Near and Far field definitions

[Andrew s]

I am aware of these term in EM / optics wher they refer to being either within a wavelength or so (near field) or several wavelengths away (far field).

Obviously, given the wide range of wavelengths in audio, it cannot be simply the same. So my question is what are the accepted distances or range of distance that are refered to as near, far and occasionally mid field when referring to monitors/speakers?

Thanks Andrew

 

[Frgirard]

It's depends if you are audiophile or pro.
In pro world a mid field is 2-2,5 m
Near field is under.
The goal is to hear the direct sound not the room.
Farfield is used in cinema studio. You need a distance to the screen.
In audiophile word, i don't no what is it.

I'm a music lover and use my neumann kh420 at 2,5 m in an optimized room.

 

[NTK]

Near field/far field can mean two different things on audio related fora such as this one.

The first is the technical meaning in acoustics or wave theories, which means the sound source is "far" enough away that it appears to be "small", and can be approximated as a point source (no different from EM and optics). The distance at which far field begins is frequency dependent (as you have already mentioned). Klippel gave a few rules of thumb (section 1.1): https://www.klippel.de/fileadmin/klippel/Files/Know_How/Literature/Papers/Holographic Nearfield Measurement of Loudspeaker Directivity.pdf

1. The listening distance must be larger than the largest dimension of the speaker -- pretty easy.
2. The listening distance must be larger than the wavelength -- difficult/impossible to achieve at low frequencies.
3. The ratio of the listening distance to largest speaker dimension must be larger than the ratio of the largest speaker dimension to wavelength -- may be difficult to achieve at high frequencies.

These are usually conservative estimates. For speakers with good diffraction control and inert cabinets, the dimension may be reduced to the size of the active driver, or the combined dimension of the two or more active drivers or ports when in the cross-over regions.

The Klippel NFS used by Amir and Erin is capable of determining where far field begins. The method it uses is to compare the real sound power to the apparent sound power (see sections 5.3 and 5.4 in the above link). Amir doesn't usually show this measurement in his reviews, but he did for the Dynaudio LYD 5 at 400 Hz. The result was 1.54 m, which was 1.8x the 400 Hz wavelength, and matched well enough to rule-of-thumb #2 (and satisfied #1 and #3).
https://www.audiosciencereview.com/...s/dynaudio-lyd-5-studio-monitor-review.15963/

The second meaning is the confusion of near field/far field as direct field/reverberant field. In almost all cases when people here talk about near field listening, they mean listening in the direct field. For a more detailed explanation, see: http://www.sengpielaudio.com/DirectFieldAndReverberantField.pdf

 

[Andrew s]

@NTK thanks for the links. The first paper is fascinating and fits what I recall from other wave physics.
@Frgirard thanks fir your practical post very helpful.
Regards Andrew

 

[RayDunzl]

So my question is what are the accepted distances or range of distance that are refered to as near, far and occasionally mid field when referring to monitors/speakers?

Uh....

I have a pair of adjacently located beamy scientifically disapproved electrostats, and a pair of JBL Linear Spatial Reference 308 biamped active DSP crossovered and tweaked speakers.

I listen to both at 10 feet.

I would call the perception of the beamy stats as more nearfield, and the dispersive JBL as closer to if not actually farfield.

This, based on the relative level of direct to reflected sound in my non audiophile approved listening space.

The stats have approximately a 10dB higher direct to reflected ratio, if I interpret this ETC plot as I am wont to do.

Red, JBL, Black, ML stats.

Time 0 = the (normalized) direct sound level (both peak at 0dB), after that is the hangover the room supplies.

 

[BoredErica]

The first is the technical meaning in acoustics or wave theories, which means the sound source is "far" enough away that it appears to be "small", and can be approximated as a point source (no different from EM and optics).

I'm assuming the distance at which a speaker's sound becomes approximated as a point source is the recommended minimum listening distance? If we sit closer that is bad for the sound?

1. The listening distance must be larger than the largest dimension of the speaker -- pretty easy.
2. The listening distance must be larger than the wavelength -- difficult/impossible to achieve at low frequencies.
3. The ratio of the listening distance to largest speaker dimension must be larger than the ratio of the largest speaker dimension to wavelength -- may be difficult to achieve at high frequencies.

These are usually conservative estimates. For speakers with good diffraction control and inert cabinets, the dimension may be reduced to the size of the active driver, or the combined dimension of the two or more active drivers or ports when in the cross-over regions.

It does seem very conservative to me. With Dynaudio LYD 5's largest cabinet dimension of .254m, the distance to fulfill rule 3 for 20khz would require a listening distance of 3.8m which is significantly over double the actual distance calculated by Klippel. To fulfill rule 2 at 100hz would require 3.432m.

I guess one factor that goes into this is where the drivers are. In a coaxial design it should allow for closer sitting distances. Of course there are limits to how far away a tweeter is from the other driver in a certain cabinet size.

The Klippel NFS used by Amir and Erin is capable of determining where far field begins. The method it uses is to compare the real sound power to the apparent sound power (see sections 5.3 and 5.4 in the above link). Amir doesn't usually show this measurement in his reviews, but he did for the Dynaudio LYD 5 at 400 Hz. The result was 1.54 m, which was 1.8x the 400 Hz wavelength, and matched well enough to rule-of-thumb #2 (and satisfied #1 and #3).
https://www.audiosciencereview.com/...s/dynaudio-lyd-5-studio-monitor-review.15963/

I wish Amir would do that test for every speaker.

The second meaning is the confusion of near field/far field as direct field/reverberant field. In almost all cases when people here talk about near field listening, they mean listening in the direct field. For a more detailed explanation, see: http://www.sengpielaudio.com/DirectFieldAndReverberantField.pdf

What would be the ideal room size for a speaker then? Is it a better experience to listen in reverberant field rather direct field? I'm guessing that's supposed to be the case, judging by how many people tend to listen far away in large to medium rooms. Are there any studies on whether people prefer direct field vs reverberant field for listening?

 

[gnarly]

The definition i've been taught, is that far field begins when all contributing drivers have settled into summation, where -6dB per doubling of distance takes hold across the spectrum.

Anything inside that distance can measure very wonky, mic position dependent....and it's nothing that you can assume min phase summation begins to apply to.

Rule of thumb i've come to respect, is try to measure 3x largest speaker dimension, to assure far field. Maybe 2x works.
But still, that's a tough nut to pull off.

 

[NTK]

I'm assuming the distance at which a speaker's sound becomes approximated as a point source is the recommended minimum listening distance? If we sit closer that is bad for the sound?

I don't know how audible these effects are. My understanding is that frequency response in the acoustic nearfield can change with distance, unlike when in the farfield there is only sound level attenuation. How much of it is audible? I have little idea. We know that because of room reflections, there is already a lot of comb filtering going on in a room that causes FR variations at differing listening locations, and our hearing is not very sensitive to them -- the reason we often do spatial averaging to smooth these variations out when measuring in-room FR.

Also, by definition, we are usually listening in the acoustic nearfield at low frequencies. What difference does that make, and do we notice?

It does seem very conservative to me. With Dynaudio LYD 5's largest cabinet dimension of .254m, the distance to fulfill rule 3 for 20khz would require a listening distance of 3.8m which is significantly over double the actual distance calculated by Klippel. To fulfill rule 2 at 100hz would require 3.432m.

I guess one factor that goes into this is where the drivers are. In a coaxial design it should allow for closer sitting distances. Of course there are limits to how far away a tweeter is from the other driver in a certain cabinet size.


I wish Amir would do that test for every speaker.

Yeah. I also think Klippel's definition is very conservative. When you look at the plot in Amir's Lyd 5 review, the "Total Apparent Power" curve levels off a long way before 1.54 m. But it is very difficult to tell the actual trend with a vertical scale of -100 dB to >200 dB. The criterion Klippel uses is the point where "Total Apparent Power" levels off to within 0.5 dB of the asymptote.

What would be the ideal room size for a speaker then? Is it a better experience to listen in reverberant field rather direct field? I'm guessing that's supposed to be the case, judging by how many people tend to listen far away in large to medium rooms. Are there any studies on whether people prefer direct field vs reverberant field for listening?

I don't know if there is an ideal room size. I'll quote Dr Toole's paper.

E-library page - AES

AES E-Library ← Back to search

www.aes.org

He was talking about speaker directivity in this paper, but I think it is applicable also to non-reflective vs reflective rooms or direct field vs reverberant field listening. When reflections are minimized, we get more pin-point imaging, but at the expense of losing spaciousness. So there is a delicate balance, which likely depends heavily on individual preference.

 

[Cote Dazur]

"The speaker should be closer to the listener's ears than it is to any boundary in the room. "
"The listener's ears should be closer to the speaker than they are to any boundary in the room."

I am a fervent near field listener and I have never seen those lines you have in your post, thank you for sharing, they make perfect sense to me. The room, any room, has such a destructive effect on what we hear, any attempt at minimizing the room effect is a great step in the right direction. Placing speakers close to the front walk in a near field set up is a rookie mistake, but it happens more often than not.

Are there any studies on whether people prefer direct field vs reverberant field for listening?

All I, or anyone else, need is a study of one, to tell me direct field is the way to go. So easy to do, if only temporatiraly, as a (free) trial. Move speakers and seat away from boundaries, don’t worry about so called integration, place speakers in a small equilateral triangle almost or at the level of your knees when sitting, sound will fill the room behind them, close your eyes, we see the music better with the ears, enjoy the music forget about “sound”. Most people do it wrong and try to fix their mistakes with considerable expense and efforts. It is not the system, but the room who is the issue. People selling speakers usually don’t agree.

 

[Cote Dazur]

I’ve found that the closer to near field I get the more immersed in the recorded soundstage and acoustics I get. But I find it starts suffering from the “headphone effect”
Like headphones it’s less room sound, but just as headphones don’t portray the physicality you get with speakers, with near fiend I start losing a sense of impact/punch/liveliness to the sound.

So for me it’s always been a trade off for how close I can get to hear more direct sound while staying far enough to keep that physicality. In my room, for my taste, that tends to be between 6.5 feet to 8 feet from the listening position, typically about 7 feet.

I will gladly continue the conversation in this thread, the other thread is confusing enough as it is , if you don’t mind.
My 2 setups are nearfield, 5’ between speakers 4.5’speakers to my ears in the big room, 3’5 between speakers and to my ears in the smaller room.
I have none of the “headphone effect” in neither room, their is 7 feet from the back of my speakers to the front wall and all the sound comes from behind the speakers. The sound stage is limited by the walls, ceiling and floor if I listen with my eyes open, but if I listen with my eyes closed, like I do 90% of the time, then the sound stage depends on the recording, most of the the iem it feels like this:

No speakers, no room, just musician somewhere playing music.
Since the speakers are so close, I have Plenty of SPL, more than I need, everything is crystal clear.
I know that it might not be for everyone, you need a dedicated room but for no money, except the sound system one presumably already own, a very spectacular stereo experience is available to anyone. Being away from boundaries is priceless.

 

[prerich]

@amirm Question...are there any requirements or special setups for Klippel measurements with speakers that are designed for use in the far-field only? I've come across speakers that state strictly intended for far-field use (namely professional PA speakers and some domestic audio gear that may have it's roots in the PA realm). Are NFS (Near-Field Scanner measurements) sufficient to measure these types of speakers, or are these the type of speakers that must be measured outside (due to their size) and in a quiet field free from ambient noise? If my question is stupid, forgive me - I'm just learning a few things here.

 

[DVDdoug]

I don't know if this helps. It's from Floyd Toole's book:

Main monitors
Large, usually in-wall installed, powerful full bandwidth systems capable of very high sound levels.

Mid-field Monitors
Medium-sized loudspeakers that may be full bandwidth or may use subwoofers, localized at moderate distance in front of the main console, positioned to minimize reflections from the working surfaces.

Near-field monitors
Small loudspeakers placed on the meter bridge of the recording console. The reflection for the working surface is part of the sound heard from these loudspeakers, and their locations may cause them to interfere with what is heard from main or mid-field monitors. Listeners are in the acoustical near field of the source, meaning that the small changes in head location cause changes in the sounds arriving at the ears.

 

[prerich]

I don't know if this helps. It's from Floyd Toole's book:

I listen exclusively in what would be the far-field (approximately 4 meters from my speakers - I've read that anything over 2.5 meters is the far-field).

 

[Waxx]

Far field exist, and the main difference is a narrow dispertion of the speakers. They can be tested like regular speakers on all, you just need to read the results different, especially on dispertion where a narrow dispertion is wanted.

 

[prerich]

Far field exist, and the main difference is a narrow dispertion of the speakers. They can be tested like regular speakers on all, you just need to read the results different, especially on dispertion where a narrow dispertion is wanted.

Thanks!!!!

 

[NTK]

@amirm Question...are there any requirements or special setups for Klippel measurements with speakers that are designed for use in the far-field only? I've come across speakers that state strictly intended for far-field use (namely professional PA speakers and some domestic audio gear that may have it's roots in the PA realm). Are NFS (Near-Field Scanner measurements) sufficient to measure these types of speakers, or are these the type of speakers that must be measured outside (due to their size) and in a quiet field free from ambient noise? If my question is stupid, forgive me - I'm just learning a few things here.

For a detailed discussion on using the NFS to measure sources such as line arrays typically used in PA and sound reinforcement applications, please take a look at this Klippel presentation.

https://www.klippel.de/fileadmin/klippel/Files/Know_How/Literature/Transparancies/Holographic%20directivity%20measurement%20of%20line%20sources%20and%20sound%20panels.pdf

 

[Ropeburn]

Practical dogetip for amateur musicians/producers (pros already know this and more):

Nearfield monitoring can be done with good results with any reasonably neutral speakers. Dedicated "studio monitors" aren't neccessary, only advantageous to a certain degree. As long as Speakers, placement, and room are above the "very low cost" variety (~1000 moneys the pair in the past, ~500 now) and a model and/or brand that is known for neutral tuning, you're good. The rest is all about the learning process about how your speakers' sound translates to other listening situations. Cross-referencing is key. Simplest approach: if it sounds good on your system, a crappy kid stereo or boombox, and your friend's/dad's/whatever 3000 moneys stereo, and in a car with a good stock soundsystem, it'll sound good everywhere.

 

[ahofer]

I don't know how audible these effects are. My understanding is that frequency response in the acoustic nearfield can change with distance, unlike when in the farfield there is only sound level attenuation. How much of it is audible? I have little idea. We know that because of room reflections, there is already a lot of comb filtering going on in a room that causes FR variations at differing listening locations, and our hearing is not very sensitive to them -- the reason we often do spatial averaging to smooth these variations out when measuring in-room FR.

Also, by definition, we are usually listening in the acoustic nearfield at low frequencies. What difference does that make, and do we notice?


Yeah. I also think Klippel's definition is very conservative. When you look at the plot in Amir's Lyd 5 review, the "Total Apparent Power" curve levels off a long way before 1.54 m. But it is very difficult to tell the actual trend with a vertical scale of -100 dB to >200 dB. The criterion Klippel uses is the point where "Total Apparent Power" levels off to within 0.5 dB of the asymptote.

I don't know if there is an ideal room size. I'll quote Dr Toole's paper.

E-library page - AES

AES E-Library ← Back to search

www.aes.org

He was talking about speaker directivity in this paper, but I think it is applicable also to non-reflective vs reflective rooms or direct field vs reverberant field listening. When reflections are minimized, we get more pin-point imaging, but at the expense of losing spaciousness. So there is a delicate balance, which likely depends heavily on individual preference.

View attachment 171179

I was unaware of this measurement. As I looked at Genelecs, and listened to my 8010s and my son’s 8030s, I began to wonder about how this property would be measured. The 8030s are in too large a room and they don’t sound as good as they should. My 8010s sound great - but only standing at my desk.

On the other hand, I have Revel F228be in a huge, reverberant, room and they sound wonderful.

 

[prerich]

For a detailed discussion on using the NFS to measure sources such as line arrays typically used in PA and sound reinforcement applications, please take a look at this Klippel presentation.

https://www.klippel.de/fileadmin/klippel/Files/Know_How/Literature/Transparancies/Holographic%20directivity%20measurement%20of%20line%20sources%20and%20sound%20panels.pdf

Thanks!

 

[Jim Taylor]

Many people (including me) routinely confuse nearfield/farfield with direct/reverberant field definitions. See here (Thanks go to @NTK )

https://sengpielaudio.com/DirectFieldAndReverberantField.pdf

Jim