of cheap music access available through services such as MOG (which has been a greater benefit to my listening than any piece of audio equipment I've ever purchased). (3) The criticisms of ABX testing heard from some quarters of the audiophile community are--at least as applied to comparisons of file formats--highly unpersuasive. The ABX testing software allows the listener to isolate and repeatedly compare the "X" version of any bit of a given track to the two known versions. You can listen as long, and as frequently as you like to any particular snippet before making your choice whether X is A or B. Having spend hours unsuccessfully trying to distinguish different versions of multiple tracks using this software, I simply don't believe that meaningful audible differences could remain unidentified. It's a humbling experience, but a very persuasive one.

Since conducting this testing, I've given up on HD downloads. In fact, I've given up on purchasing music. At a cost of about $5/month, I'm now able to explore virtually any music currently in print in any genre without having to worry that I might be missing something sonically. Good recordings streamed through my system offer all of the "audiophile" virtues--clarity, musicality, imaging, beautiful tone, strong dynamics, solid bass, etc. It's really quite wonderful.

I find the debate over resolution and format choices fascinating. I have heard great sounding Red Book and pretty mediocre 24/96 …and the reverse. I don’t currently have the ability to listen at 192 due to the limitations of my system. Given my 56 years, I know that my hearing is only good to about 14 or 15 khz on a good day. Still, I can easily discern differences when many don’t. My brain must be able to make sense out of the timing cues and other attributes that are spoken of less often than frequency extension.

The connection to photography is also useful. I can see very clear differences in the look of a print from one 12 megapixel camera to another. Color tonality, micro contrast and many other attributes DO separate the “look” that you get with one camera’s sensor vs. another. Reviewers are always struggling to put words to what they see …it can get pretty “woo woo” sometimes. However, they DO find differences and are simply trying to articulate what they see.

The higher end of audio has always been a bit of “woo woo” for most. But, that doesn’t take away the fact that there are very specific perceived differences that many people are able to identify. Hard to articulate sometimes, but the differences absolutely exist. Yes, there is the “placebo” effect and a host of other human perceptual issues that can influence perception. Still, I believe that we can safely say that all equipment does not sound the same even when it “measures” that way.

I was hoping for the definitive scientific explanation that lays out the “why” in these debates. It doesn’t exist because we still don’t really know all the attributes that point us in the direction of one approach vs. another. There is a great deal of speculation and even dogmatic rhetoric that accompanies those writings that try to “scientifically” describe what we hear.

I think that for me, it is a bit like the “Bull Duram” monologue: “I believe that:

- Less distortion is better than more
- Extended flat frequency response wins over “colored”
- Shorter signal paths vs. lots of extra boxes and wires usually sounds better
- More data (24/96) vs. less data (16/44.1) is usually better sounding (assuming the same master and a DAC that “plays” well with both) irrespective of the critics
- Transports DO matter
- Most rooms influence the sound more than the electronics
- Synergy exists in some systems and not in others (otherwise good components can sound bad when integrated with the wrong playmates)
- Bass sounds better with a larger driver
- Damping factor can create a sonic difference
- Sometimes discrete vs. IC chips sounds a lot better, especially at the more “pedestrian” price points …Burson products come to mind here
- Satellites “image” better than full range …if you can manage to fill in the bass properly
- Interconnects need to be at a “quality level” higher than what comes in the box. However, at some point, there is a diminishing return on the investment
- Speaker cables sound good and are less of an issue when the run is short making mono-block amps a good choice when possible
- Tubes can sound “warmer” and even “better” than solid state, but often at the expense of something (articulate bass, noise floor, transparency, etc.)
- Solid state is usually better than tubes when price is factored in, but the solid state product must be at a “quality level” way above normal mass market products
- “Grainy” digital is the Devil’s curse
- LP’s can sound really great (better than digital) or really bad …but cleaning them is a bitch either way
- Simple A/B comparisons rarely mean anything once a certain level of quality has been achieved …it takes time to really understand the nuances of what change has occurred …often months of listening
- Irrespective of what you read or are told by an “expert”, you must ALWAYS trust your own perception when it comes to high end audio and photography (and a lot of other things, as well)
- When it sounds “good” …don’t overthink it, just ENJOY!”

So, I will continue to enjoy these debates and the links to some fine reading (I really do appreciate the references). The elements that impact our ability to “hear the music” are important and meaningful to understand.

The comments to this entry are closed.



Blog powered by Typepad
Member since 06/2007