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

[BenB]

*Yes, agreed. And in small rooms it is difficult to get long time delays. So what you think about heavy toe-in to increase the lateral reflections of the opposing speaker wall and at the same time reducing the levels at the nearest speaker wall?

I have never been a fan of this approach. I suspect it works well for things centered in the mix, but for something panned significantly left or right it has an odd effect that can actually decrease the apparent source width. As an example, consider a sound panned fully to the left. It comes out of the left speaker. But with sever toe in of a directional speaker, the right wall reflection is now louder than the left wall reflection. To maximize the soundstage, we want things panned to the left to be accentuated by reflections on the left, not the right.

 

[Floyd Toole]

I have never been a fan of this approach. I suspect it works well for things centered in the mix, but for something panned significantly left or right it has an odd effect that can actually decrease the apparent source width. As an example, consider a sound panned fully to the left. It comes out of the left speaker. But with sever toe in of a directional speaker, the right wall reflection is now louder than the left wall reflection. To maximize the soundstage, we want things panned to the left to be accentuated by reflections on the left, not the right.

Agreed. Add to this the fact that sound quality is substantially established by the first arriving sound. That should be the smoothest, flattest sound radiated by a loudspeaker - in good designs this is the on-axis response. Also, good designs show minimal difference between on-axis and listening window (average of 9 curves within the ranges +/- 30 deg. her, +/- 10 deg. vert) responses. Toe in of such speakers will do little to achieve time/intensity trades to expand the sweet spot. Such trades don't work well in any event because the trading ratio varies enormously with frequency and temporal envelope. Best to start with good speakers, aim them at the sweet spot, and sit there. Just sayin'

 

[Thomas_A]

I have never been a fan of this approach. I suspect it works well for things centered in the mix, but for something panned significantly left or right it has an odd effect that can actually decrease the apparent source width. As an example, consider a sound panned fully to the left. It comes out of the left speaker. But with sever toe in of a directional speaker, the right wall reflection is now louder than the left wall reflection. To maximize the soundstage, we want things panned to the left to be accentuated by reflections on the left, not the right.

I personally use wide dispersion speakers heavily toed in. I am not sure if the ear/brain knows how many times the lateral sound have bounced between the side walls if it is late enough, only that it is correlated and detectable at a certain level. Or? The nearest primary wall reflection would never be zero, of you use reasonably good dispersion speakers (and not narrow ones).

 

[Wes]

Darkness makes speakers "disappear " and can be measured with a light meter.

 

[Duke]

Agreed. Add to this the fact that sound quality is substantially established by the first arriving sound. That should be the smoothest, flattest sound radiated by a loudspeaker - in good designs this is the on-axis response. Also, good designs show minimal difference between on-axis and listening window (average of 9 curves within the ranges +/- 30 deg. her, +/- 10 deg. vert) responses. Toe in of such speakers will do little to achieve time/intensity trades to expand the sweet spot. Such trades don't work well in any event because the trading ratio varies enormously with frequency and temporal envelope. Best to start with good speakers, aim them at the sweet spot, and sit there. Just sayin'

I have never had good results with time-intensity trading using speaker which were not deliberately designed to be used that way. Ime the difference in results between speakers designed for it, and ones which are not, is significant.

I learned about the technique from Earl Geddes. He uses a 90-degree constant-directivity axisymmetric waveguide (crossed over to a large midwoofer where their patterns match), toed-in by about 45 degrees. The designed-for listening axis is actually about 20 degrees off-axis, so the increased high frequency energy directly on-axis is not actually part of the direct sound unless you are well off to one side of the room's centerline, in which case it is part of the direct sound from the FAR speaker. This axis is shown by the straight black line in the polar map, and by the red line in the polar graph:

This setup DOES forego the increase in apparent source width due to the early same-side-wall reflection, but Geddes claims that the upside is improved image precision, depth, and increased useable sweet spot width. Ime a bit wider than normal spacing can restore some of the lost soundstage width.

 

[Thomas_A]

The info so far still does not explain the apparent fixed center phantom image when I am listening in the sofa in front of the left speaker. It falls apart when moving closer to the speaker, but at nominal listening distance the center phantom image is dead center between the speakers. Without a center speaker. Is it reflections and toe-in?

 

[Duke]

The info so far still does not explain the apparent fixed center phantom image when I am listening in the sofa in front of the left speaker. It falls apart when moving closer to the speaker, but at nominal listening distance the center phantom image is dead center between the speakers. Without a center speaker. Is it reflections and toe-in?

Imo the KEY to time-intensity trading working well is for the output of the near speaker to fall off smoothly (and rather rapidly) as you move off-axis. This way as you move off to one side, the far speaker becomes louder than the near speaker in part because you are moving more on-axis of the far speaker, but MOSTLY because you are moving well off-axis of the near speaker. You want that near speaker's off-axis response to stay smooth so that the center image doesn't wander with changes in pitch.

When it's working right (which ime is at normal listening distances as you have found), my understanding is that the earlier arrival TIME of the near speaker is approximately offset by the greater INTENSITY of the far speaker.

I have multiple customers who sold their center-channel speakers after buying left and right speakers I designed because the dialogue was still on-screen for viewers well off to either side of the centerline, while the timbre of the phantom center channel was superior to that of their physical center channel speakers. And I'm talking about high-end center channel speakers which cost as much or more than one of my full-sized speakers.

There is one situation in which time-intensity trading does not work well even with suitable speakers, and that is for people who have a hearing imbalance (one ear hears significantly better than the other). For these people, imo a dedicated center-channel speaker is needed to anchor the dialogue on-screen.

 

[Newman]

I was going to suggest earlier that the speakers would best 'disappear' if there is more side wall reflection, not less, but I got the impression people thought more the opposite, so I let it be.

My thinking was that, with no side wall reflection, hard-panned 2-channel music would inevitably be 'nailed to' the one speaker and it can't 'disappear'. Add side wall sound and it 'blurs'.

 

[Newman]

loudspeakers with well behaved (smooth and fundamentally similar) off axis early reflected sounds (around 60 +/- degrees off axis) are awarded the highest sound quality ratings in double-blind listening tests - done in rooms with NO side-wall absorption.

Hi Floyd,

is there an optimal distance between the reflective side wall and the wide-dispersion speaker? Too close, and it becomes a problematic early-reflection 'second source' instead of merely adding apparent width: too far, and the amplitude drops off along with the effect of width?

cheers

 

[richard12511]

5. Floor reflection. If present it may reveal position, especially height localisation. This will challenge your credibility. See Section 7.4.7 in the 3rd edition. The most definitive test I am aware of was done in the Fraunhofer Institute flexible acoustics room. They concluded: "Regarding the floor reflection, the audible influence by removing this with absorbers around the listener is negative - unnatural sounding. No normal room has an absorbent floor. The human brain seems to be used to this." Humans evolved with something reflective below the feet.

I tested this for myself awhile back and came to the same conclusion. Placing a 6" absorber on the ground in between the speaker and the listening position actually eliminated a null in the response, but the sound was just off in a way that's tough to describe.

 

[Duke]

I tested this for myself awhile back and came to the same conclusion. Placing a 6" absorber on the ground in between the speaker and the listening position actually eliminated a null in the response, but the sound was just off in a way that's tough to describe.

I find this fascinating!

So if I understand correctly, objectively the measured response seemed to be improved, but subjectively the response was degraded?

In your opinion could this have been because the foam changed the spectral balance of the reflection (in addition to reducing its overall energy) so much that what remained was no longer clearly identifiable as "signal" (in this case a reflection) but rather became "noise" (no longer identifiable by the ears as a reflection)?

Another possible (i.e. speculative) explanation might be that the floor and ceiling bounces tend to partially "fill in" one another's dips, so when one of them is eliminated, the other's dip is no longer being partially filled in... ?

Or maybe the ear just hears the global removal of acoustic energy (from the initial and subsequent bounces of later reflections), whereas the original dip was a local event as two energy streams pass through one another with a time differential, but the net in-room energy is not actually reduced?

Apologies for the stream of conjecture, but this is very interesting to me.

 

[richard12511]

I use multichannel now and then as well, which works well with some but far from all recordings. Again, I think this is related to what model you wish to adapt to when listening to music. Do you want to be transferred to the event and ignore your room - go multichannel as far as possible. Do you want to transfer you room to the event, effectively using your room as a lounge with an transparent opening against the event? Go two- or three-channel audio, make sure that the wall behind the speakers are acoustically treated (within reasons!).

Have you tried Auro3D for upmixing stereo? I actually upmix all my music with Auro3D now. IMO, I think it does pretty much everything better than pure stereo. It's not a huge difference(which is why I like it). I almost never listen to stereo anymore since getting it. Unfortunately it requires 4 height channels, and Auro2D seems to be a completely different algorithm that I don't like at all . In fact, I think I prefer stereo over every other upmixer. What I don't like about most of them is that they put parts of the mix behind and/or above you. I like that Auro3D leaves the stereo image basically untouched and just adds additional envelopment via delayed reflected sound. I also love how customizable it is. You can customize both the delay of the reflections as well as their strength.

I hate that it requires 4 heights though. That makes it a no go for most. I wish Auro2D used same algorithm(just without the heights).

 

[Newman]

objectively the measured response seemed to be improved, but subjectively the response was degraded?

Not really. Someone is just working to the wrong definition of 'objective improvement', to say the above.

 

[richard12511]

This setup DOES forego the increase in apparent source width due to the early same-side-wall reflection, but Geddes claims that the upside is improved image precision, depth, and increased useable sweet spot width. Ime a bit wider than normal spacing can restore some of the lost soundstage width.

This is true ime. My JTR (+/- 30°) aimed like this definitely have a much better(precise) center image than my Revel speakers, and I think you might argue that the sweet spot is wider(due to time intensity trading). Where the JTRs lack is in that sense of spaciousness. The soundstage and spaciousness are really lacking, especially compared to the Revels.

 

[Newman]

I also recall @Floyd Toole in his first book, IIRC, concluding that all this talk about side wall reflections is really just for stereo reproduction, because of the inherent limitations of stereo at creating impressions of apparent source width and envelopment. In other words, it's a kludge.

If we want next-level experiences in soundscapes at home, then go multichannel, point the speakers at the listener(s), and give no regard to reflections off the near side walls, other than ensuring that they don't contribute negatively.

(And now I'm hoping I have not misrepresented Dr Toole.)

cheers

 

[JWAmerica]

Good to see real data on room response and psychoacoustics. If home consumers can design rooms with good acoustic traits, speakers should follow soon after.

 

[JWAmerica]

After reading Mr Toole's response, would it be fair to say that older speaker designs emulated headphone design, with a particular emphasis for on axis and an emphasis on eliminating off axis sound? It seems to me that modern speaker design accommodates for the room, whereas in the past there was a concerted effort to eliminate the room from the equation.

 

[MattHooper]

I mentioned in another thread: I have floor standing speakers that do a very good job of disapearing. But when I put spring-based pods under my speakers to decouple then from the wood floor they seemed to completely disappear in almost an electrostatic speaker way. The sound also tightened up in the bass and became cleaner and less congested seemingly throughout the full frequency range. It’s pretty amazing.
The thing is whenever I do this I am very pleased with the results but after a while I miss the sense of physical connection with the sound that I get when they are allowed to vibrate on the floor. There’s more physical presence and punch, so I take the spring footers back out.

 

[JWAmerica]

I mentioned in another thread: I have floor standing speakers that do a very good job of disapearing. But when I put spring-based pods under my speakers to decouple then from the wood floor they seemed to completely disappear in almost an electrostatic speaker way. The sound also tightened up in the bass and became cleaner and less congested seemingly throughout the full frequency range. It’s pretty amazing.
The thing is whenever I do this I am very pleased with the results but after a while I miss the sense of physical connection with the sound that I get when they are allowed to vibrate on the floor. There’s more physical presence and punch, so I take the spring footers back out.

I've also had good results decoupling full range floorstanders from the room. I use sorbothane pads but the effect was similar.

 

[Thomas_A]

I tested this for myself awhile back and came to the same conclusion. Placing a 6" absorber on the ground in between the speaker and the listening position actually eliminated a null in the response, but the sound was just off in a way that's tough to describe.

Have you tried a table instead?