What makes speakers "disappear " and can it be measured?

[Pearljam5000]

I noticed that some speakers just disappear and it feels like the sound appears out of thin air, and others don't feel like that at all, and the sound is just coming out of 2 speakers.
What makes speakers "disappear" and can it be measured?

 

[ppataki]

This is just my 2 cents based on experience after having close to a dozen different speakers

I managed to get holographic image if:

- the speaker is point-source or appears to be point-source (=in a multiway system the drivers are time aligned to the listening position). And you can measure this with the Step Response curve

- there is enough physical space (air) around the speakers (=at least 50cm away from all the walls, and the more the better)

 

[AnalogSteph]

- there is enough physical space (air) around the speakers (=at least 50cm away from all the walls, and the more the better)

Makes sense - I tend to associate "imaging stick to the speakers" with diffraction issues.

Some of the best-imaging speakers tend to be ones that are as smooth as the proverbial baby's behind, with waveguides to take care of the time alignment (they can also be used to keep HF from "sticking" to the surface, so any edges that do exist will be less visible to the tweeter, hence less diffraction).

I imagine some of the worst candidates would have to be early-'70s constructions with domes mounted flat on an inset baffle (guess why lining the outside of the tweeter with foam got popular) and woofer mounting from the top.

 

[tecnogadget]

Point source + minimum phase system = invisible speakers

 

[Chrispy]

What speakers in what room situation did you "experience" this with? No other way?

 

[Snarfie]

Time aligent/phase coherent build speakers combined with roomcorrection is my experience. However setup is crucial.

 

[BenB]

In my experience, adding anti-diffraction frames to the Sony SS-CS5 improved the imaging of the speaker. Details here:
https://www.avsforum.com/threads/improving-the-sony-ss-cs5.3068362/

The frames change the on-axis and off-axis frequency response (and lower the baffle step frequency). I can't say that I see much difference in the measured impulse response, but perhaps a step response would have revealed something. There can't be different frequency domain behavior without also having different time domain behavior. I can't isolate on-axis changes (to the direct sound) from off-axis changes (to the reflected sound). I suspect the imaging improvements were associated with the on-axis changes, because that speaker is fairly well behaved off-axis even without the modification.

Still, even with diffraction mitigated, they don't disappear as well as my on-wall multi-way line-array speakers, which disappear better than any other speaker I've heard. The room reflections give us information about the location of the speaker in the room. My arrays do a good job of mitigating floor and ceiling reflections, and the placement of them on the wall (which they were specifically designed for, with a wide, shallow, wedge shape) effectively eliminates a delayed front-wall reflection. The result is music with no apparent attachment or association with the speakers.

In my experience, stereo sound reproduction helps speakers disappear. It's much harder for one speaker played by itself to disappear in a room. The result can be an inconsistent: sounds toward the middle of the soundstage are perceived as separated from the speakers, but sounds panned significantly to one side or the other are perceived as coming from the speaker. This is an area where my on-wall line arrays have significantly more consistency than other speakers.

 

[Plcamp]

there is enough physical space (air) around the speaker

Linkwitz suggested one metre or more from walls in order to prevent reflections earlier than 6 ms.

 

[sigbergaudio]

Actually locating the speakers away from walls isn't necessary to achieve this, despite that the need for space around the speakers feels like a logical requirement.

This invisibility and large soundstage was one of the design goals of our new active speakers. They are point sources (coax) and in phase across all drivers, but are designed to be located close to the back wall. And they achieve this "disappearing" act as well as large and accurate soundstage both with regards to stereo perspective and perceived depth, even when they're just 10 cm (4 inches) from the wall.

 

[Plcamp]

Actually locating the speakers away from walls isn't necessary to achieve this, despite that the need for space around the speakers feels like a logical requirement.

This invisibility and large soundstage was one of the design goals of our new active speakers. They are point sources (coax) and in phase across all drivers, but are designed to be located close to the back wall. And they achieve this "disappearing" act as well as large and accurate soundstage both with regards to stereo perspective and perceived depth, even when they're just 10 cm (4 inches) from the wall.

That’s very interesting. Why don’t early reflections mess with brain’s ability to detect location of the point source?

 

[sigbergaudio]

That’s very interesting. Why don’t early reflections mess with brain’s ability to detect location of the point source?

There's a few things going on. First the frequencies of the speakers don't spread evenly in all directions. Behind the speaker it's mainly the lower frequencies that are not attenuated, so most of the reflections from that wall will be low frequency content. Related subject: This leads to the SBIR effect causing dips typically in the 100-200hz area, and to combat that, having the speakers close to the wall is actually beneficial (ideally they should be IN the walls).

With regards to reflections in general, what creates good imaging and accurate perspective is a speaker with an even dispersion pattern, not the distance from the wall. A predictable off-axis response helps us recognize reflections from the walls for what they are (reflections, not direct sound), and allows us to maintain the illusion of a point source and accurate imaging.

Here's a pretty good article on the subject of speaker placement https://sonicscoop.com/2017/12/14/the-1-speaker-placement-tip-speaker-manuals-get-completely-wrong/

 

[Marc v E]

Some observations I made with my own speaker that easily disappear:

They image better when 50 cm from the wall
Same observation when I moved the tv that stands between them closer to the wall.
They beam the sound wide and evenly without altering the pitch of the sound.
(Cones are alligned too of course)

 

[Plcamp]

A predictable off-axis response helps us recognize reflections from the walls for what they are (reflections, not direct sound), and allows us to maintain the illusion of a point source and accurate imaging.

Is this essentially saying that the time delay of reflections doesn’t matter if the reflection frequency response is very similar to the direct?

 

[sigbergaudio]

Is this essentially saying that the time delay of reflections doesn’t matter if the reflection frequency response is very similar to the direct?

I wouldn't go that far. For instance having one speaker close to the side wall wouldn't be ideal, but even in such a situation our brains are quite good at compensating and understanding what is going on. And in general less reflections is better (as in dampen your listening space).

 

[Marc v E]

In my experience, adding anti-diffraction frames to the Sony SS-CS5 improved the imaging of the speaker. Details here:
https://www.avsforum.com/threads/improving-the-sony-ss-cs5.3068362/

The frames change the on-axis and off-axis frequency response (and lower the baffle step frequency). I can't say that I see much difference in the measured impulse response, but perhaps a step response would have revealed something. There can't be different frequency domain behavior without also having different time domain behavior. I can't isolate on-axis changes (to the direct sound) from off-axis changes (to the reflected sound). I suspect the imaging improvements were associated with the on-axis changes, because that speaker is fairly well behaved off-axis even without the modification.

Still, even with diffraction mitigated, they don't disappear as well as my on-wall multi-way line-array speakers, which disappear better than any other speaker I've heard. The room reflections give us information about the location of the speaker in the room. My arrays do a good job of mitigating floor and ceiling reflections, and the placement of them on the wall (which they were specifically designed for, with a wide, shallow, wedge shape) effectively eliminates a delayed front-wall reflection. The result is music with no apparent attachment or association with the speakers.

In my experience, stereo sound reproduction helps speakers disappear. It's much harder for one speaker played by itself to disappear in a room. The result can be an inconsistent: sounds toward the middle of the soundstage are perceived as separated from the speakers, but sounds panned significantly to one side or the other are perceived as coming from the speaker. This is an area where my on-wall line arrays have significantly more consistency than other speakers.

Maybe that's a thing I overlooked. My speakers are oval shaped, without any sharp edges.

 

[gsp1971]

I wouldn't go that far. For instance having one speaker close to the side wall wouldn't be ideal, but even in such a situation our brains are quite good at compensating and understanding what is going on. And in general less reflections is better (as in dampen your listening space).

@sigbergaudio

Speaker manufacturer Focal recommends that the surface behind the speakers should be reflective (to allow the sound to fully develop), whereas the surface behind the listening position should be absorbing.

Would you agree with that or is it speaker-specific, i.e. it holds true for the specific design and performance of Focal speakers?

Thx in advance.

 

[anphex]

Considering all the speakers I had, the most "invisible" ones were the ones with optimized enclosure (rounded corners, non-rectangular case), super low distortion and perfect timing of the chassis and crossovers. I love vertical Dàppolitos for sure because of the fewer reflections from ceiling and floor and the sound kind of "appears" in your head if done well.

Edit: Oh, and the sound center shouldn't be close to the ground or any hard surface if possible.

 

[YSC]

Off topic, invite me to your home, then you walk away for 15min, and you can have your speaker disappear

back to topic, I think it's to have the proper recording as your source (say point source with some coaxial speaker, or some classical with electro stats etc.) then do proper room treatment to absorb reflection, and that will replay pretty convincingly

 

[gsp1971]

I noticed that some speakers just disappear and it feels like the sound appears out of thin air, and others don't feel like that at all, and the sound is just coming out of 2 speakers.
What makes speakers "disappear" and can it be measured?

Closing you eyes?

 

[Balle Clorin]

This one is wonderful.
Not many can beat the Audio Physic Virgo II , Even today
https://www.stereophile.com/content/audio-physic-virgo-loudspeaker-measurements-part-3