Updated , Sam Vafaei

Our Soundstage Score and Tests
Heaphones

What it is: Soundstage qualities are not inherent to the audio content, the headphones have to 'create' them rather than 'reproduce' them. They determine whether the sound is perceived to be coming from inside or in front of the head, how open and spacious the soundstage is, how much the headphones acoustically interact with the environment, and how strong the phantom center is.
When it matters: When an accurately produced, large and spacious soundstage, similar to that of a stereo loudspeaker setup is desired.
Score distribution

Soundstage determines the space and environment of sound, as created by the headphones. That is, it determines the perceived location and size of the sound field itself, whereas imaging determines the location and size of the objects within the sound field. In other words, soundstage is the localization and spatial cues not inherent to the audio content (music), and headphones have to 'create' them rather than 'reproduce' them. This differs from imaging, which is the localization and spatial cues inherent to the audio content.

Our soundstage test evaluates PRTF Accuracy (Std. Dev.), PRTF Size (Avg.), PRTF Distance, Openness, Acoustic Space Excitation, and Correlated Crosstalk, for the tested headphones below.

Test Results

When it matters

A large, out-of-head, and in-front soundstage is mostly important to critical listening, home theater, and gaming use cases. A headphone with a good soundstage gives the sense that the music is happening in front of you, in a room you are sitting in, and not in a vacuum in your head. Creating a good soundstage is one of the hardest things to do for a headphone, and almost no headphone currently available can mimic a speaker-like soundstage.

Unless a large and speaker-like soundstage is needed or highly desired, most users will be perfectly happy with a headphone with a non-existent soundstage. It is possible that some people won't even notice the lack of it, or not be bothered by a small and inside-the-head soundstage, until they compare it with a good stereo loudspeaker setup or a pair of headphones with a good soundstage.

How it works

Our reference for the ideal headphone's soundstage is the ideal stereo loudspeaker setup, in the ideal room. Although no such ideal setup exists in the real world, this provides a good theoretical basis for determining the components of soundstage. Based on this framework, the differences between loudspeakers and headphones in terms of localization and spatial qualities, i.e. soundstage, can be divided into three:

  1. Room effects: With loudspeakers, the music is happening in a room where the sound is affected by reflections off the walls and other surfaces, as opposed to headphones where the music is happening at or in your ears.
  2. Angle and the sound source: Loudspeakers are positioned out in front and at a 30-degree angle, but headphones are usually at a 90-degree angle and very close to the eardrum. The pinna (the outer ear), which is part of the individual's HRTF (Head-Related Transfer Function), has a major role in helping the human auditory system to localize the angle and source of the sound source. Since the shape of the pinna is very complex and asymmetrical, both vertically and horizontally depending on the location of the source, different pinna resonances become active. These resonances add direction-specific patterns into the frequency response of the ears, which is then recognized by the auditory system for up/down, and front/back localization.
  3. Crosstalk: With a stereo loudspeakers setup, some of the sound from the left speaker gets to the right ear, and vice versa, because there’s no separating barrier between them. But with headphones, the sound of the L driver does not get to the right ear, due to the shadow created by the head. In a stereo speaker setup, the role of crosstalk is to strengthen the phantom center and to make the stereo image more cohesive; in other words, to remove the hole in the middle of the sound field, which is a characteristic of headphone soundstage.

Room Effects
(source: arqen.com)

Distance and Angle of the Source

Pinna Filtering Effects
(source: ucdavis.edu)

Crosstalk
(source: dirac.com)

Our tests

PRTF

In order to measure the amount of interaction between the pinna and headphones, a test was devised that we call PRTF, short for pinna-related transfer function. For this test, we got an extra ear for our dummy head, chopped its pinna off, and measured the frequency response of each headphone twice. Once with the intact ear, and once with the ear with the missing pinna. The difference between the two frequency responses is the PRTF of the headphone.

Headphone's PRTF = (Frequency Response w/ Pinna) - (Frequency Response w/o Pinna)

To come up with a reference to score our PRTF measurements against, a loudspeaker was used. It was positioned 1 meter in front of the HMS (our dummy head) and frequency response measurements were performed with and without pinna every 30 degrees. We picked the loudspeaker PRTF at 30 degrees as the reference, because that’s where loudspeakers are positioned in a stereo setup, but as the graph below shows, there's a clear pattern emerging as the head is rotated.

Reference PRTF Measurement Setup

Reference PRTF Measurement Setup

Loudspeaker PRTF at different angles

In the soundstage category in our reviews, PRTF calculations are divided into there: Accuracy, Size and Distance.

PRTF Accuracy (Std. Dev.)

What it is: The standard deviation of the PRTF (Pinna-related transfer function) of the headphones compared to a reference loudspeaker's PRTF at 30°. This quality is monaural and can be perceived even with one ear. This test does not apply to in-ears and earbuds, due to the lack of pinna interaction.
When it matters: An accurate pinna activation is mainly responsible for how natural and speaker-like the soundstage is perceived to be. The less error in the shape of the PRTF, the more natrual and speaker-like the perception of the soundstage will be. High amounts of error may indicate a soundstage that is unnatural or odd.
Good value: <2.5
Noticeable difference: 0.1
Score distribution

The accuracy value is derived by performing a standard deviation calculation of the PRTF response compared to the speaker reference PRTF, between 2KHz and 7KHz. In other words, it shows how accurate the shape of the response is in that region, compared to the reference PRTF response, and therefore, is mostly responsible for how natural the perception of soundstage will be.

As a rule of thumb, headphones with angled drivers and large and deep enclosures have the best PRTF responses. That's why over-ear headphones tend to do better than on-ears, earbuds, and in-ears in this test. Also, some headphones show unusual peaks and dips in their PRTF responses. One possibility is that these headphones have small ear cups, and they pinch or mangle the pinna in a way that creates an unusual response and the soundstage of these headphones may be perceived as odd and unnatural.

High PRTF accuracy

High accuracy but low activation

Decent accuracy with sharp peaks

Poor PRTF accuracy

PRTF Size (Avg.)

What it is: The average amplitude of the PRTF (Pinna-related transfer function) of the headphones compared to that of a reference loudspeaker's PRTF at 30°. This quality is monaural and can be perceived even with one ear. This test does not apply to in-ears and earbuds, due to the lack of pinna interaction.
When it matters: This value is responsible for the perceived size of the soundstage. The higher the value, the larger the perceived size of the soundstage. However, values above the reference (5.0dB) could result in a soundstage that is perceived as unnatural or odd.
Good value: >3.7
Noticeable difference: 0.1
Score distribution

The size value is derived by performing an average amplitude calculation on the PRTF response between 2KHz and 7KHz. In other words, it shows the amount of interaction between the headphone and the pinna, regardless of its accuracy, and therefore, is mostly responsible for how large the soundstage will be perceived.

As a rule of thumb, headphones with angled drivers and large and deep enclosures have the best PRTF responses. That's why over-ear headphones tend to do better than on-ears, earbuds, and in-ears in this test. Also, some headphones show an unusually high amount of pinna interaction in their PRTF responses. One possibility is that these headphones use the pinna to dampen the resonances of the enclosure, and when the pinna is removed for the PRTF test, those resonances are able to ring freely. In this case, the extra activation can be considered as noise in our measurements.

Good pinna activation

Mediocre pinna activation

Poor pinna activation

Excessive resonance in PRTF

PRTF Distance

What it is: The depth of the "10KHz notch" of the headphone's PRTF, which is caused by phase cancellations at the concha. This quality is monaural and can be perceived even with one ear. This test does not apply to in-ears and earbuds, due to the lack of pinna interaction.
When it matters: This value is mainly responsible for the perceived distance and elevation of the soundstage. A small distance value may result in a soundstage that is perceived to be located inside the head. Larger values may help pull the soundstage out from inside of the head and bring it to the front.
Good value: >13
Noticeable difference: 1
Score distribution

The distance value is derived by calculating the difference between the "average PRTF amplitude of the 2KHz-7KHz range" (i.e. PRTF Size), and the "lowest PRTF amplitude value in the 8KHz-12KHz range". In other words, PRTF distance looks for the depth of the 10KHz notch, which is mostly responsible for cues regarding the angle and elevation of the sound source.

As a rule of thumb, headphones with angled drivers and large and deep enclosures have the best PRTF responses. That's why over-ear headphones tend to do better than on-ears, earbuds, and in-ears in this test.

 

Good 10KHz notch

Good 10KHz notch but upshifted

Mediocre 10KHz notch

Poor 10KHz notch

It should be noted that with some headphones, the center frequency of the "10KHz notch" is slightly shifted upwards or downwards. Based on our tests, an upshift in the center frequency of the notch could actually represent cues for the elevation of the sound source. That is, the auditory system will localize the sound source in-front and slightly above, rather than just in front.

Blue: Loudspeaker PRTF at 30° in-front and 0° elevation
Red: Loudspeaker PRTF at 30° in-front and 15° elevation

PRTF | The 10KHz notch is upshifted in frequency

Openness

What it is: How open the headphones are, and how open and spacious they sound. This quality is monaural and can be perceived even with one ear. This test differentiates between acoustically and electronically produced crosstalk and only takes the acoustically generated crosstalk into account. This value is the inverse of the Noise Isolation test score, and could be indirectly related the acoustic impedance of the headphones.
When it matters: When a headphone with a sense of an open, and spacious soundstage is desired. A value of 10 indicates a fully open headphone, and a value of 0 indicates a fully closed headphone.
Good value: >7.5
Noticeable difference: 0.5

This value is the inverse of the noise isolation score. That is, the less isolating a headphone, the higher their openness value will be. Openness is included in the soundstage category, since open-back headphones, compared to closed-back headphones, tend to give the impression of a more open and spacious soundstage. However, the reputation that open-back headphones have regarding their soundstage hasn't been quite validated in scientific studies yet, and the sense of a more spacious soundstage may actually be related to the acoustic impedance of the headphones, rather than their enclosure properties.

 

 

 

Sennheiser HD 800 S Bass Frequencies

Isolation | Passive Open-back

Sennheiser Momentum 2.0 Bass Frequencies

Isolation | Passive Closed-back

Beats BeatsX Bass Frequencies

Isolation | Passive Closed-back

Isolation | Closed-back with ANC

 

Acoustic Space Excitation

What it is: How loud the headphones are, and how much they excite their environment acoustically. If the headphones are loud and open enough, the sound leaking from the headphones will be affected by the environment (reflections/reverb) before reflecting back into the open headphones and to the listener's ears. This quality is monaural and can be perceived even with one ear. This value is the inverse of the Leakage test score.
When it matters: Headphones with higher excitation values, similar to openness, tend to have soundstages that are perceived as more open and spacious.
Good value: >7.5
Noticeable difference: 0.5

This value is the inverse of the leakage score. That is, the more leaky a headphone, the more they acoustically excite they room they are in. Acoustic Space Excitation (ASE) is included in the soundstage category since the soundstage of certain headphones can be subtly affected by the room they are located in. For example, with very open and leaky (loud) headphones, it is possible to hear a small change in soundstage as one moves from a large dead environment to a small and reflective room.

As a rule of thumb, the smaller the driver of a headphone, the less their ASE will be. Also, open-back headphones excite their environment more than their closed-back counterparts. Therefore, closed-back in-ears usually have the least amount of ASE, and open-back planar magnetic headphones tend to have the highest amount of ASE.

Leakage | Closed-back In-Ear

Leakage | Open-back Earbuds

Leakage | Open-back Over-Ear

Leakage | Open Planar Magnetic

Correlated Crosstalk 

What it is: How strong and solid the phantom center is. This is mostly a stereo quality and its effects on mono content are minimal. This test is sensitive to the phase of the crosstalk and whether it is produced acoustically or electronically.
When it matters: When a true reproduction of the stereo image is desired. A value of 0 indicates no crosstalk, or that the existing crosstalk is not correlated enough to affect the phantom center. A negative score means the crosstalk is out of phase with the original signal, resulting in a slightly wider stereo image at the expense of creating a 'hole' at the center of the stereo field. A positive score means the crosstalk is in phase and positively affecting the phantom center.
Good value: >1dB
Noticeable difference: 0.3dB
Score distribution

This value is derived by calculating the amount of correlated crosstalk between the L and the R channels of a headphone. That's because uncorrelated crosstalk, which does exist to some amount in open-back headphones, doesn’t really affect with the phantom center and stereo cohesion.

So far, we haven't measured any headphones that would produce any significant amounts of correlated crosstalk.

Additional Information

At the moment, PRTF measurements are only performed on over-ear and on-ear headphones. That's because earbuds and in-ears fully bypass the pinna and don't interact with it, and their PRTF response would basically be a flat line, as can be seen in the examples below:

PRTF - Earbuds | No pinna interaction

PRTF - In-Ear | No pinna interaction

What's not included

  • Room effects
  • Head tracking
  • Distance of the sound source
  • Headphone acoustic impedance
  • DSP: soundstage effects created using digital signal processing

If you feel there is an item missing that should be included, please let us know in the Q&A section or via email.

Questions Found an error?

Let us know what is wrong in this question or in the answer.

Email:

Questions & Answers

2 ANSWERED QUESTIONS
1
I plan to purchase the Audeze Isine 20 which is a open back, planar magnetic, in ear, with an frequency range of 10hz-50khz. I absolutely love the system you guys use to evaluate headphones and would very much like to see it and it's baby brother the isine 10 reviewed on your website.
Thanks for your suggestions. We already have the Audeze headphones on our to-do list, but we don't have an estimate for when we will be able to review them.
0
My understanding is that your soundstage score is based on how closely the tested headphones match the sound of a stereo speaker setup that is located in front of the user and at a 30-degree angle. However, as a gamer I am interested in purchasing headphones that immerse me in a game's virtual world as accurately as possible (assuming the developers created spatialized audio for the game). I know that "Imaging" is the most important score for this, but I was wondering if soundstage has any relevancy in this case as well. If so, are there specific soundstage characteristics that I should look for if I'm mostly interested in experiencing the spatialized audio of a video game?
Unfortunately we haven't investigated the immersive features are video games yet, so can't give you a definitive answer. Most likely though, as you said, it is the imaging that would be most important for that. However, if the video game is using binnaural audio to create a 3D soundstage, then chances are in-ears/earbuds would actually perform better. This is because binaural audio tries to mimic pinna interaction, and if your headphones interact with your pinna as well, then it could distort the binarual information of the video game (it would be as if the sound has gone through two sets of pinnae instead of just one).
Questions Have a question?

Before asking a question, make sure you use the search function of our website. The majority of the answers are already here.

Current estimated response time, based on # of pending questions: 3.9 business days.

:
:
A valid email is required. We answer most questions directly by email to prevent cluttering the site.