What Does A Song That Costs $5 Sound Like? | Hacker News Reader

What Does A Song That Costs $5 Sound Like?

npr.org

26 pointsjoe_bleau48 comments

It would be nice if someone could demonstrate that DSD sounds better than PCM in a double-blind experiment. Just as a sanity check, you could record the PCM from the DSD output to discover whether PCM is perceptually transparent. Advocates of DSD are often skeptical of double-blind experiments, which itself is a red flag.

> Advocates of DSD are often skeptical of double-blind experiments, which itself is a red flag.

Exactly. There's a lot of wackos out there in the music biz though. People have religious beliefs about certain products/technologies/sounds without any solid reasoning behind it.

Whats wrong with that? We don't question tastes in art, why audio? I don't need logic or research or the agreement of others, I just know what sounds right to me, and that is perfectly solidly reasonable. Ears are as weird and personal as eyes, as is our interpretation. You wont necessarily hear what I hear. On top of that, over time they change.

No need for words like wacko and religious. Unless of course you doing mind if I fire back with a few judgmental words of my own?

I think you have a different sort of 'taste' in mind than busterarm does.

It is a very reasonable thing to prefer record over CD. They have different sounds, people have different likes, easy, done.

But if someone tells you they prefer the look of HDMI when you use gold-plated wires, they're clearly deceiving themselves, because as a digital signal there is no difference whatsoever.

So the important question to ask is whether people can hear the difference. I will accept that some people have better/different ears from others, but the best way to demonstrate that ability is a double-blind test.

Your standard is what 'sounds right'. Well, if you can reliably tell me that X technology sounds better than Y technology, without labels on the two sets of headphones in front of you, then great, that's all I wanted. I don't need any kind of research showing that a format is objectively 'better'. I just want someone to show that it's objectively different to human ears.

If someone says they have a preference, but doesn't actually prefer either one in a blinded test, then the labels 'wacko' and 'religious' fit.

Right. I'm trying to refer to the self-deception that people have when it comes to audio gear. A really common example is how many musicians and audiophiles end up buying power filters and claim they make the sound better. Often these people don't need the power filters and you can prove with documentation and test equipment that they a) don't need them and b) aren't doing anything to the sound. Are you playing out live in random clubs? Sure get a power filter (but that's not as important as a surge protector). Are you playing at home with modern wiring and not overloading your circuit? Don't waste your fucking money.

So many musicians end up developing biases about what certain equipment is supposed to sound like. There are a few videos floating around comparing guitar amp/cab modeling gear to the real thing. A lot friends of mine have trouble picking them apart or getting them right in a blind listening test. Frequently the model has some tweaks that make it sound better... or a better description would be to say that it sounds more like what the musician expects that gear to sound like. Some amp X was used on such and such album and has Y sound and the model sounds like that amp does on the album.

I'm frequently right in these blind tests and I end up telling people is that it's because I don't care about the gear and don't have any preconceived notions about what it should sound like. I just listen to the timbre and make my best guess.

Because Monster Cables use those emotions to rip people off. Unless you really and truly think there's no moral problem with selling cables at a multi-thousand-percent markup based on claims which are trivially disproven, in which case I doubt we'll be able to have a reasonable conversation.

What's wrong wrong with it is simple: The question "Is PCM transparent?" is not subject to taste. It either is or it isn't. If listeners can't tell the difference between a source that is being reproduced from PCM or one that is being reproduced from DSD, then we can objectively state that PCM does not degrade sound quality.

You and I can disagree about whether impressionism is a "good" style of painting. The art world spent a good deal of energy debating that point. That is something which is subject to taste. But if you say "yes" and I say "no" to the question of whether van Gogh created his impressionist painting on canvas, then only one of us can be right. In this case, my insistence that van Gogh didn't paint on canvas would put me squarely in the loony or wacko category.

High end audio delivers an experience which cannot be easily quantified for the listener. It's just like wine, most people (even many wine experts) can't tell wines apart, let alone rank them in double blind experiments, but drinking wine, from the presentation to the bottle to the company you're keeping at the time is what brings it together.

It could be anything in that bottle at that point, but the emotions of that holistic experience are what matter.

If you are summing up the whole experience rather than the audioformat, then you really don't have anything comparable. If one format really is better (and even that is subjective), then there needs to be a repeatable experiment that proves it.

Unlike wine, music really comes down to knowledge, experience and training. Sound and our interpretation of it by the human ear have had rigorous study. Much of it is quantifiable. We know the optimum frequency response of our ears is around 85dBSPL, for example.

Things that are just peoples perceptions (like "brightness") have technical terms (amplitude of harmonics).

Sound engineers aren't doing black magic. If you reduce dynamic range too much through compression you can end up with a sound that people find "boring".

I'm sorry but there are things that very quantifiably sound terrible.

(I am a musician, I have run live sound and mix my own music.)

I totally agree, there's certainly going to be stuff that most people, audiophiles included, would consider better or worse.

What I mean though is there are certain aspects that can't be reduced to technical terms- like going to listen to a live performance from an artist who isn't going to perform any longer or listening to a piece on a rare, hard to produce system. The value isn't coming directly from the technical aspects of the production for the listener at that point.

For sure, but I wouldn't call somebody wacko based on that sort of aesthetic difference.

I reserve wacko purely for people who claim performance benefits from things which cannot be proven or measured in any way...or are provably false.

Its painful sometimes to read audiophile forums. Often you'll see people claim that FLAC audio doesn't sound as good as raw PCM. Some people seem so proud of their ignorance that they ignore the benefits instead of trying to understand.

That stuff is funny in a sad way but easy to ignore. I find it hardest to deal with the people who think that spending more money on something makes it better.

My little $100 tripath amp and the cheap but appropriate speakers that I picked for it absolutely smoke a lot of $3000 amps (and similarly expensive speakers) that I come across and people tend to not believe unless they're blindfolded (and still try to poo-poo it when they find out how little my setup cost).

Also it crushes my spirit when people have some $10k HiFi setup and the speaker placement/room accoustics aren't right at all.

I believe there's a good bit of psychological evidence that spending more money actually makes you perceive things as better.

For sure, but that's a cognitive bias.

These people go out claiming these things as fact...and unfortunately there's an entire industry out there supporting and benefiting from their claims.

Right, saying something is better because it's quality - absent price - is better than something else's quality is different than saying something is just plain old better (which it might be just because it's more expensive).

Seems to me that there is something people want - a better audio experience. They may actually get that by paying more, but the preferable outcome would be to get better objective quality. In fact, I think it would be preferable to even pay more to get the better objective product and know you weren't duped. I think there is probably a market for consulting:

value1 = objective_value1 + value_of_knowing_you_werent_duped

value2 = objective_value2 + cognitive_bias_for_paying_more

If it's true that objective_value2 is actually << objective_value1, and cognitive_bias_for_paying_more is not too significant, value_of_knowing_you_werent_duped could be a lucrative business.

I loved that fact that Monster Audio cable was designed with an 'arrow' telling you which end should be on the speaker and which at the amplifier because they were optimized for sound flowing "to" the speaker.

It doesn't even stop at audio cable. Ever seen those "High quality HDMI cables? The bits on THOSE wires are purer than any other 1s and 0s. There is real money to be made in selling the emperor his clothes.

What I find interesting is the positioning. Fry's has the $150 hdmi cable at eye level, then a $15 hdmi cable a little bit below that, and then on the bottom shelf, you can dig around and find a $1.50 hdmi cable.

I suspect that the $150 cable is there to make the $15 cable look like the "reasonable" or "middle of the road" choice.

(from experience, the $1.50 cable works just fine.)

You're lucky. Back when I still worked in an electronics store, the $15 was your only other option. Except back then it was a $30-45 cable.

Aside: Anybody else remember $50 6' USB A-B cables?

I bet best buy and office depot still have $15 as the minimum price for a cable. Last time I really needed an ethernet cable and everything but office depot was closed.... I paid like $15 for a 6' cable.

I think that's just the best buy/office depot "minimum price"

Fry's is one of those places that targets, well, me; They are more expensive than monoprice, but not by so much that I would delay a project to get the monoprice price.

I think they are pretty good at pricing high enough that they are making some money (I mean, I /expect/ to pay more in person) but low enough that I can still go in and buy a hundred cables for a project without it being thousands of dollars.

oh man, I remember back when my only keyboard failed during my shift (i work from home) and I had to rush out to Staples to get a short-term replacement. There wasn't any available option under $80.

Though I should just shut up. I use a Topre Realforce now and most would consider the cost to be excessive.

There actually are differences in digital cables (like HDMI, cat6 utp, etc.), though -- they become apparent in longer runs. I've seen really crappy 2m HDMI which wouldn't actually sync.

The price point for decent cables isn't high, it's just not the absolute cheapest cables. I'd probably exceed the spec for long runs or noisy environments, too. I mainly go with monoprice for short, and blue jeans cable for longer.

Yes, there has been testing that there are quality differences in HDMI cables over 30'. But it's not Monster's gold plated contacts that makes them better, it's alwayas cable materials and shielding.

Every objective test that I've seen though with tested-good HDMI cables shows no difference until that 30' mark. The HDMI licensing board recommends that past 30' you use signal amplifiers or HDMI-to-Ethernet anyway.

Maybe out past 10' you'd want to worry about cable insulation? I haven't been in a situation to worry about it yet. At 2m I'd think any cable that doesn't sync is defective. I'm not interested in paying 10x-100x the price for higher manufacturing quality control standards.

I'm mostly talking about $3-8 cables (CL2 2m from Monoprice) vs. $1-3 cables (chinese mystery crap). I also really prefer 24 to 28 awg.

I sold electronics in college. The only difference I could ever detect was the money that I made off the sale.

Audiophiles are all about impressing rich friends with the amount of money they wasted. It's just like the corporate VP who spends an extra $40k on a V-12 Mercedes that is used to commute 2 minutes to work in rush hour traffic.

It's not because you don't have the hear that nobody has it. That's perception bias at best.

I worked as a recording engineer you definitely can detect a difference between audio gear. However, it's likely that you weren't selling products for an audience that actually cared.

But those performance differences are measurable and are independent of price. There's data (and better yet, graphics) that can be supplied to help explain those differences but they're almost never presented to the salesman or the customer...if either is knowledgeable to interpret them anyway.

Audiophiles (more specifically audiophile communities and publications), traditionally do have a tendency to prefer gear that his higher in price-range/markup over gear that may be better but isn't.

Reviews of products that say "hey, this product that costs <$250 is measurably better than this 5-star product that costs 10x as much" are rare and make waves when they happen but such product differences always exist.

Disclosure: I'm an 'agnostic' audiophile. Meaning I optimize my music experience as much as possible (given time, space & money constraints) - FLAC -- wired-only streaming -- audiophile grade sound card -- decent & no-nonsense stereo amplifier -- floorstanders...but I draw my own line when things like power conditioners & gold/jelly cables come on.

Yet, I call myself an agnostic because I keep myself open to the possibility that some audiophiles might get true value from their esoteric equipment choices. Appreciation of any art is necessarily subjective. Double blind tests have no use there at a personal level, unless you're an academic or a critic because you'd use them to analyze or critique other's choices.

Why should I, or any reasonably intelligent person with his/her own biases and choices, make decisions by artificially isolating some of their senses (sight, smell etc.) via double-blind tests? The idea itself is preposterous and insulting. Worse, if a double-blind test tells me option A was 'better' than option 'B', and I change my preference after that, was it even my choice?

There's a world of subjective and personal choices we make everyday from wine to clothing brands to books to music to personal technology. Are we confident that we can duplicate those choices via double-blind tests? I don't think so. Sure, I might try a fee tests once in a while to 'validate' or 'challenge' my own choices, but I'm never going to allow the tests to alter my individual choices.

> The idea itself is preposterous and insulting.

I'm not asking you to make decisions according to my criteria. I can even sympathize with the multisensory approach: I like the tactile experience that CDs offer, for example, and only reluctantly switched to downloaded music in 2013 because I simply couldn't find physical copies of the music I wanted.

However, you seem to misunderstand the meaning of "double-blind". It does not mean "blind" in the literal sense. It only means that neither the experimenter nor the subject are consciously aware of which piece of equipment is under test.

The reason I like making some choices based on double-blind tests is because I know I have a ton of cognitive distortions. I want to believe that the more expensive amplifier is better, I want to believe that the open-source codec is more transparent, and I want to believe that my gear is better than my friend's gear. Sometimes (not always) I actually want to know the answers to these questions. So, science!

I'm all for embracing subjectivity, but if you are selling me a $1,000 or $10,000 audio system, I want some more objective justification that it is worth the cost.

Or, put another way, you may want all of your vinyl blessed by the pope, but I'm not going to pay extra for it.

> Worse, if a double-blind test tells me option A was 'better' than option 'B', and I change my preference after that, was it even my choice?

This is pretty esoteric stuff. I can say that my goals are closer to "I want a decent sound system without paying through the nose", not "I want some kind of existential assurance that I used my free will to make this particular decision." I change my preferences all the time, based on all sorts of internal and external data.

Or, put another way, the person that exists five minutes from now may have my name, but he's not me, and I trust him to make his own decisions.

> However, you seem to misunderstand the meaning of "double-blind". It does not mean "blind" in the literal sense. It only means that neither the experimenter nor the subject are consciously aware of which piece of equipment is under test.

I'm sorry if I wasn't being clear enough. Allow me to restate in somewhat better terms (I hope).

I maintain that our choices are uniquely and indivisibly (a) personal, (b) subjective, and (c) the product of all our senses. Take one away, and what's left isn't really a 'choice' any longer.

I agree that a double-blind test isn't about literal blinding, but it does blind one or more of your senses. Let me illustrate using two hypothetical examples:

- if I do a double-blind test b/w the iPhone 5, 5s and 5c, masking everything else but their screens, are we all so confident that we'd pick the very same model that we picked up in a store or a website? All I've done is to 'blind' you to the exterior appearance of the phone, which means (per the double-blind enthusiasts), you're then judging the phones based on only features like speed, animation smoothness, screen quality etc. etc. Worse, if you end up picking a different model than what you'd done IRL, did that mean Apple tricked you?

- say there are two different flavours of ice-cream, both of which look the same and come from the same company. But one of them is 25% more expensive, and contains an additional ingredient that makes it smell 'better' than the other to you. Now what if I design a blind test that blocks your smell and I ask you choose the better ice-cream. Would you still pick the same? I don't know. Because your choice of ice-cream was made using all your senses. Now it's entirely possible that the company is making more profits from the more expensive flavour merely by adding some relatively inexpensive flavour. Should you then change back to the other flavour because that's what your 'objective' test tells you? If you did, are you sure you'd enjoy the ice-cream just as much as you did earlier?

I'm not defending snake-oil peddling audiophile equipment makers, just saying that the value of what they (or anyone else, like, heck Bose!) sell is personal and subjective. Like most HN skeptics, I stay clear of them, but I also remain open-minded to the idea that there may be folks who find value in them and that I will never be able to truly understand why.

> Or, put another way, you may want all of your vinyl blessed by the pope, but I'm not going to pay extra for it.

Totally agree. Sure, do any number of tests and research before making your own choices. But don't expect that others will feel bound to those tests the same way. Which is what the original discussion was about, namely that "advocates of DSD are skeptical of double-blind experiments".

They have a right to. And we cannot "prove" them wrong.

> it does blind one or more of your senses

Wow, you really don't know what a double-blind study is. It is NOT about the senses.

From Wikipedia (http://en.wikipedia.org/wiki/Blind_experiment)

> The terms blind (adjective) or to blind (transitive verb) when used in this sense are figurative extensions of the literal idea of blindfolding someone.

Note the word "figurative". Senses are not actually being blinded.

You're entirely missing the point. The only difference between DSD and CD is the waveform. There are no other factors in the decision of sound quality. Someone claiming that DSD is better does so not because of artist choice, or headphone fit, or lacquered-wood smell. They are claiming that the waveform is more pleasing in some manner. So to test this you must isolate the waveform in an otherwise-ideal environment.

Nobody objects to a person preferring a specific device because of other senses. But it's provable whether DSD itself is playing a role in the experience.

> Worse, if a double-blind test tells me option A was 'better' than option 'B', and I change my preference after that, was it even my choice?

Someone informs you, and documents impeccably, that your girlfriend / wife is cheating on you. As a result, your opinion of her is lowered. Is that your choice? Suppose that the informant's specific goal is to get you to think worse of your girlfriend, for reasons of spite and malice. Does that matter?

Apples and oranges. We're talking double-blind tests between consumer products. Let's not turn it into a wider philosophical debate on human beings and relationships :)

So DSD is just storing [the equivalent of] the digital pulse part of a class-D amplifier? That seems completely pointless...

My feeling is there's a one to one transform between the 1 bit world and the multi-bit world, which means that they're mathematically the same and any perceived differences are merely the result of poor A/B testing.

Pretty much spot on here.

It sounds a lot like someone making money. You play these over your $150,000 speakers with $10,000 speaker wire and you can pretty much hear the money hitting the table.

I heard this story on my way to work this morning and was just shaking my head. There are probably going to be a lot of people wanting to try out this newfangled $50 album, though.

Whether DSD is even good or not is still something hotly debated. It's not at all suitable for editing. I'm really scratching my head why NPR reported a story that is so one-sided.

The fact-checking here is poor. They're presenting it as DSD vs MP3 when really it's DSD vs PCM.

But no one's talking about FLAC/DSD for archiving and MP3 for casual listening... which should be the point.

The history of music (even if you don't like today's pop) should not be stored in 320 kilobit/s CBR.

The very fact that 5.6MHz sampling rates are even a thing intended for human listening makes me doubt this lady's sanity.

It is intended for encoding, not for hearing.

Here: http://en.wikipedia.org/wiki/Pulse-density_modulation

You're ignoring that the encoding is only a single bit deep. Compared to a CD, it's only four times the bitrate. It just happens to be drastically reformatted.

Here is a link to what must apparently be the new range of Sony hi-fi equipment the article talks about (but doesn't link to) in the first paragraph: http://www.whathifi.com/news/sony-launches-high-resolution-a...

Are people even able to audibly tell a difference? Like in an ABX test?

Audiophile (the unhealthy obsession with audio) has made it on NPR

Reply on news.ycombinator.com