This page is intended to be a guide for those seeking to expand their knowledge with regards to headphones. For general information on headphones, please visit Wikipedia:Headphones. Guides are located at the bottom section of the page.
While discussion of headphones is welcomed, generally threads based on which headphones you should buy are discouraged. Those threads frequently result in the same guides being posted, many of which can be found on this page.
There is no such thing as the objectively best headphone. While we can talk about how a headphone is objectively more accurate than all the others, it does not make it better for everyone, since some people do not like the sound of accurate headphones. They might, for example, prefer a bassy sound with scooped treble. However, it should be noted that as a generalization listeners tend to prefer accurate sound reproduction and therefore a neutral signature under ideal conditions 1 2 3 4 5 6.
When seeking a headphone that reproduces sound in a way that fits your subjective preferences, the term "sound signature" will be of great use to you. Sound signature refers to the way a headphone reproduces sound, for example, a bassy headphone will reproduce sound with more bass than other frequencies, and a bright headphone will emphasize the high frequencies. Despite the differences in subjective preferences, there are still objective differences like build quality, weight and distortion, which will also be useful to you when looking for a headphone. Note: Comfort, is in fact subjective. The comfort a user experiences will be related to head size, hair length and other factors. While some headphones are terrible for everyone, others may only be uncomfortable to a few.
The world of audio is filled with snake oil products, such as magical stones, hangers and silver cables. If something sounds too good to be true, do some research on them.
- 1 Amplifiers and DACs
- 2 Impedance/resistance and sensitivity
- 3 Frequency response charts
- 4 Soundstage
- 5 Form factor
- 6 Some things to look out for when buying a headphone
- 7 Types of drivers
- 8 Accessories and mods
- 9 Equalization
- 10 Digital
- 11 Surround sound
- 12 External links
Amplifiers and DACs
Amplifiers and DACs are a necessary part of your digital audio system. The DAC converts your music from 0s and 1s into analog signals, and then the signal is amplified by your amp. People often ask if they need an external amp or DAC, and the answer will depend on the following:
- Are your current headphones underpowered?
- Does your source produce audible noise/hissing/EMI?
There are a few types of amp and DACs, and they serve different purposes. An amp or a DAC can be portable, meaning you can simply put them in your pocket, and then enjoy the music on the go. Desktop amps or DACs can go from the tiny USB powered FiiO E10 and all the way up to the size of a Beta22. Amps and DACs are either be two stand alone units, or a single device. For example the FiiO E17 is a portable amp/DAC two in one, and the WooAudio WES is a vacuum tube based desktop amplifier for STAX electrostatic headphones.
With amplifiers, most people want a neutral and transparent sound signature, meaning the amp doesn't change the sound in anyway. However some people prefer amps that change the sound, because of this, many vacuum tube based amps are made with the intention of reproducing a colored sound.
Professional amps are often multi-channel, meaning it can power many headphones at once. Headphones with high impedance is recommended with this kind of amps, for more on impedance scroll down.
To utilise your amp correctly and minimising damage and stress, be sure to read up on gain structure. To simplify correct gain structure for playback, keep the volume bar at 100% or as close to it on your computer, then the same on your DAC and only adjust the volume at the end of the chain (AKA your amp).
Impedance/resistance and sensitivity
Headphones will often come with a few specifications like frequency range, sensitivity and impedance/resistance. The most common impedance would be 32ohms. Although some headphones will have impedance all the way up to 600 ohms. As a generalisation, more impedance will require more voltage and less current, and visa versa. The reason that many portable players can not drive a 600ohm headphone well is due to the limited voltage it is capable of outputting. However for the purpose studio monitoring, many 600ohm headphones in one amp is much better than many 32ohm headphones in one amp as it will require less current.
Improved damping factor is also often associated with high impedance headphones but it's effects is minimal if not unnoticeable when using an amp with near zero output impedance. However when amps with very high output impedance is involved high impedance headphones will fair better than their lower impedance counter parts.
Sensitivity is often measured in SPL/mW @ 1kHz. What this essentially means is that at 1mW, a 1kHz tone played through this headphone will be in *Insert SPL here*. As the output voltage (but not power) of a headphone amplifier is essentially constant for most common headphones, dB/mW is often more useful if converted into dB/V using Ohm's Law:
Once the sensitivity per volt is known, the maximum volume for a pair of headphones can be easily calculated from the maximum amplifier output voltage. For example, for a headphone with a sensitivity of 100 dB (SPL)/V, an amplifier with an output of 1 V RMS will produce a maximum volume of 100 dB.
Frequency response charts
Before we can get into frequency response charts, you have to know what frequencies contain what sounds. If you think 20 kHz is where the bass is, you are going to have a bad time. This frequency chart should help you improve your understanding.
A frequency response graph shows how a headphone will reproduce sound, whether it'll have emphasized bass or treble, or relatively neutral, but it does not necessarily tell you the whole story. There are other factors such as non-linear distortion involved.
Measurements are often displayed with a compensation where the measured frequency response adjusted on a specific target and then normalized to it. Raw responses are available in some places and show how the headphone measures without any compensation and is useful for comparing different sets of measurements to each other from different sources when you don't know what compensation is used. It also allows you to apply your compensation curve of choice to the measurements.
Compensation curves vary a fair bit and understanding what compensation is used and when is important. The idea of compensation is to remove the response of the dummy ear to give you a better estimate of how the headphones' response is actually perceived by your ears. The common diffuse field (DF) compensation is based on the average response of sound from all directions at once. Free field (FF) compensation is from facing a speaker in anechoic conditions, where there are no reflections from the room. DF is considered the more accurate of the two for representing real conditions, but has been criticized for sounding bright and unnatural 1 2 3.
Golden Ears base their curve on the X-curve, which was originally developed for usage in theaters for tuning their sound systems. It was developed using pink noise in the average cinema because of the reverberation effects in a large room. GE found that 'flat' speakers were modeled similarly. Thus, they modeled their curve off the small room X-curve to try and get it as close to what would be expected in the studio.
The preliminary Olive-Welti (Harman) curves were created based on measurements from a calibrated loudspeaker arrangement in a listening room. They compared it in two sets of blind tests against diffuse-field and free-field curves, in-room target curves, and the unequalized headphone. They found that the equalization with their in-room curve was subjectively better than all of the other equalization curves (as well as the unequalized headphones). To refine the target curve, 249 listeners from 4 countries adjusted the bass and treble level of a headphone according to preference after it was equalized to match the flat in-response of the loudspeaker. The preferred headphone target response closely matches the preferred in-room response of an accurate loudspeaker in a reference listening room having about a 10 dB downward slope from 20 Hz-20 kHz. Research is still in progress.
Raw response of Philips Fidelio X2 + GE curve
Some sources of FR graphs:
- Korean GoldenEars / GoldenEars
- Speakerphone Headphones Measurement Blog
- Rtings Headphones Reviews
- Rin Choi's Headphone Analysis Blog
- ヘッドホン・イヤホン 周波数特性 測定結果
- Reference Audio Analyzer
To begin, we must define soundstage. Two popular definitions from Head-Fi and Stereophile are included here:
- "Soundstage - The area between two speakers that appears to the listener to be occupied by sonic images. Like a real stage, a soundstage should have width, depth, and height."
- "Soundstaging, soundstage presentation The accuracy with which a reproducing system conveys audible information about the size, shape, and acoustical characteristics of the original recording space and the placement of the performers within it."
Humans can perceive the direction of a sound, that much is known. This ability is based on the delay for sound to reach each ear, the amount of high frequencies and the volume of sound in each ear. All of these phenomenons can be demonstrated in a DAW (digital audio workstation).
Now that it is established that humans can perceive soundstage, we can move on to perception of soundstage in audio equipment. 2.0 speakers, for example, when placed together, project sound almost directly into both ears, at the same angle. When placed apart, several factors come in effect.
- The treble will have a slight reductions as the distance between the listener's ears and the speakers increase.
- The sound will reach the listener's ears at a different angle, which will be perceived by the listener
- The delay between the sound from reaching the closer ear and further ear increases as it takes more time to reach the other ear, again heard and perceived by the listener. (Think of the distances between the two ears and one speaker as an almost isosceles triangle at the start, and as it moves further to the sides the closer ear side is decreased as the further ear side is increased)
Moving on to headphones, headphones are similar to the speakers, but there are a few differences. Sound on one side will not be heard by the other ear due the nature of headphones, so the delay between the ears are only caused by the recording itself here and soundstage relies on the distance between the ears and the drivers. Where the drivers are pointing at is also different to speakers, as speakers are generally at the front of the listener, while a headphone driver can pointed either centred on the ear canal or slightly off of it. This difference in pointing the drivers will also have an effect on the sound, as shown in Tyll's measurements at Innerfidelity, where even moving a headphone a few mm, an acceptable difference similar to human error, produced differences in sound.
Headphones come in many forms, and some may not suit your needs, so always be careful when making a purchase with your hard earned money.
Headphones have different form factors, such as on-ear (supra-aural), around ear (circumaural), in-ear (IEM) and earbuds.
- An on-ear headphone rests on the ear, and does not surround it. Examples include Beats Solos and Audeze SINE. Some on-ear headphones, such as the Koss KSC75, do not have headbands and secures itself with a clip, these are known as clip-ons.
- Around ear headphones are far more popular in the high end headphone market, examples include the AKG K612 and Audio-Technica M40x.
- In-ears or IEMs (In ear monitor) are the type of earphones that is inserted into your ear canal. How far in an IEM goes varies, but they tend to provide decent isolation against outside noise. Examples include Etymotic ER4, Shure SE215.
- Earbuds are earphone that do not go inside your ears, like the junk you see included with older iPods. They are almost always terrible sounding and offer no isolation.
A headphone can also be open or closed. An open headphone usually have a mesh at the back of the earcups, which allows the sound from the headphones to travel outside, but also allows external noise to enter. For these reasons, open headphones should never be used in public. Open headphones are said to have a larger perceived soundstage and are highly recommended if you don't need isolation. Closed back headphones are the most common type. They do not have openings, offer isolation and leak less sound than open headphones.
Headbands also come in a variety of forms:
- Split - see Beyerdynamic DT 1350 and Sennheiser HD25-1 II, splits open and some find them more secure
- Single - the normal headband
- Behind the neck - goes around the back of the head
Some things to look out for when buying a headphone
Headphones are usually wired, and wireless headphones are normally thought of as inferior, due to the need to add extra components into the headphone. Sometimes companies make wireless versions of the headphones previously available in wired versions, such as the Sony MDR-R1BT. The cables for wired headphones come in detachable, and attached. Detachable cables offer more convenience, allowing different lengths of cables and cables with mics to be used.
Sometimes headphones will be rebranded by a several small headphone companies, and the price point of it will varies greatly. A great deal can often be found. The Yoga CD-880 is an example of rebrands being sold at various prices. ($50 at Jaycar, $150 from Brainwavz)
A feature often seen in luxury and fashion headphones is active noise cancellation (ANC). While most closed headphones offer decent amounts of isolation, ANC headphones aims to reduce the noise form outside even further by producing inverted soundwaves to cancel out the unwanted noise. ANC headphones use a microphone to pick up ambient noise, and then add the inverted soundwaves into the reproduced sounds. Because of physical limitations, the noise canceling will not be 100% accurate, and often there will be high frequency noise added to the sound. The main disadvantage of ANC however, is its need for batteries to operate.
Headphones are often marketed by their uses. For example, if you see a headphone marketed as a "DJ" headphone, you can expect it to be heavy on the bass, and occasionally have a roll off in the bass region similar to speakers (The Pioneer HDJ-2000 for example). A studio headphone on the other hand, will have relatively neutral frequency response, and they sometimes have the same kind of speaker-like roll off found in DJ headphones (The Sony V6 for example). Audiophile headphones tend not to have a general sound signature. Anything with a rapper's name on it should be treated like Xtreme Gaemer l33t Sniper Military Mountain Dew.
Types of drivers
There are three types of drivers that you're likely to see when shopping for a headphone. While there are others, they tend to be less common, so they're are not included in this wiki.
The three main types are:
- Dynamic/moving coil headphones are the most common type of drivers, found in $1 earbuds to $1000+ headphones. To put it simply, a dynamic driver is a piece of voice coil is attached to a diaphragm, with a magnet that pushes the voice coil to produce movement, which becomes sound. Of course not all dynamic drivers are made equal, and companies tend to add their own fancy technology to their drivers.
- Planar-magnetic/orthodynamic headphones are less common, but tend to be compatible with other consumer sources, and does not require the consumer to purchase a special amplifier. The drivers are generally thin plastic, with the voice coil spread onto the diaphragm in a manner that allows the driver to be driven evenly across its surface. Most planar-magnetic headphones have two magnets, however some only have one
- Electrostatic drivers are drivers that uses a diaphragm coated with conductive materials, and the diaphragm is moved by stators on either side of the drivers. Electrostatic headphones tend to require their own special amps. STAX and Koss currently makes electrostatic headphones. Sennheiser once made two electrostatic headphones, known as the HE90 Orpheus and HE60 baby Orpheus. The HE90 is an almost legendary headphone among hobbyists due to its price and rarity.
Accessories and mods
Aftermarket earpads and tips is frequently purchased to improve comfort or for other reasons. Foam tips for IEMs for example, improves the isolation provided by IEMs. Common earpad materials include velour, pleather and leather. IEMs tips are available in single-flange, bi-flange, tri-flange, foam and many other forms.
Cables are also modified and replaced. Some believe that they improve sound but no evidence supports this belief. V-moda, a headphone company, sells the cable intended for their line of headphones. DIY solutions are also popular.
Comply Foam tips are a common tip. They seal well, and are very comfy. The downside is they only last a month or so of normal use.
An Equalizer simply changes how strong a headphone responds to certain frequencies.
Analog EQs tend to have multiple bands representing each region of the frequency range, going from 3 to 32 bands in most situations. Digital EQs tends to emulate the same setup. EQs are available in graphic and parametric versions. A graphic EQ is a simple multi-band equalizer that controls the level of certain frequencies, where as parametric EQs can also control the central frequency of each band, and its range. Meaning it can control the frequency that it'll boost the most and the range of surrounding frequencies. For this reason, parametric equalizers are recommended.
The go-to software for equalizing your headphones is Equalizer APO along with the interface Peace GUI.
How to equalize your headphones using Equalizer APO + Peace GUI
Bands: Each band contains a number of elements that you can make changes to in order to alter the frequency response. You can add bands by clicking the plus sign or remove bands by clicking the minus sign.
- Frequencies: A band has a frequency that your boosts and cuts in gain are based around. They are evenly spread by default, and it's best to leave it that way until you know what you are doing.
- Gain Values: Here you choose how much you want to boost/cut a frequency. You can change the value by moving the slider or typing in a new one. By default, it's set to lock at 0.5 dB increments, which can be lowered for more flexibility in Settings > Snap dB gain.
- Quality: The quality determines how much an amplitude change to a frequency affects the gain in frequencies around it. A lower Q-value is going to affect a broader range of frequencies and vice versa, meaning a low Q-value will make a boosts and cuts wider and a high Q-value make them narrower. For this reason high Q is used when you want to EQ the treble, while it can be left by default (1.41) for the rest of the frequencies.
- Filter: Leave this as default (Peakfilter). Do not change it.
Pre Amplifying: To avoid clipping, never use boost, either lower the frequencies you don't want, or lower the preamp the absolute value of your highest boost.
Graph window: You actually have no idea what the changes you make to the bands have on your frequency response until you open the graph window and have a look. To do this, click on icon with a vertical black line, a green curve and a black horizontal line going through the curve. The graph window can be left open while making adjustments.
Saving: You need to save your EQ as a file or you're going to waste a lot of time on nothing. Also be careful not to click on any of the other presets while EQing your headphones, as this will lead to your changes to being lost.
On/off: In the top right corner, there's a on/off-switch. Clicking that helps you compare what your headphones sound like with and without EQ.
The reason you want to equalize your headphones is most likely because they have one or more flaws that you want fixed. There are generally two ways to go about fixing them: Dick around until you've come up with something that sounds good to you, or try to EQ your headphones to a specific compensation curve. The later requires that you have a basic understanding frequency responses charts and how they are produced.
Regardless of your method, you're not going to get a good result when trying to fix the treble, for several reasons that won't be discussed here. So what you want to do, is download another piece of software called SineGen. It allows you to listen to single frequencies at a constant sound pressure level. As you slowly move up the frequencies and get to the treble, you'll start to hear sudden peaks and dips. SineGen helps you single out those inconsistencies, which can then be removed with your EQ. Remember always to use high Q-values when equalizing the treble.
Equalizing comes with a price. When you lower your preamp, you reduce the amount of headroom you have at your disposal. This can be a problem if your source does not have the required power to drive your headphones to satisfactory listening levels after you have EQed them.
Audible distortion is another problem that can be caused by equalization, specifically when boosting frequencies. Conversely, cutting certain frequencies can lower the distortion. You can check out InnerFidelity's headphone measurements on THD+N, however, a single THD number is inadequate to specify audibility, and must be interpreted with care. Lower THD is preferred over higher THD.
Needless to say, your EQ settings are limited to the device you're using. An equalizer does not permanently change the frequency response of the headphones or any of the physical properties for that matter.
Sound files come in different digital formats. .mp3 is by far one of the most common type. .mp3 is a lossy form of compression and 320kbps is recommended for listening purposes as lower bit rates can sometimes have audible degradation. .ogg and .aac are also popular lossy compression formats, and follow roughly the same rules as .mp3. Two forms of popular formats for storage is .wav and .flac. A .wav is often the default export format used in DAWs, and can be uncompressed. .flac is a popular lossless compression format.
The difference between difference bit rate, sample rate, bit depth and other factors has been debated endlessly, at the end of the day, the only thing that matters is what the listener himself can hear, rather than what magical golden ears on the Internet say they could hear. A testing method known as ABX is commonly used to test if there is an audible difference between two or more files. An ABX test in this context, is where two samples A and B, are played to the tester, and then a third sample, either A or B, known as the X sample is played. In the case of A and B sample, the user is told which sample is which, however with the X sample, the user would have to guess which sample is the X sample. Note that this methods requires repetition, often up to 10 to 20 times for the sake of accuracy. ABX tests can easily be performed as double-blind trials, eliminating any possible unconscious influence from the researcher or the test supervisor. Because samples A and B are provided just prior to sample X, the difference does not have to be discerned from assumption based on long-term memory or past experience. Thus, the ABX test answers whether or not, under ideal circumstances, a perceptual difference can be found.
A rather interesting form of recording known as "Binaural recording" are available from records such as " Chesky, these are made with special binaural dummyheads that simulate the human ear, giving ultra realistic imaging. An example would be the virtual barbershop from QSound Labs.
Surround sound sources, usually from movies and games, can be downsampled to stereo with filters that simulate the human ears, just like dummyhead recordings. One example of a technology accomplishing this is Dolby Headphone.
A DAC, as it is used to convert the digital to analog, is not required for anything that is analog, like vinyl records. Your computer however, will require a DAC.
If you frequently play video games or watch movies, you may want virtual surround for your headphones. Virtual surround converts the multi-channel signal from movies or games into 3D stereo, simulating the experience from a real surround sound setup. This should not be confused with physical surround headphones, which carry multiple speakers in each cup and do not enhance the experience, so they should be avoided. To get virtual surround, a DAC that can process virtual surround must be used. The most widespread technology that achieves this is named Dolby Headphone, which is supported by numerous DACs, including ASUS Xonar sound cards. 3D sound works by applying properties to the audio that real sound has in a spatial environment. This is formally known as head-related transfer function (HRTF).
For gaming on consoles, you'll need a device that converts your Optical Out surround sound audio into Dolby Headphone, such as a Mixamp. Recommended you read this thread for more information.
- A list of videos on headphones
- Stack of AES papers, documents and books on headphones/loudspeakers
- Measurement Repeatability and Reproducibility @InnerFidelity
- Headphone measurements explained @InnerFidelity
- Amplifier measurements explained @InnerFidelity
- WHO | Make Listening Safe
- NwAvGuy's blog (inactive since May 2012)
- Sean Olive's blog
- Harman How to Listen
/g/uides and infographs
- Joker’s mutli-IEM review thread. More than 200+ IEM reviews
- Tranquilizer's Universal IEM Guide (based on IEM_GUY's, but with some modifications)
- $200 and over headphones
- $200 and under headphones
- Microphone /g/uide