• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required as is 20 years of participation in forums (not all true). Come here to have fun, be ready to be teased and not take online life too seriously. We now measure and review equipment for free! Click here for details.

About the "sweet spot"

Theo

Active Member
Joined
Mar 31, 2018
Messages
161
Likes
79
#1
Several manufacturers used to advertise for a wide dispersion pattern of their loudspeaker, like the Devialet Phantom (1 speaker being enough?) or omnidirectional ones like Ohmspeakers or Bose in the last century...
There is however a contradiction between a wide dispersion and a proper stereo image due to the interaction with the room. This is probably not a big deal if the system is used for background or wallpaper music.
The opposite choice is a focused pattern like those of Kii and D&D or more generally of monitor speakers. These come with the constraint of producing a very narrow "sweet spot", making the listening activity a selfish one.
I would personally like to have a system where the dispersion pattern is adjustable like Beolab 90 speakers, kind of out of my league though...
What would you choose?
 

SIY

Addicted to Fun and Learning
Technical Expert
Joined
Apr 6, 2018
Messages
919
Likes
1,048
Location
Chicago-Milwaukee axis
#2
c) None of the above.

We have to start with the admission that if we take a complex 3D soundfield and compress this to two channels, then try to reproduce that field again using two channels, that we're already massively handicapped. At that point, different paradigms of what the two channels should look like are questions of preference, not absolutes. Within these paradigms, we can come up with rules to optimize them (e.g., Toole's approach to point source reproducers), but they will ALWAYS sound different from one another, and NONE of them will be correct in the sense of reproducing a 3 dimensional sound field.

I've owned dipoles (panel electrostatics and magnetic planars) and point source speakers, and heard a variety of omnis. They're all wrong in different ways. True multichannel (not just "surround") is the step forward, but has not proved commercially viable.
 

Sal1950

Major Contributor
The Chicago Crusher
Joined
Mar 1, 2016
Messages
4,422
Likes
1,560
Location
Central Fl
#3
The opposite choice is a focused pattern like those of Kii and D&D or more generally of monitor speakers. These come with the constraint of producing a very narrow "sweet spot", making the listening activity a selfish one.
It's that "focus" that can create soundstaging that is eerily real.
I don't believe you'll ever be able to have it both ways no matter how much DSP and shotgun scattering, ie Beo90's, you throw in.
 

andreasmaaan

Major Contributor
Patreon Donor
Joined
Jun 19, 2018
Messages
1,801
Likes
1,049
#4
c) None of the above.

We have to start with the admission that if we take a complex 3D soundfield and compress this to two channels, then try to reproduce that field again using two channels, that we're already massively handicapped. At that point, different paradigms of what the two channels should look like are questions of preference, not absolutes. Within these paradigms, we can come up with rules to optimize them (e.g., Toole's approach to point source reproducers), but they will ALWAYS sound different from one another, and NONE of them will be correct in the sense of reproducing a 3 dimensional sound field.

I've owned dipoles (panel electrostatics and magnetic planars) and point source speakers, and heard a variety of omnis. They're all wrong in different ways. True multichannel (not just "surround") is the step forward, but has not proved commercially viable.
THIS^^^
 

andreasmaaan

Major Contributor
Patreon Donor
Joined
Jun 19, 2018
Messages
1,801
Likes
1,049
#5
The opposite choice is a focused pattern like those of Kii and D&D or more generally of monitor speakers. These come with the constraint of producing a very narrow "sweet spot", making the listening activity a selfish one.
Geddes' interesting approach is to actually utilise the narrower directivity of constant directivity horns to try to widen the sweet spot by pointing the mains across each other, so that they are aimed at a point in front of the listening plane.

This way, the response from each speaker is flat in the centre of the listening plane i.e. the prime listening position. As one moves to either side of the centre listening position, the response from the closer speaker starts to drop off, as the listener is now further off-axis from it, while the response from the further speaker remains flat. This tends to compensate for the difference in relative SPLs resulting from the listener being closer to one speaker.

As the off-centre listener is further from the speaker which is producing higher SPL on that axis, and closer to the speaker producing lower SPL, the inter-channel level balance is closer to equal.

Just whipped this up in MS Paint, it's obviously not to scale, but you can see how, for the listener in the far left position for example, the drop in SPL from the R speaker resulting from it being further away will be somewhat mitigated by the drop in SPL from the L speaker resulting from the listener being off-axis to it. For listeners closer to the centre position, the SPL from both speakers will be roughly the same. Such a setup can actually create a wider sweet spot than if both speakers were very wide directivity.

This would work even better in theory with a true constant directivity speaker like the 8C for example (but of course it can't go any way towards mitigating the inter-channel time difference for off-centre listeners).

1544095630342.png
 
Last edited:
Joined
Oct 25, 2018
Messages
60
Likes
137
Location
Stockholm, Sweden
#6
Here's an excellent test for speaker placement. Get hold of the album Amused to Death by Roger Waters. It's one of a handful of recordings using QSound technology that creates a 3d soundstage from two speakers. If your setup is correct, you'll hear a dog barking from the rear right and an old radio playing from the rear left on the first track. If you have everything set up correctly, the effect is uncanny. Any misalignment of the speakers and the result is gone.
 

Frank Dernie

Addicted to Fun and Learning
Patreon Donor
Joined
Mar 24, 2016
Messages
915
Likes
720
Location
Oxfordshire
#7
Here's an excellent test for speaker placement. Get hold of the album Amused to Death by Roger Waters. It's one of a handful of recordings using QSound technology that creates a 3d soundstage from two speakers. If your setup is correct, you'll hear a dog barking from the rear right and an old radio playing from the rear left on the first track. If you have everything set up correctly, the effect is uncanny. Any misalignment of the speakers and the result is gone.
I use this track too. The old soldier's narration is also way outside the left speaker much nearer the listener too when my system is well set up.
One of the early Devialet firmware updates for their dual monoblock setup didn't work well and it was blindingly obvious with this CD.
 

Dogen

Member
Patreon Donor
Joined
Aug 31, 2018
Messages
36
Likes
24
#8
I’ll take a system that sounds good from multiple locations in the room. Yes, a system focused on the sweet spot can yield stunning results, but music is a part of life, and in life, I move around! There’s something sad to me about clamping my head in a vise to enjoy the music.
 

RayDunzl

Major Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
5,553
Likes
1,511
Location
Riverview, FL
#9
I would personally like to have a system where the dispersion pattern is adjustable like Beolab 90 speakers, kind of out of my league though...

What would you choose?
I guess I have both, though i haven't thought about it that way. Focused ML/Electrostat/Dipole for the spooky stuff. Dispersant JBL for casual.

Selectable from remote. If it has to be adjustable, I can blend the two via the 2-zone preamp with separate volume settings.
 

Sal1950

Major Contributor
The Chicago Crusher
Joined
Mar 1, 2016
Messages
4,422
Likes
1,560
Location
Central Fl
#10
Here's an excellent test for speaker placement. Get hold of the album Amused to Death by Roger Waters. It's one of a handful of recordings using QSound technology that creates a 3d soundstage from two speakers. If your setup is correct, you'll hear a dog barking from the rear right and an old radio playing from the rear left on the first track. If you have everything set up correctly, the effect is uncanny. Any misalignment of the speakers and the result is gone.
I'd not found the images being placed much to the rear, but that dog bark was mid-wall on the right and I could swear he was about 2 blocks down the street, so far away it sounds, close to the same on the left radio broadcast.
Always used those tracks to amaze/amuse friends back when it was first released. I don't remember trying it on my current rig, give it a go later on when the house quiets down.
 
Joined
Oct 25, 2018
Messages
60
Likes
137
Location
Stockholm, Sweden
#11
I'd not found the images being placed much to the rear, but that dog bark was mid-wall on the right and I could swear he was about 2 blocks down the street, so far away it sounds, close to the same on the left radio broadcast.
Always used those tracks to amaze/amuse friends back when it was first released. I don't remember trying it on my current rig, give it a go later on when the house quiets down.
Yeah, you're right, it should come from the sides. You can find a list of all (not many) of the recordings using Qsound. A lot of crap, but now I need to get hold of some of them.
http://www.qsound.com/spotlight/users/recording-artists.htm
 

Sal1950

Major Contributor
The Chicago Crusher
Joined
Mar 1, 2016
Messages
4,422
Likes
1,560
Location
Central Fl
#12
Yeah, you're right, it should come from the sides. You can find a list of all (not many) of the recordings using Qsound. A lot of crap, but now I need to get hold of some of them.
http://www.qsound.com/spotlight/users/recording-artists.htm
I did just give it a quick try and it's not performing as it should on my current system. I'll do some testing with different speaker aiming/placement later on.
But Holly Cow, you should hear how Dolby Surround / DTS Neural X code latches onto the QSound matrix. Damn near sounds like a discrete 5.2.4 multich recording. LOL
 
Joined
Apr 28, 2016
Messages
335
Likes
150
Location
Northeast US
#13
Geddes' interesting approach is to actually utilise the narrower directivity of constant directivity horns to try to widen the sweet spot by pointing the mains across each other, so that they are aimed at a point in front of the listening plane.

This way, the response from each speaker is flat in the centre of the listening plane i.e. the prime listening position. As one moves to either side of the centre listening position, the response from the closer speaker starts to drop off, as the listener is now further off-axis from it, while the response from the further speaker remains flat. This tends to compensate for the difference in relative SPLs resulting from the listener being closer to one speaker.

As the off-centre listener is further from the speaker which is producing higher SPL on that axis, and closer to the speaker producing lower SPL, the inter-channel level balance is closer to equal.

Just whipped this up in MS Paint, it's obviously not to scale, but you can see how, for the listener in the far left position for example, the drop in SPL from the R speaker resulting from it being further away will be somewhat mitigated by the drop in SPL from the L speaker resulting from the listener being off-axis to it. For listeners closer to the centre position, the SPL from both speakers will be roughly the same. Such a setup can actually create a wider sweet spot than if both speakers were very wide directivity.

This would work even better in theory with a true constant directivity speaker like the 8C for example (but of course it can't go any way towards mitigating the inter-channel time difference for off-centre listeners).

View attachment 18479
Any chance Geddes wrote a paper on this, or any links to where he goes into more detail? Thanks.
 

svart-hvitt

Major Contributor
Joined
Aug 31, 2017
Messages
1,575
Likes
628
#14
The Big Elephant in the room is point source. Nobody is interested in it because nobody - except the exception(s) - focus on it (heard about the drunkard who lost his keys in the night...).

As soon as you have a point source speaker, this discussion would take another direction, to illuminate other nuances.

Obviously, you will NOT experience stereo outside of sweet spot with a point source. But you would be better positioned (sic!) to understand the shortcomings of stereo if you had a point source.
 

andreasmaaan

Major Contributor
Patreon Donor
Joined
Jun 19, 2018
Messages
1,801
Likes
1,049
#15
Any chance Geddes wrote a paper on this, or any links to where he goes into more detail? Thanks.
TBH, I can't recall specifically what I read originally. And I can't find much online now. Might be one of those things you need to scurry down the audio forum rabbit hole for :confused: He is or was very active on DIYaudio IIRC...
 

Wombat

Major Contributor
Joined
Nov 5, 2017
Messages
2,886
Likes
1,179
Location
Australia
#16
I’ll take a system that sounds good from multiple locations in the room. Yes, a system focused on the sweet spot can yield stunning results, but music is a part of life, and in life, I move around! There’s something sad to me about clamping my head in a vise to enjoy the music.
My avatar in words.
 

andreasmaaan

Major Contributor
Patreon Donor
Joined
Jun 19, 2018
Messages
1,801
Likes
1,049
#17
The Big Elephant in the room is point source. Nobody is interested in it because nobody - except the exception(s) - focus on it (heard about the drunkard who lost his keys in the night...).

As soon as you have a point source speaker, this discussion would take another direction, to illuminate other nuances.

Obviously, you will NOT experience stereo outside of sweet spot with a point source. But you would be better positioned (sic!) to understand the shortcomings of stereo if you had a point source.
@svart-hvitt, I remember you posted a link to a Genelec presentation on this topic that I'd been meaning to watch but have now lost... Do you remember what it was I was talking about?
 

Blumlein 88

Major Contributor
Joined
Feb 23, 2016
Messages
4,343
Likes
2,624
#18
Yeah, you're right, it should come from the sides. You can find a list of all (not many) of the recordings using Qsound. A lot of crap, but now I need to get hold of some of them.
http://www.qsound.com/spotlight/users/recording-artists.htm
I have Amused to Death. Was just messing with it in my video system, using front stereo channels only.
At my main seating area they are 75 degrees apart. I can move to 90 degrees or move backward until they are 40 degrees.
Seems to me the Qsound is more consistent and works best at speaker angles of no more than 60 degrees. Less seems better and more dramatic. At wider angles the sound moves to different locations. Standing also makes a big difference vs sitting when speaker angles are greater than 60 degrees. So maybe they optimized this for 60 degree speaker angles or less. Far back in this room the effects are very noticeable and huge in apparent size.

I didn't know Shawn Colvin's a Few small repairs was Qsound. I don't recall it being stated on the CD liner notes. I just listened and it is a good sounding album, but no imaging effects were notable as being beyond stereo capabilities. I've seen her live a couple times just before this album. One was in horrendous music hall which was converted from an A&P grocery store. The sound in there that night was atrocious. I felt embarrassed for her. She and the previous performer started the first song and then stuttered, stopped and started over. The sound was that bad. The place still had vinyl tile flooring, and poles with no sound treatment. A make shift stage and bars right and left. She did a good job and gave a good performance though.
 

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
1,414
Likes
542
Location
Zg, Cro
#19
I'd not found the images being placed much to the rear, but that dog bark was mid-wall on the right and I could swear he was about 2 blocks down the street, so far away it sounds, close to the same on the left radio broadcast.

I have that recording and it also sounds very "wide" to me. Nothing is coming from the rear but there are many sounds coming from the sides of the speakers.

Btw, I also find phone ringing at the end of "What God Wants, Part III" very realistic.
 
Last edited:

Wombat

Major Contributor
Joined
Nov 5, 2017
Messages
2,886
Likes
1,179
Location
Australia
#20
Shall we have some scientific input or just have a vote on it?
 

Similar threads

Top Bottom