Something alluded to but not said explicitly is that specs don’t tell you if an audio product is “good.” Very few people can look at a spec sheet for any product and tell you if it’s “good” or “better” than something else. You need to listen to it.
Impedance/sensitivity just tell you how well it will play with your system. Stuff like “high impedance == more clarity” is snake oil.
I say this as a former pro audio specialist who has read and possibly written a bit of snake oil in their time.
Headphone specs are not straight up horseshit like the rest of their product page, but they are next to useless. Take your daily driver and a reference track you know and listen to the cans before you buy them.
There is a huge amount of unspecified details. I have recently changed pads on my Beyerdynamic DT990s from cloth to leather. I had to quickly revert back. The sound became very boomy, practically unbearable. I suspect the tight seal provided by leather to be responsible.
Unless you do a blind listening test, most consumers will fool themselves into believing that the more expensive or attractive looking headphones sound "better".
Company idea (or maybe Consumer Reports is good enough):
Don't hire experts to review products. Do a double blind study with like 100 participants. I pay you $5 to access this years study of 80 different popular/common headphones. But of course do this for all kinds of products that need this kind of review.
I would happily pay for a sample of how the population feels about products and how they on average rank them.
Let me x/y plot things like rating vs. Price/warranty/whatever
Don't ruin your pages with advertising.
Be annoyingly clear and repetitive that you only make money from the fee and everything else is done in a way that reduces bias wherever possible.
And don't screw up the UX (looking at you, graphics cards and CPU sites stuck in 2003)
There are market research firms that do things like this, and their studies cost at least 100 times more than that. It's actually pretty lucrative to keep up to date on details like that if you consult for investment bankers on a particular product category/market segment but that can be difficult to break into.
There's nothing wrong with people buying things that they like more than other things for whatever reason. Audio is especially subjective compared to other products, possibly one of the most subjective product categories in consumer electronics.
Blind listening tests also aren't a silver bullet for headphones. Fidelity is just one factor in deciding whether the product is right for you, other things like headband quality, general fit, comfort, weight, etc all play into whether or not you'll like it. But the overall point I'm making is that none of this is obvious from looking at a spec sheet, you need to try it yourself, IRL.
The first thing you need to do is get one that matches your audio output.
In "theory", you get higher impedance as a result of thinner/lighter wires in your voice coil, which "could" result in better sound reproduction.
In practice, I don't think that's really relevant for most usecases (in any case, you would have to measure the performance of the specific hardware anyways). So the most important thing is just get something that your output driver can handle.
If you don't have an extra amp, you'll probably not want the 600. The safe bet is the 32, anything can drive that, including your phone.
Traditional wisdom is that higher-impedance models have better sound, but at least I wouldn't be able to hear that, and there is a certain bias in that comparison, because higher impedance models are bought by audiophiles, so companies cannot cut too many corners there.
In addition to understanding impedance and sensitivity, doing actual tests of the headphones to see how they match up with your hearing and tastes is important (and getting harder to do with high end headphones). I personally use the following:
The audio characteristics of headphones can also be objectively measured using test gear. rtings.com has a very nice database with comprehensive measurements for a large number of headphones.
True - there an objective method of measuring headphones. That said, what headphones sound good to me, with my "slight" hearing loss and a taste for both classical and dubstep, will be different than what sounds good to you.
I found rtings' database was good for regular consumer stuff like gaming headsets and noise cancelling stuff for planes but it was quite limited when it comes to "audiophile" brands.
The UI is a bit of a mess though, you need to select a "User" or "Pro" report on the device's page (which otherwise doesn't tell you anytihng). Here are some example reports:
I honestly have no clue what I'm looking at here though.
While I'm at it, the AudioScienceReview forums (http://audiosciencereview.com/) have nice comprehensive measurements of the other things you might find in an audio setup (e.g. amps, players, DACs etc., even USB-C dongles).
Is there something equivalent for computer parts as in "computer parts specs explained"? Like it should explain how to choose a processor/RAM/(I/O)/battery and other components which make up a computer.
I mean, basically just read reviews and look at benchmarks on sites like anandtech. there's not a lot of mystery in computer hardware specs. find a benchmark that approximates your workload and buy the top scoring part that you can afford. in theory, you could delve into the architectural details of different CPUs but it will take a lot of knowledge for this to be a better use of your time than just looking at benchmarks.
I think you would have a similar problem as audio gear. You can know all of the specs but there is some hidden info in the processor/GPU architecture that is only determined by performance testing. Same as audio gear is mostly affected by implementation in the design than the raw specs of drivers
I really like some of the non-sound related measurements they perform -- the clamping force and breathability are pretty important for comfort. For instance, as someone with glasses there's a good chance those HP50s would not be remotely comfortable as the high clamping force on the glasses arms tends to get pretty irritating after a while.
Their more thorough sound tests are really invaluable as well -- for instance, apparently the bass response goes down considerably with the HP50s when the cups can't form a good seal (e.g., if you're wearing glasses).
This article is really inaccurate. The impedance isn't a good signal for how much power needs to be pumped into the headphone. The impedance also doesn't act as a good signal for what devices could generate enough volume. It also isn't a good signal for being able to "more accurately and vividly reproduce sound." Almost everything in this section isn't accurate.
Lower impedance headphones are "harder" to drive at the same voltage because they require more current. Often, amplifiers will have lower maximum outputs when driving into a lower impedance.
The sensitivity section isn't too bad but a lot of this article is pure snake oil and bad science. A lot of the amplifier talk in this article also doesn't hold up to scrutiny.
>>> Impedance is electrical resistance – essentially, how well something resists an electrical current that passes through it. It’s a “measure of the opposition that a circuit presents to a current when a voltage is applied.” (Thanks, Wikipedia). Even if that makes no sense, headphone impedance is still very easy to interpret. It’s measured in ohms (Ω) and what it tells you is how much power your headphones will need to get to a reasonable listening volume.
For someone who understands what voltage, current, impedance, and power are, this paragraph is going to look like someone either doesn't understand the subject matter at all, or that something has been lost in an attempt to simplify it for laypeople.
I would have started with the relationship between voltage and current in a simple circuit. For instance, the more voltage is applied, the more current typically flows. Resistance and impedance are terms for the ratio between voltage and current in some kinds of simple circuits.
And so forth. I'm using the fiction of a "simple" circuit to avoid going down a rabbit hole of detail that would be unnecessary here.
The ratio of voltage and current in the complex plane.
Drop the “harder” and “easier” analogies with Ohmic resistors. They’re confusing, with impedance you’ve gone up a level and the analogy breaks down. Even “matching” Z is kinda meaningless if you don’t express it as a complex quantity: what you’re trying to match is not |Z| but Z* (otherwise it is possible to match |Z| And deliver no power! In practice matching |Z| is good enough)
What if you don’t know about phasors and Fourier transforms?
Doesn’t matter, don’t look down on people to over simplify. Use the magnitude and angle formulation of complex numbers and build on that with pretty pictures.
26 comments
[ 2.4 ms ] story [ 54.3 ms ] threadImpedance/sensitivity just tell you how well it will play with your system. Stuff like “high impedance == more clarity” is snake oil.
I say this as a former pro audio specialist who has read and possibly written a bit of snake oil in their time.
Headphone specs are not straight up horseshit like the rest of their product page, but they are next to useless. Take your daily driver and a reference track you know and listen to the cans before you buy them.
Don't hire experts to review products. Do a double blind study with like 100 participants. I pay you $5 to access this years study of 80 different popular/common headphones. But of course do this for all kinds of products that need this kind of review.
I would happily pay for a sample of how the population feels about products and how they on average rank them.
Let me x/y plot things like rating vs. Price/warranty/whatever
Don't ruin your pages with advertising.
Be annoyingly clear and repetitive that you only make money from the fee and everything else is done in a way that reduces bias wherever possible.
And don't screw up the UX (looking at you, graphics cards and CPU sites stuck in 2003)
Blind listening tests also aren't a silver bullet for headphones. Fidelity is just one factor in deciding whether the product is right for you, other things like headband quality, general fit, comfort, weight, etc all play into whether or not you'll like it. But the overall point I'm making is that none of this is obvious from looking at a spec sheet, you need to try it yourself, IRL.
The opposite really. The higher the impedance the less power they will consume for a particular input voltage.
That's why lower-impedance headphones are used for mobile phones, and only "real" audio gear is used with higher impedance.
For example, the Beyerdynamic DT-880 is sold in three versions: 32 Ohm, 250 Ohm and 600 Ohm. The 600 Ohm on an iPhone won't make you happy.
But which one is better? Should I go for a lower impedance or higher impedance?
In "theory", you get higher impedance as a result of thinner/lighter wires in your voice coil, which "could" result in better sound reproduction.
In practice, I don't think that's really relevant for most usecases (in any case, you would have to measure the performance of the specific hardware anyways). So the most important thing is just get something that your output driver can handle.
Traditional wisdom is that higher-impedance models have better sound, but at least I wouldn't be able to hear that, and there is a certain bias in that comparison, because higher impedance models are bought by audiophiles, so companies cannot cut too many corners there.
https://www.audiocheck.net/soundtests_headphones.php
This website had much better coverage of those brands: https://reference-audio-analyzer.pro/en/catalog-reports.php?...
The UI is a bit of a mess though, you need to select a "User" or "Pro" report on the device's page (which otherwise doesn't tell you anytihng). Here are some example reports:
User: https://reference-audio-analyzer.pro/en/user-report.php?id=7...
Pro: https://reference-audio-analyzer.pro/en/report/hp/meze-99-cl...
I honestly have no clue what I'm looking at here though.
While I'm at it, the AudioScienceReview forums (http://audiosciencereview.com/) have nice comprehensive measurements of the other things you might find in an audio setup (e.g. amps, players, DACs etc., even USB-C dongles).
https://www.innerfidelity.com/headphone-measurements
I like my NAD VISO HP50.
I really like some of the non-sound related measurements they perform -- the clamping force and breathability are pretty important for comfort. For instance, as someone with glasses there's a good chance those HP50s would not be remotely comfortable as the high clamping force on the glasses arms tends to get pretty irritating after a while.
Their more thorough sound tests are really invaluable as well -- for instance, apparently the bass response goes down considerably with the HP50s when the cups can't form a good seal (e.g., if you're wearing glasses).
Lower impedance headphones are "harder" to drive at the same voltage because they require more current. Often, amplifiers will have lower maximum outputs when driving into a lower impedance.
The sensitivity section isn't too bad but a lot of this article is pure snake oil and bad science. A lot of the amplifier talk in this article also doesn't hold up to scrutiny.
For someone who understands what voltage, current, impedance, and power are, this paragraph is going to look like someone either doesn't understand the subject matter at all, or that something has been lost in an attempt to simplify it for laypeople.
And so forth. I'm using the fiction of a "simple" circuit to avoid going down a rabbit hole of detail that would be unnecessary here.
Drop the “harder” and “easier” analogies with Ohmic resistors. They’re confusing, with impedance you’ve gone up a level and the analogy breaks down. Even “matching” Z is kinda meaningless if you don’t express it as a complex quantity: what you’re trying to match is not |Z| but Z* (otherwise it is possible to match |Z| And deliver no power! In practice matching |Z| is good enough)
What if you don’t know about phasors and Fourier transforms?
Doesn’t matter, don’t look down on people to over simplify. Use the magnitude and angle formulation of complex numbers and build on that with pretty pictures.