Jump to content
IGNORED

How to disguise science with snake oil !


cs

Recommended Posts

Bob, I agree this is a good debate although I must say I understand little of the technical arguements! But what you say about the way the appreciation of the sound differs from the science rings true with me.....and there is a lot more to it as well in my opinion. Why does music always sound better to me in the morning, for example.....why do I listen to a track one day and think that's just great, play it again and think hmmm, maybe not so great? Do ears need to warm up like hi fi components, etc, etc.

 

Also an interesting point you make about the potentially shrinking market. "When I were a lad" evenings were spent round at friends' houses listening to music/hi fi systems....Deep Purple, Pink Floyd on everything from Bush stereograms through to Pioneer (then the bees's knees) and even (the holy grail) a Linn Sondek! My parents had a B&O system which was the envy of all my friends. We now have generation ipod. I think it is inevitable that the only way generation ipod are going to appreciate good sound is through upgrading to good computer audio systems, hence the importance of this site. Vinyl and valves (although I have no experience of the latter) are aesthetically beautiful and can sound great I am sure, but inevitably I am sure it will become more and more of a niche market.

 

Phil

 

Link to comment

Ashley thanks for the kind words. You may be right. Sharing an e-cup of coffee is kind of tough, though.

 

Bob, you're probably not familiar with me but actually this posting was a departure for me and was meant as an offering to those who place a great deal of importance on measurements. Although I do own some relatively sophisticated measurement gear to assist in my speaker design, I have owned a high end store for many years and am a strong proponent of subjective evaluation and trusting yourself when it comes to good sound. We are coming from the same place.

 

Take care

 

Rick

 

Audio Research DAC8, Mac mini w/8g ram, SSD, Amarra full version, Audio Research REF 5SE Preamp, Sutherland Phd, Ayre V-5, Vandersteen 5A\'s, Audioquest Wild and Redwood cabling, VPI Classic 3 w/Dynavector XX2MkII

Link to comment

Hi Rick,

 

You're right of course, I am/was unaware of your industry interest. Those of us not 'in the loop' as it were can sometimes be disadvantaged by not knowing to whom we are talking, and of their expertise, but it can still be fun putting the proverbial foot in it, now and again!

 

I must take issue with you, however, over the coffee comment. Surely that would be an iCup, not an e-cup?!! I mean. 'e'-anything is just sooo yesterday!!

 

Seriously though, I love the passion of these disussions mainly because they highlight just how fragile meaurements can be, when it comes down to the all important 'system synergy'. Sure, a scientific starting point is paramount in the design of quality hi-fi and, without it, we would not progress. But when it comes down to choice and what sounds good, science HAS to take a back seat to what my ears/brain tell me. I have listened to many bits of kit over the years and now know roughly what I like and what I don't. And I don't care if I'm the only one left in the world - more than one speaker in a box is just wrong!!

 

But there again .............!! And it's that 'BUT' that keeps us all on our toes and makes for a much more interesting hobby!

 

Keep up the good work and all te best to you.

 

ps - And I don't like Apple Mac's either, or Linux. I am beyond help!

 

Link to comment

Hi Phil,

 

You and me both! Those were the days - first time I ever heard Jimi Hendrix' 'Hey Joe'! Distorting it's way to me via a mono valve radiogram - all 1w/3" speaker of it! The arm used to jump in time with the bass guitar riff! And the needle doubled up for darning socks!

 

Well, maybe that last bit was a tad made up! But music was about as low-fi as it got and almost as exciting as it's ever been. Makes you wonder doesn't it, when we sit drooling over pictures of the latest £10,000 super Dac!

 

Ah well - as they say, the older we get the wiser we get. It's just that, how old do I have to get before I start getting my share of this 'wise' stuff!

 

Best hi-fi ever? - Jimi Hendrix - Live - Isle of Wight, 1970. Now that's what I call distortion :) !!

 

Link to comment

Bob,

 

Right on! You were at IoW in 1970 (or are you refering to a rcording)? I would only have been 11, but my favourite band when I was a teenager (Free) played there - think it was 1970 - some released on DVD now. Did you catch them (if you were there)?

 

My favourite rock performances of all time are captured on the Woodstock footage of 1969 - Joe Cocker & Santana. There's poor sound quality for you, but still knocks me out. The cinematography of the Woodstock movie is still wonderful, mind you.

 

...but we digress from snake oil - where did that term come from - never heard of it before I subscribed to audiophile forums?

 

Phil

 

Phil

 

Link to comment

Back in the 19th century, hucksters who sold miracle cure remedies off the backs of wagons were often referred to as "snake oil salesmen." I assume there must have been a common product reported to be made with snake oil that they were identified with.

 

One more thing, back OT. I trust my ears over measurement. There are things that will show up in testing that can't be heard/don't have a significant effect on quality, etc. An old SS amp or a tube amp really can sound better than a modern one with incredible specs. Where I take exception is with the audiophile world's rejection of AB/X listening tests. If we truly trust our ears, we should welcome such tests, not make excuses for their lack of validity. Psychological bias is a very real, well-documented phenomenon. AB/X testing is a proven way to take it out of the equation. If we can't hear something often enough to make that hearing a statistically significant event, it is, at the least, insignificant. If a whole bunch of us over multiple trials can't hear it, it is probably not audible.

 

It's really that simple. If we have such great trust in our ears we would invite the confirmation. I think we doth protest too much.

 

Tim

 

I confess. I\'m an audiophool.

Link to comment

Thanks for the info on the origins of snake oil, Tim. Very interesting. Someone told me once of a "blind" test of CD players in the UK, which included several high end brands. The panel (drawn from the general public I believe) picked a £100 Pioneer as the best sounding. The organisers were shocked and discounted the results as refelecting the fact that the non-audiophile panel did not know what to listen for. Hmmm.......

 

BTW is that you or Van Morrison in the pic?

 

Phil

 

Link to comment

I have correct this misnomer of scientific theory - and sorry if I am late to the discussion and taking it back to an area that has been covered already, but I would like to add my 2c. And this is not meant to offend any faith, faith is not science, and I'm discussing science.

 

In scientific terms, evolution is a fact, demonstrated by masses of evidence such as fossil records, e.g. the visible changes in bone structure of reptile jaws as they become ear bones in mammals, or more recent evidence such as the changes seen in butterfly colouration during the English industrial revolution. Argue why it happens if you must, but it happens. We can even make it happen ourselves through selective breeding.

 

Are there gaps in the fossil records? In fact they are pretty complete, but even if there were gaps, not having 100% history of the world wouldn't make the rest of the evidence invalid. So far we have Darwin's theory of why evolution happens, it has evidence supporting it and it has undergone peer review. It seems to be a good explanation. It may be wrong and there may be aliens tweaking species from time to time when we aren't looking, but so far no one has found any evidence for that. If you really want to argue against Darwin, some evidence to support an alternative would be a good starting point and believe me, scientists love nothing more than disruptive evidence that rocks the boat and means the beginning of a whole new line of research. It's why they build things like the LHC.

 

To be precise, a scientific theory is an explanation of fact. Theories may change, as with subjects like Newton's laws which don't work at subatomic levels, or more recent theories like Einstein's that have made way for string theory for quantum physics and cosmology. Theories may develop, but the facts or process they are explaining do not alter, even if they may be better understood or measured. Nor does that make a scientific theory a best guess, it requires peer review and evidence. It's this lack of understanding of the rigor of scientific theory that fuels so many heated discussions. They aren't just ideas, they have to be carefully considered concepts backed by solid evidence to pass muster.

 

This applies all science, not just those areas where faith gets involved. We are discussing audio technology here, based on theories of solid state physics, and the way in which quantum physics works is still not understood entirely, but we accept the facts we do know of electronics, and without it we'd have no computers, or stereos and no internet to be discussing this. We don't dispute our doctor's advice because not all of human biology is understood, nor do we refuse to get in a plane when we know that aeronautical theory doesn't have all the answers. We trust those scientists, so why have some regard for all scientists who follow the same rigor and peer review?

 

But to get back to audio, I personally got tired of audiophiles and the magazines ignoring evidence and science when it came to reviewing technology and easily measurable effects. They would conjure up romantic and flowery language to disguise their lack of process and rigor and an unnerving desire to recommend based on price tag. More than that, they started making pseudo-scientific judgments on aspects of hardware, such as hearing data loss in digital networks, where clearly such an event would be catastrophic for computers in general. When you read that WiFi is sonically inferior to wired cables, or lossless formats sound worse than WAV, you know a wheel has come off the audio bandwagon.

 

Having spent thousands in gear. I eventually settled on mid-fi and digital reproduction via first a squeezebox and then a macbook and just started enjoying the music. It's very easy to lose sight of that, recorded music is never true fidelity, but a studio artifact, so chasing the dream of ultimate reproduction is simply trying to recreate an artiface that never existed outside of a soundproofed room. Today's gear is more than adequate at reproducing popular music in high enough fidelity for most of our aging ears. I might have been able to tell the difference between esoteric brands of high end gear years ago, I know I can't now, so I won't try to fool myself.

 

Link to comment

Hi Paul - Thanks for your very thorough and well articulated post. One thing I disagree with is your statement:

 

"When you read that WiFi is sonically inferior to wired cables, or lossless formats sound worse than WAV, you know a wheel has come off the audio bandwagon."

 

I have proven to myself that the wireless Apple TV into my Berkeley Audio Design ALpha DAC via Toslink is inferior sounding to my Lynx AES16e card into the Alpha via wired AES/EBU. In a blind listening test I identified sonic signature of the Apple TV 8 out of 10 times. I don't care for the sound of current wireless transports and to me they don't sound as good as wired methods.

 

If you use scientific measurements there is a measurable difference between the wireless Apple TV and the wired Lynx/Alpha DAC. Thus, science and my ears agree. Science measures a difference (jitter etc...) and my ears hear a difference.

 

Let me know your thoughts on this. I am very interested in your opinion as you as obviously a smart guy.

 

Founder of Audiophile Style | My Audio Systems AudiophileStyleStickerWhite2.0.png AudiophileStyleStickerWhite7.1.4.png

Link to comment

Paul - fantastic post - I am a geologist by training myself although I have forgotten most of my "pure" geology - I am now a professional hydrogeologist. I could not agree more with your observations and I myself have gone down the same route now using a Macbook, etc. I have rediscovered enjoyment of music after years of fussing about components and much of my enjoyment comes from use of an ipod set up, something I had always dismissed on the lines "it would never sound good". It does, with a decent headphone amp and phones, and compliments the Mac home set up perfectly.

 

Chris - I use wireless via Airport Express and can't tell a difference to it being hard wired via toslink. Also, I thought it was agreed the AE is bit perfect to 16/44.1? What are your views on this?

 

Tim, buy yourself a hat ... treat yourself ... you might as well as we will all have no money left soon....

 

Phil

 

Link to comment

My point about the comparison was that you can easily validate that the data received is the same as the data sent. If it weren't wireless networks wouldn't work. A file goes in one end and comes out the same the other. The text I'm writing will be the same when it appears on server that creates this page.

 

What you describe isn't the same as I was saying, you can hear the difference caused by a measurable phenomenon, jitter. I'm talking about reviewers who swear that they can tell the difference between a lossless file and a WAV file when they are decoded by the player, or when the player uses WiFi instead of wired connections. The data *is* the same. The output from the player is the same, it's using the same internal hardware to decode the same data, likely from the same buffer.

 

So in your case, the comparison would be using the AppleTV to play back music from another Mac library first using wireless connections to the music server, and then wired. One will be more reliable than the other, but better sounding - that simply does not make sense. Unless the gear was so badly designed that power supplies drooped or interference was not suppressed adequately, the incoming data when decoded is the same and it goes through the same DAC.

 

Link to comment

There was a time, I too, was convinced that not all we can hear, can be measured. Indeed, I am not an engineer. However, since I am working on XXHighEnd, these thoughts have changed drastically;

 

For instance, on the cable matter (and I didn't look at that video) this is so easy. That is, for LS cables it is. Connect a scope to the output of the amp and to the input of the speaker terminal, and there you go. There is no way you can imagine the speaker can produce music, looking at the degradation of the wave towards the speaker.

 

A one zillion things are wrong with audio

 

This is my motto for a couple of years now, and subsequently I solved one or two myself. And you know what ? at my older age I *still* didn't do anything about my cabling. I always make them myself just to get them at the proper length, and I never bothered about the quality of that. Remember, I know the results of "random" cables (explained above), and might one be better than the other ... it is unimportant ! Why ? because there are still the other zillion minus 1 things being wrong ...

 

I got into the strange habit of trying to squeeze out that wave as it resides on disk, from the speaker. I know, a complete clinical approach, but since I learned that the better this is achieved, the better color the sound receives (oh yes), I kept on hunting for that.

One thing I can tell you : when you start off with not following that wave precisely, all what happens after that is amplification. Amplification of the error.

 

I think, what we all should realize, is that the "music" as such lives by the harmonics. You could well say that when harmonics are not there, all instruments sound alike. Those who create instruments by means on synthesizers will understand ...

Those harmonics are the most fragile tiny (wavelength) pieces which must fit. Which must ... harmonize. A zillion things can be improved on that. Err, *should* be improved on that.

 

BEEMB :

Whilst we probably can't hear certain low frequencies, do low frequencies have the ability to make a differences to frequencies above them ?

[...]

Do lower frequencies help higher frequencies travel ??

 

"Help" is one (maybe not correct) angle, but that they just do is another.

In order to see this, one must twist his mind a bit. I mean, sound waves "travel" in 3D space, while the medium they are stored on I think always is in 1D (I'm not sure about viny actually, could be 2D).

Now, looking at the visualisation of the wave file (which for sure is 1D), you can easily see in there, the higher frequencies ride on the low. This is just how it works physically in the whole chain, up to the loudspeaker driver (think on doppler distortion in the driver, and you will recognize it from that angle).

 

For me there is no reason not te believe that in mid air the higher frequencies again ride on the low frequencies, although this might be more a matter of interaction, also knowing that sound waves travel longitudinal, which is *not* how they are represented on storage media like vinyl or the visual representation of the digital wave for that matter. Anyway, look at a water wave, and there you see it happining, although here too this is not longitudinal.

 

Thinking of sound travelling by means of compression and distraction (I hope I say it correct in english), you can imagine that low frequencies influence higher frequencies. You could say that they get blown out of the way more or less. Anyway, usually adding bass audibly influence the higher frequencies, although it is hard to determine whether this is because of the higher frequencies get overvoiced, or because things just better fit.

 

Better fit ?

Oh, I couldn't emphasize more on this. I am 100% sure the best musical representation through loudspeakers is achieved when *nature* is achieved. This again comes down to the harmonics I mentioned, and the very fragile "fit" that may happen in your system and room.

 

Btw, I am not blattering complete nonsense here, and telling you that I designed a local position system (like GPS) with an accuracy of 0,1 mm in a 300m2 3D space (working with phase difference from light waves), and a working prototype from that with ultrasonic sound waves ... could kind of prove what I'm working with. You know, listening to sound is just the other way around : at certain positions in the 3D space, waves meet and interact (double, phase out) and that is the music we are listenining to.

 

Back to the bass waves influencing the higher frequency waves ... as I told elsewhere, from experience/knowledge I could reason things another way around once I heard "gaps" at a certain frequency : I added SPL in the higher frequencies in order to fill out those gaps, and it just worked.

 

Since all my work is focused on these phenomena, I also know that what you hear for results, can NOT be measured. However, this is for an unexpected reason (maybe) :

If I *claim* that the interaction of the individual frequencies are of the most importance, then how would we measure this through loudspeakers ? A sine sweep won't do the job, nor would pink noise do.

Instead, we have to measure the sound of a violin etc., because only there the special interaction takes place. And when we accomplished this impossible job, we can start with the next instrument.

 

For all this, it is important to know that any influence you might have on the sound hardly can be applied in a linear fashion. Take jitter as an example : it can be seen as a resonance (like undithered downgrading from bits), hence the error is repetitive and therewith influences sound. However, since it is a resonance, you never know where the highest impact is, and it is dependent on the music as well. I think it is commonly known that with changing jitter *or* the bass gets better *or* the highs get better.

The resonance just shifts to other frequencies, and create their own frequencies (distortion) overthere.

 

If you indeed might believe in this snake oil story, it is rather easy to see that we don't know of any means to measure the musical effects. The only thing we can do is measure the basic elements (like jitter can be measured at each frequency), and think less jitter gives the best result.

Things are much more complicated than e.g. just having a low amount of jitter; What would sound best ? 1000ps of jitter throughout the whole spectrum, or 100ps of jitter at 1200Hz only ? I don't know the answer, but I assume the latter sounds much more annoying.

 

So, the nos-DAC measures much worse, right ? yes it does. But following me with my stribing for 1:1 following the waves it is impossible that the oversampling DAC is able to do that. Now, why half of everyone thinks nos sounds better ? because it measures wrongly ?

So far, I never saw anyone explain in a decent manner why the nos-DAC sounds better. I do (all over, so I won't repeat that here), and the moral is : it is all one big pile of reasoning about things which cannot be measured but for their elements, and it comes down to which of those few things have prevalence out of those zillion things actually being wrong in the first place.

 

Yes, I am convinced a cable can matter. But not before many other things - those being much more important - got out of the way. Remember, a first thing a cable may do is filtering. Filtering out those other anomalies ...

 

Peter

 

Lush^3-e      Lush^2      Blaxius^2.5      Ethernet^3     HDMI^2     XLR^2

XXHighEnd (developer)

Phasure NOS1 24/768 Async USB DAC (manufacturer)

Phasure Mach III Audio PC with Linear PSU (manufacturer)

Orelino & Orelo MKII Speakers (designer/supplier)

Link to comment

"Now, why half of everyone thinks nos sounds better ?"

 

Is this in blind tests or when audiophiles are expressing opinion? In blind testing, approximately half of the subjects hearing a difference means there is no real difference being heard. If it is audiophiles' opinions, it may be even less meaningful.

 

You gave a detailed and well-thought-out treatise on how that which can be measured, but should be outside the range of human hearing, and that which cannot be measured at all, might still affect what we hear. It sounds much more scientific than "it is musical," but it still lacks the necessary and easily obtainable evidence. Using your own example, do you know of any blind listening tests in which a statistically significant number of people, beyond the half stated above, could hear the difference between similar nos and os DACs?

 

Because it doesn't matter if humans can't hear it.

 

Tim

 

I confess. I\'m an audiophool.

Link to comment

I'm not an engineer, but I am surrounded by them. Therefore I'm not allowed to make statements about things being audible that aren't measurable, in fact I'm regularly told rather emphatically that you can measure a great deal more than you can hear, that it's very easy to show how easily fooled the ear is. Follow this experiment and all should become clear: http://www.mrc-cbu.cam.ac.uk/~mattd/sine-wave-speech/ and that's before you factor in preconceptions based on price or prejudice (rife in audio!)

 

We've been able to show that difference DACs sound different with different sources and different amplifiers and we believe we know why. It's because not all DACs cope well with jitter yet and most DACs have a little bit of RFI on their output.

If this is the case, then it's very difficult for anyone to decide what's influencing the sound of the system he's building. He might think he's orchestrating "synergy" by assembling together a disparate group of components whose faults may complement each other if he's lucky, but he won't be doing as well as a decent engineer and the right test equipment.

 

Measurement are only flawed when they are made by people who don't have sufficient knowledge all the factors that are necessary to draw the correct conclusions, or so engineers we know will argue.

 

Science dictates the the digital output from a wi fi device or from a computer should be the same, therefore if there is a sonic difference, the most likely explanation is level differences. Any blind listening test done by us, using ears others than our own who don't know what they are listening to or for, is conducted with instant switchability of exactly matched source components because it's well known that as little as a 1/10 dB difference will bias judgement in favour of the loudest of the group.

 

Measurements in audio have suffer terribly from the people making them IMO. They mean all if they come from the best engineers.

 

We live in an age when it's "cool" to say "science doesn't know everything", however more often than not, it's a statement made by people who don't know much else either. They just believe.

 

It's my belief that if these discussions show healthy respect for science and engineering that they will attract scientists and engineers who have much to contribute, but have stayed away from audio in droves for decades, rather get embroiled in spats with those locked in to the more lunatic aspects of audiophilery and who may not understand rational objection to irrational belief.

 

I suppose I'm saying that we should look and eliminate for logical explanations before we assume there are none.

 

sincerely

 

Ashley

 

 

 

 

Link to comment

During our experiments we've built a number of DACs where it's possible to change the sample rate whilst listening. We were curious, but unable to hear any differences, so we asked younger and more acute listeners if they could. They couldn't either, in fact they couldn't even hear one when we applied a brick wall filter at 16 kHz! We tried this on quite a few people too.

 

The problem I see with these comparisons is that there are a number of factors that could screw around with the result, whereas in our case, it was simply a question of making one adjustment and leaving everything else the same.

 

I hope this helps.

 

Ashley

 

Link to comment

"...in our case, it was simply a question of making one adjustment and leaving everything else the same."

 

Which is, of course, the way any valid test is conducted. At the other end we have my recent experience, which involves spending many hours a day in the presence of lots of pretty good audio equipment and a switching system that allows me to A/B to my heart's content. The sources all have DACs, which are used for the stereo amps. Most of the amps are 5.1 and 7.1, include their own DACs and take digital from sources, so those DACs are in play. Which means every time I switch amps (receivers, actually - Pioneer Elites and Denons), I switch DACs as well. And that, of course, means I don't know if the differences I'm hearing are the amps or the DACs or a combination of the two.

 

I really don't think I could overstate how meaningless it is in actual listening.

 

Why? Because if you go through and equalize the volumes between the amplifiers and run them all in straight stereo so there are no processors engaged, turn on a pair of speakers that the least of them can drive well enough (most in the room), then start pushing the buttons, the differences are so insignificant that they fall into the "I think I hear...." realm, into the realm where civilians, even civilians who love music and listen intently, will not care.

 

Now, switch modes, keep the amp/DAC the same and start switching speakers. You'll very quickly understand what matters.

 

Tim

 

I confess. I\'m an audiophool.

Link to comment

Tim

 

You have a very good point! It always amazed me that people would buy better and more and more expensive amplifiers and CD players to drive older and quite often awful loudspeakers. I think it's because small, incremental improvements in sound quality are easier to recognise and accept than the dramatic ones that speakers can often give.

 

The experiment I provided a link to in an earlier posting shows how we (subconsciously) process the information we hear from our hi fi so that the sound seems excellent to us, but may not to others who visit. Before we can accept a big change/improvement we have to realise our present system isn't up to scratch and even then, some people have trouble, which is why I think turntables are still with some of us.

 

Ash

 

 

 

 

Link to comment

Agreed Lee.

 

I heard from so many people this weekend at RMAF2008 that said Computer Audiophile really has something going with the constructive tone of all the readers and the lack of "pissing matches."

 

I must thank everyone for all the restraint and very solid posts thus far, especially in a thread that has the term snake oil in the title.

 

Founder of Audiophile Style | My Audio Systems AudiophileStyleStickerWhite2.0.png AudiophileStyleStickerWhite7.1.4.png

Link to comment

Chris,

 

The other admirable thing about CA (and a refelection of the way you run it) is the way starters like myself are embraced and given fantastic, neutral advice without any hint of condesention - long may it continue.

 

Phil

 

P.S. Have you ever thought about having a page where folk can buy and sell used gear. Put together computers and audio equipment and there must be a lot of turn over of gear. Or is this just too much hassle?

 

Link to comment

Haha, great; After reading your posts Tim and Ashley - and not following the link from Ashley at first, I thought to write the below, but then followed the link afterall. Well, my text below doesn't change from it, but I guess I use the same subject the opposite. :-)

 

[better skip this post, because again it is too long and mostly off topic. Sorry.]

 

Tim, I can be wrong of course, but it looks like you have the perception that feeding a system with higher frequencies and/or being able to perceive those higher frequencies better, is the heart of the matter. Of course you go against that by stating that most people don't hear the difference, and people of "our age" (hehe) won't be able to hear above 12K anyhow. My (maybe vague) point is : it just is not about that. I mean, a well performing audio chain is not about being able to show the higher frequencies (I hope you will understand better what I mean after having read this whole post).

 

*If* it is about higher frequencies afterall, it is merely about how those higher frequencies integrate with the other frequencies, but I think this is a slightly different matter. Example : a widebander like a Lowther which stops at 13K in a nice horn system to me is as dull as hell. It just doesn't work. Add a tweeter, and I am fine. Still I don't hear above 12K ...

The same stories float around about adding ultrasonic tweeters which for sure nobody can hear. However, they do influence the sound (I never tried that, but believe the stories).

 

Right. Where above is a more or less valid example about higher frequencies doing a job or not, I say it is not about this at all. It is merely about resolution i.e. the way the system is able to follow all changes in amplitude, which most certainly is not equal to being able to follow a 20KHz sine. Or a 12KHz sine for that matter. So here I come again with my "square" hobby :

 

If a speaker driver is able to follow a 20 KHz sine, it doesn't have to do more than smoothly push out and pull back the diaphragm. Nature even helps with this because at maximum (desired) excursion, it will go back automatically. Ok, 20,000 times per second, but no big deal nowadays. But :

Let's make life easy for understanding, and take a recording of a synthesizer, which *IS* able to direct perfect square waves into the chain. How do you think a diaphragm must (de)form to ever get out those square waves ? (btw, do not think this is impossible because it just is -> I captured the (of course not perfect) square waves by microphone once output by a speaker). Well, you get the hunch : the faster the driver is, the better this will work. But where are the specs that tell about this (resolution) ?

 

So, with my big hobby in the foreground, something is lacking in audio chain specs : the resolution. Or ... the fastness. Indeed, you now recognize that the only part of the chain dealing with fastness are amplifyer manufaturers. And you know how they present it : show an xxxxKHz square wave going through it, and how less it rings etc. ...

 

Never think square waves do not exist in music; it is full with it !

 

Using your own example, do you know of any blind listening tests in which a statistically significant number of people, beyond the half stated above, could hear the difference between similar nos and os DACs?

 

The fun is, this is not necessary. So, on to the title of this post, it just requires the knowledge of what to listen for. Or more funny : what to play to squeeze out the difference best. And some technical insight of course;

Things which have profound square elements in them : trumpet and in fact all instruments that produce sound through vibration (dispersion) by lips or reed (? like sax) in the base, timbre, any very fast transient like small bells, ride cymbal (at the tick) and cymbals in general depending on how they are hit, etc.

So, everything with a most steep transient, and that is where the attention should be when listening to a nos DAC. And remember, the oversampling DAC will round squares into sines. -> Keep in mind that the captured sound of whatever squarish part of the music never can be a 100% square because that cannot exist electrically. Thus, the smallest rounding in the corners of the square will be rounded more and more and more at x times oversampling.

 

Am I still on the subject ? yes I think so, because in the end all comes down to knowing some stupid theories, and the urge to get the waves the most 1:1 from the speakers (and please keep in mind, this just works for the better -> the more I can realize that, the better the sounds get, really).

 

The reference

 

At tweaking at the level I do, and where software versions show various elements one by one being better, once you heard such an element at his best, you remember that (this is about what I said elsewhere, that it is so important to being able to remember such things from even months ago). Examples are : virbrating bass strings (versus bass sound), color in cymbals, normal cymbals sounding like Chine cymbals, straigtness of strings and woman voices, deeper bass, distorted "sounds" (like all JVC XRCDs do !), timbre, the presence of standing waves (mostly known from the under 40Hz area, but just as well present in the whole frequency range).

 

Each of these can show at their best and at their worst. Thus, in the end it is "just" a matter of bringing them all together at their best.

 

All these elements can be listened for, and yes, you are listening technically then, which, to my experience, most people can't. Most people listen to the whole, and of course it is about the whole. However, there is no way the elements can be improved by listening to the whole; you wouldn't know what to do, and what comes from it is that a cable might improve ...

 

Believe it or not, but most of these elements show (for me) within two seconds really. Not because they show in absolute sense, but because I have the reference. E.g. if I would be in a random room like "yours", within seconds I can tell whether things are ready for improvement. Key always is the standing waves or buzzing sound. -> buzzing : waves which meet and add up or phase out where they should not, "buzz" through the room and emerge at random places. If you have this while using my player, it is 100% guaranteed that you have an impedance problem. I know, you won't believe this, but for me it has become a very handy absolute measurement device. A reference (which works everywhere).

 

An example in another leage is the XRCD. As said, I can easily hear this, which obviously began one time, and which kind of things btw often emerge from a user complaining hearing anomalies, for me to find out what it is. Well, look at the wave of such random XRCD. It is one big pile of clipping.

This is in another leage, because most probably "you" never complained about an XRCD, and the only thing you might have noticed is that it sounds more saturated. Well, this could be a complaint by itself, because music shouldn't be more saturated than it was in reality. It's just a hoax (and I'm sure we agree upon that one). But supposed you don't hear the distortion, please take some time to look into the wave and tell me : *should* you just not hear the distortion ? and if you don't, is that a virtue of your system ? And no, "I don't hear it, so we can't hear that" is no answer, because I can, and it is not related to personally inaudible high frequencies. Again it is a reference, and if I'd point it out while both listening, there is no way you can avoid it the next time. Perceptual insight ...

 

I think on this forum I referred to this link earlier : http://www.phasure.com/index.php?topic=88.0 (Test track with distortion which isn't there...)

This is such a beautiful example of finding a "measurable" piece of music which shows what harmonics should do, and how they are destroyed just by stupid software.

It also shows at which level of detail one should listen, to perceive what will be (!) going on at the larger level (meaning : throughout).

From this I learned how the harmonics of one instrument can destroy the harmonics of the other when things are not right, and this too is perceiveable within seconds *with the proper music*.

 

When things really start to be allright ...

 

This is key !

For me, just over two years ago, music started to emerge from the speakers of which could be stated that no further tweaking was necessary to eliminate disturbing things.

Most probably everybody except a few maybe, are still working on eliminating disturbances.

Think of harsh highs, shouting ladies, standing waves, and in general, a system which just has a maximum on the volume knob, because otherwise the music starts to disturb one way or the other.

 

The above is far behind me, and instead I now concentrate how each instrument or voice can be made even more real. Cymbals do not hiss anymore, and instead they are not loud enough or have the wrong color, could be more metal, etc.

This is so, so, so much more easy to work with, you have no idea ...

 

 

Of course I must add some On Topic lines now;

Not all snake oil is snake oil I think, but often it takes "insight" we don't have, and things can be too complicated to judge. But there are two sides, and here again a reference is in order :

 

I do my very best to judge my own system against yours, and I don't even know what systems you have, let alone that I know how they sound. But, since you all will be having commercially available systems, the same I can listen to in high end shops everywhere, or what could be heard at the past RMAF for that matter (where my spies fly around :-) I tend to *know* how your systems can change for the better. This is one side.

The other side is the oter way around : if the guy with the snake oil cable etc. tells us his cable is better and even the best, how would he know how this turns out for our systems. Or at least that would be my perception.

 

The real story of the latter is this :

150m from my house the respected mr. Van den Hul lives (known from The Frog and also from his cables). Mr. Van den Hul is pro-pro-pro vinyl, and a few months ago I made the agreement with him that I would be able to convince him that vinyl is history really (those sessions still must be held). I can tell you that the Van den Hul company is far from a fraud, but don't ask me whether a 500 euro cable should cost 100 really. This is a slightly other matter I think. But :

While Van den Hul does not sell snake oil, there is no way even one brain cell in me (might I have one) will think these cables are going to help me. Why ? I know his reference !

I was there at the meetings with the 30K turntables and 70K speakers and all. And it was NOTHING.

 

But was it nothing really ? every visitor raved about the systems listened to. And hmm ... 3 years ago I think I would have liked what I heard there. But not anymore. My reference has changed.

Now what must I do with a cable which has systems for a reference I don't like (at all) ? I would call that snake oil, or possibly a fraud. But is it really ? it depends on the receiver I'm afraid ... It even depends on the seller who genuinly may think he has the best while the world around him proceeded faster than him.

The SQ of audio playback currently is improving as fast as it never did before. It needs computers to do it, though. Today for many this is snake oil by itself. Four years ago I would be shot when even starting the PC and audio subject.

 

Peter

 

Lush^3-e      Lush^2      Blaxius^2.5      Ethernet^3     HDMI^2     XLR^2

XXHighEnd (developer)

Phasure NOS1 24/768 Async USB DAC (manufacturer)

Phasure Mach III Audio PC with Linear PSU (manufacturer)

Orelino & Orelo MKII Speakers (designer/supplier)

Link to comment

paulb,

 

I'm talking about reviewers who swear that they can tell the difference between a lossless file and a WAV file when they are decoded by the player, or when the player uses WiFi instead of wired connections. The data *is* the same. The output from the player is the same, it's using the same internal hardware to decode the same data, likely from the same buffer.

[...]

but better sounding - that simply does not make sense.

 

You wouldn't know how much sense it makes ...

 

Chris, allow me to take the opportunity to feed this thread by real snake oil. That is, if you want to appreciate it like that. And if so indeed, just throw out this post (but it would be better and more honest to first try things out then).

 

So here is a snake oil salesman, and it is ... me.

As people might know, I started XXHighEnd because

a. I was convinced that computer playback could eliminate the baddies from CDPlayers

b. All software sounded different, and none sounded right. Besides that, nothing sounded "stable", hence what was okay one day, was not the other. Generally most albums did not benefit from PC Playback.

 

Thus, I started to write my own player, so at least I would know what was in there that would molest sound.

Surprisingly enough the first version (which ran on XP because Vista wasn't there yet) already sounded better than anything else, and everybody agreed upon it.

Because the way I set it up (playing directly from memory) the downside was that it could use "unsigned" RME drivers only to be "bit perfect", and it was not stable with files above 64MB because of those drivers not 100% compliant (unsigned).

 

Then came Vista, and despite all in there was half cooked (which it still is) it gave me great opportunities in de audio area. Tough job, because nobody had done it before, nobody to help, and again, the means to use were half cooked.

 

From that point on I learned that software could influence sound (not to forget, all is bit perfect, or IOW what is spit out digitally can be read back and compared with a 100% reault). And moreover, I learned how to influence the sound, hence in what direction.

 

On a downside, I learned that not *everything* was under my control, and when something in the user interface was changed, that influenced sound as well.

With that knowledge as a base, I created a playback means which is totally independent of the user interface (so called Unattended Playback), and from that point on things started to roll (fast) and I always said that SQ improved by a factor of two each 3 months or so.

Right now it has been over 6 months ago that SQ improved hence I paid attention to that, which is merely because many functionalities had to be created and improved as well.

 

All 'n all software can improve sound, and all is as bit perfect as can be. Yes, if something is snake oil, it is this, and it works for all DACs except the one or two that won't be influenced by software (hence the PC environment).

 

This snake oil can be tried for free. That is, the player runs for 30 minutes on average (between 6 and 60 minutes), and it just needs a restart to have another go. The only thing which lacks in the trial (which works forever) is "Processor Core Appointment", which does matter to the sound and worksout differently for everyone because it highly depends on the configuration of the PC and the processor used.

 

Might you want to try this snake oil (which you really should just because of my own enthusiasm :-) take your time to get to the merits which can't be perceived in one hour unless you are very experienced. Note that you need Vista for it, and must use Engine#3.

I most certainly do NOT want you to buy this snake oil, which even is beta snake oil. What I do want, however, is all the input possible to improve SQ further; Grab the opportunity, already for yourself.

 

-----

Well, this was my attempt to throw some snake oil on the thread fire. Possibly you don't appreciate this funny meant combination with something which typically will be looked at as snake oil, but since CA is about audiophile computer playback I thought maybe it is legit to post this. If not, throw it out.

 

I would love to see responses of the kind that indeed this is snake oil, hence everyone believing in it is placebo'd, me on the first row. Good fire for the thread I think.

It is *not* a good idea to start asking questions about the ins and outs or settings of this special snake oil, because I'm sure this thread, nor this forum is meant for that.

 

Peter

(http://www.phasure.com/index.php?topic=592.0)

 

Lush^3-e      Lush^2      Blaxius^2.5      Ethernet^3     HDMI^2     XLR^2

XXHighEnd (developer)

Phasure NOS1 24/768 Async USB DAC (manufacturer)

Phasure Mach III Audio PC with Linear PSU (manufacturer)

Orelino & Orelo MKII Speakers (designer/supplier)

Link to comment

Peter, I actually read all of that. I even followed it, though I moved painfully slowly through parts, with my lips moving as I read :). Thank you for the thoughtful, informed response.

 

First, one clarification: When I say if you can't hear it, it doesn't matter, I'm not talking about our limited ability to hear high frequencies as we age. I'm talking about what can and cannot be heard in scientifically controlled and statistically valid listening tests. I think all manufacturers of gear, including cables, should run them. I think the fact that those seem to be the only kinds of tests that they almost never use is very telling. And I think it is a bit of a scandal that the audio press, paper and virtual, is not consistently running such tests, as they are the only tests that will inform and protect their readers. I consider it an abdication of responsibility. I'm not talking about the small not-for profit enterprises like this site, of course. But for major audiophile (and A/V) publications to enthusiastically recommend technologies that ONLY perform competitively in the context of listening (vinyl, tubes...) then turn around a recommend expensive cables based largely on measurement or one reviewer's hearing without verifying through controlled listening tests is pretty outrageous in my view.

 

Second: I don't doubt for a moment that you hear what you say you hear. I would just say that if one must train oneself to hear it, it is probably not relevant to the music lover. To use one of your examples, I know what hissy, hashy cymbals sound like, and I don't hear them (not when they're recorded well, anyway) in my very modest system (Lossless files from a Mac>Trends UD-10 digital transport used as a DAC>vintage Harman Kardon integrated amp>Sennheiser HD580s). And my reference for realism is irrefutable: It is cymbals, heard at close range over 3 decades of playing live music. I know what they're supposed to sound like.

 

Can reproduction get better than my modest listening station? Yes. A lot better. But I still don't hear hissy or hashy, which kind of makes my point: Audiophiles and engineers have trained themselves to hear. I sometimes wonder if mostly they've trained themselves to hear problems and if, sometimes, they are as guilty as the rest of us of hearing what they expect to hear. I expect to hear cymbals, and perhaps my mind fills in the blanks to make cymbals out of hash. You look at an unremarkable system and expect to hear hash, and so you do. Which one of us is right, if either of us are, is not relevant. What we hear is. And if, in a scientifically conducted blind listening test, a statistical sample of music lovers or audiophiles or engineers cannot hear the difference between $2,000 worth of speaker wire and lamp cord consistently enough to mitigate the margin for error, for all practical purposes, the difference does not exist; the difference is theoretical.

 

And so I believe that listening tests (blind and statistically significant, not personal and subjective) are the only tests that, in the end, matter. The others may be useful in helping to design gear, but if their findings are not verified by listening tests you will, essentially, be designing playback equipment for your testing equipment. A very limited market.

 

MHO, YMMV and all of that.

 

Tim

 

I confess. I\'m an audiophool.

Link to comment

I would just like to add one point to this very interesting and remarkably restrained thread if I may...

 

First off, to re-iterate my standpoint on cables... I am a skeptic, that believes there ARE differences in cables, but that the differences are very small, and certainly have been exaggerated by the cable companies AND the mainstream Hifi press. I run some NVA LS3 cables and have no intention of any further upgrades.

 

However, my point is one that sort of sides with the cable lovers. And I am amazed people rarely mention it.

It regards ABX testing. Often people on my side, or complete disbelievers say "Blind test it then...". The problem here is, that often 2 components whether cables, or electronics cannot be easily blind tested. I bought a new DAC, and upon first listen the improvement was very minor. I think often it is only when listening to a new component for a while (a week say) that you really start to benefit what are often very small changes. I think I can ABX it now, but it takes a while. The improvement is subtle, but shows itself very clearly through headphones when listening for extended periods. The sound is less fatiguing, easier on the ears, and mind. Reverting back to the older system isn't an immediate degradation, but consistently is more tiring.

This type of improvement is definitely there, but is hard to quickly identify.

So ABXing cables has got to be pretty difficult, especially if you are comparing 2 sets of decent cables. Yes, I believe they will be different, but not enough so that an excerpt from a piece of music repeated for each is going to prove anything.

Another point... Higher fidelity isn't always better... I tried a new headphone amp, a £500 one, and it was immediately fantastically detailed...An ABX on it would have been easy for a partially deaf individual... the sound was stunning, running from FLAC to DAC to Headphone amp...But after 20 or 30 minutes I found it VERY fatiguing. I couldn't put my finger on it. It was more detailed, better soundstage, better sound in every department, but in the end I preferred the softer, less detailed, less hifi sound of the beresford DACs headphone output! I have since bought a very cheap £80 headphone amp, which isn't close to the expensive one I tried, in quality, but gives me the extra detail, without the fatigue.

 

 

 

 

“Music and rhythm find their way into the secret places of the soul” - Plato[br]

Link to comment

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now



×
×
  • Create New...