Speaker toe in

Has anyone heard of The Tannoy Method used for speaker tie in? I have a picture I wish I could upload showing this method used on some Acoustic Research speakers. The speakers are toed in quite a bit past the listening axis. Is there a benefit? One person claims it take the room out of the equation. Thoughts?
What Kenjit meant to say is do what he did and sell your system and buy a casio radio because no speaker has been tuned to your ear and personal preferences. 

The casio radio requires no toe in...
It won't take the room completely out of the equation, but it may reduce the intensity of the side wall reflections.

Also, many speakers have tweeters which just sound better off-axis, instead of pointed directly at you. Toeing in so they cross in front of you can benefit both issues, and is a good trick for small space with live rooms.  Too much of this and you risk the same issue but from the back wall.

Of course, one of the first things you should consider is room acoustics, and I always recommend GIK acoustics products and advice.


"Has anyone heard of The Tannoy Method used for speaker toe in?"

I have not heard that name before.

"The speakers are toed in quite a bit past the listening axis."

I have been making speakers intended to work well with this sort of configuration for about fifteen years.

"Is there a benefit?"

The short answer is, there are at least three benefits when done right: Less coloration, deeper soundstage, and wider sweet spot. There is one possible detriment: The soundstage may not be as wide.  And, the technique probably will not work well with most speakers.

Warning, the long answer gets a bit geeky and technical. Please skip if that sort of thing doesn’t appeal to you (I post this because my sometimes geeky posts have offended people).

First a bit of background on relevant speaker design, which probably applies to the Tannoys: Imo the kind of speakers you want for this are ones with a fairly narrow and uniform radiation pattern over much of the spectrum. For example, you could use a 90 degree pattern (-6 dB at 45 degrees off-axis) constant-directivity horn crossed over to a large diameter midwoofer at the frequency the midwoofer’s pattern has narrowed to 90 degrees. Tannoy’s concentrics probably have similar radiation patterns, with the midwoofer cone acting as a constant-directivity horn.

Next, a bit of background on the effects of early sidewall reflections: Early reflections are more likely to cause coloration and degrade clarity than later ones, and strong early same-side-wall reflections play a role telling the ear you are in a small room which tends to overlay the soundstage on the recording, and in particular soundstage depth. Strong early same-side reflections do however broaden the soundstage’s apparent width, albeit at the expense of some image specificity, because the early sidewall reflection is smearing the image widths a little. (Personally I prefer to just move the speakers a bit further apart.)

Finally, a bit of background on how the ear localizes sound: The ear uses two mechanisms to localize sound: arrival time, and intensity. With a normal speaker setup (mild toe-in), when you are in the center sweet spot, the arrival times and intensities are the same from both speakers. If you move your listening chair to one side, the near speaker naturally "wins" arrival time, but it also "wins" intensity, because you are now on-axis (or nearly so) of the near speaker but you have moved further off-axis of the far speaker. So the image tends to shift towards the near speaker even further than your chair has shifted.

Now imagine speakers having the proper radiation pattern and 45 degrees of toe-in, such that their axes criss-cross in front of the listening area. For the listener off to one side, the near speaker naturally "wins" arrival time, but the far speaker "wins" intensity! This is because you are now on-axis (or nearly so) of the far speaker but well off-axis of the near speaker. These two localization mechanisms offset one another somewhat such that you still get a decent soundstage, though of course imaging is best up and down the centerline. The key is, the output of that near speaker must fall off smoothly and rapidly as you move off-axis. Relatively few speakers meet this criterion... most speakers have a pattern that is too wide and changes too much with frequency for this technique to give good results.

This well-controlled radiation pattern combined with the extreme toe-in essentially eliminates the early same-side-wall reflection. In fact the first significant sidewall reflection for the left speaker will be the long, across-the-room bounce off the right side wall, and vice-versa. So the first significant lateral reflections arrive at the opposite ear from the first-arrival sound, and this relatively late-onset cross-correlated reflection is generally interpreted by the ear/brain system as spaciousness rather than coloration and image widening. And thus with less small-room signature superimposed atop the soundstage on the recording, we hear more of the recording and less of the room.

I wouldn’t say the technique "takes the room out of the equation", but imo it does usefully reduce detrimental room effects. At an audio show where we used this sort of setup, an electronics manufacturer who was sharing the room with us for the first time remarked that it was the first show where they were not "fighting the room the whole time". (We were doing something unorthodox in the bass region as well, which is beyond the scope of this thread.)


So this doesn't make the soundstage laterally narrower??

If you mean the quality of the perceived image, this is a good trade-off.

Ideally, you'd be listening in a much wider room, allowing you to point the tweeters straight ahead, and you could listen in between them, but if the room is your problem, this is a possible compromise.

However, if you mean the listening location, well,  yeah, very much so, but you are in a narrow room anyway for you to consider this setup.
Very interesting! I’m trying to dial a larger speaker in a smaller room with Cape Cod ceilings. I have an older pair of B&W 801’s S3. These speakers perform unbelievably well in a larger room where I once had them. Since then I purchased a pair of Ariel Acoustics #9’s which is being used in that system. I’m not willing to part with the 801’s and this is why I’m still searching different avenues to get the best sound I can get from them in their new space. The 801’s don’t sound horrible but I know their potential and don’t believe I’m anywhere near it or probably will get to it in this room. I have acoustic panels set up behind the speakers on the back wall and ant the first reflection points. I saw a pic of a room at a show with this Tannoy set up and thought I’d throw it out to the community to get thoughts from the experts lol!
Duke, that was an interesting explanation of toe-in effects. I'm using a pair of KEF Ref 1s with Uni-Q drivers previously aimed straight forward in a 14’ X 19’ room with bookcases and hanging art, but now I’m trying toe-in aimed just behind the centered listening position. Not certain which I prefer. Do you have any suggestions for toe-in with Uni-Q drivers?

The Ref 1s are superb for the chamber and jazz music I prefer, but seem lack the slam my KEF Ref 107/2s had for large orchestrations. I suppose it’s trade-off.

Yep, that’s how I toe-in my speakers. I also use corner placement and sealed bookshelves. In a sense, I don’t really have sidewalls. There’s an article somewhere online that talks about this. The guy talks about how he went from a heavily treated room with a traditional speaker setup to a corner placement and it sounded better without any treatment.
"So this doesn’t make the soundstage laterally narrower??"

With the right kind of speakers (described in my post), not at all, except for the loss of the image broadening effect of early sidewall reflections. The soundstage width is normal, and the depth and sense of immersion in the acoustic space of the recording have been reported as better than normal. If desired, soundstage width can be increased by increasing the speaker spacing.

I get many questions about (and sometimes well-meaning offers to "fix") my speaker setup at audio shows. They are usually skeptical when I explain why my setup is correct for my speakers. Then I ask them to close their eyes and tell me whether they hear whatever problem they expected, and they never say they do. Then I ask them to sit in whichever chair is farthest from the centerline and tell me if they still hear a decent spread to the instruments. And they always say they do.

In my experience, intermediate amounts of toe-in usually do not work well. I would suggest either going all the way to axes criss-crossing in front of the center sweet spot (assuming you have suitable speakers), or just sticking with a normal configuration.

"The guy talks about how he went from a heavily treated room with a traditional speaker setup to a corner placement and it sounded better without any treatment."

This makes sense to me, assuming by "heavily treated" he means "lots of absorption". Here’s why:

The ear derives timbre from both the first-arrival sound and the reverberant energy. When there is a significant spectral discrepancy between the two, like when most of the highs in the reverberant sound have been absorbed by foam, the net result sounds unnatural and can become fatiguing. Sometimes using a lot of absorption is the lesser of two evils, but imo it makes sense to seek other solutions first.

I’ve said it before, most speakers are placed too far apart, if they were closer together toe-in would be unnecessary in most cases. I’d opine 4-5 feet would be correct or at least a good starting place for most speakers in most rooms. This can be verified using the speaker set up track on the XLO Test CD.
And in a long-wall set-up?  Presumably you aren't so worried about side-wall reflections...

Twoleftears said: "And in a long-wall set-up? Presumably you aren’t so worried about side-wall reflections..."

Very true.

However the proximity of the wall behind the speakers could constrain the soundstage depth (assuming you can’t pull them out into the room as far with a long-wall setup). I’m not saying the one configuration is more right than the other, but ime there are tradeoffs which often require trying it both ways in order to evaluate reliably.

Imo the arguments for speakers with good radiation pattern control go well beyond minimizing sidewall reflections, but that is one of them.

You know, if your ears weren’t both on the left-hand side of your head, all this talk about soundstaging might make more sense...

Sorry, I couldn’t resist!!!


Great job of explaining Duke. Before I became committed to ESLs I had several speakers that did best with extreme toe in. It all depends on the dispersion pattern of the speaker. The brains method of localization, phase and volume also explain why line source speakers can have a wider sweet spot if their dispersion is uniform throughout the listening area. Their volume does not drop off near as fast with distance and if they are dipoles there is no radiation to the side walls. This is one of the reasons both Duke and I are attracted to Soundlabs speakers.
In the end it comes down to what you like best so try toeing your speakers in farther and pay attention to what the image does as you move away from the sweet spot. High frequencies may drop off as you toe in the speaker. You may have to compensate. If the speakers were bright to start with they may sound better. Much better. 
Duke, we all know you have two left ears. But, your two right ears balance it out fine:)

Post removed 
Mijostyn wrote, regarding a side effect of extreme toe-in: "High frequencies may drop off as you toe in the speaker."

Yup! Severe toe-in can result in less high frequency energy in the first-arrival sound, but correspondingly a bit more in the reverberant field (which is typically starved of high frequencies relative to the first-arrival sound).

The solution is, tilt the tweeter’s response up a bit! This kills two birds with one stone: It corrects the spectral balance of the first-arrival sound, and further improves the spectral balance of the reverberant field. So the tonal balance is now better throughout the room.

Since 2006 I have been making speakers with user-adjustable high frequency tilt, via a single high-quality changeable resistor (in an dedicated terminal cup) on the back of the speaker cabinet. Imo this is superior to a variable L-pad. Typically this external resistor is bypassing a resistor in a particular location on the crossover board.

Anyway you correctly identified one of the potential downsides of severe toe-in, and that downside is something I should have thought to mention earlier. It hasn’t been an issue for me from the beginning because the tilt adjustment has been built into all of my controlled-pattern speaker systems, and that’s probably why I didn’t think of it.

Thanks for bringing it up! 

 I have an electrostatic hybrid Martin Logan summit X speakers.  My room is well actively long and narrow with the speaker sitting on the long wall in the listening position obviously being on the opposite side long wall. I have my speakers slightly towed in. I wonder if anyone has any thoughts on whether the idea of towing or not is Different for an electro static design?
rbodner, It is extremely different. ESLs beam like crazy. There are ZERO high frequencies beyond the border of the panel.  MLs are curved to increase their dispersion a little but if you toe the speaker in to far the high end will jump off a cliff. ESLs should be aimed directly at the listening position in most installations.
Ohm suggest that their Walsh series be toed in for an increase in lows, toe out for an increase in highs.  I rotated my Walsh 5000s and found the image popped into focus with a slight toe-in.  I have always found that it is best to close your eyes and trust your ears when making adjustments.  As all things audio, your mileage may vary.

When I had my Spendors, the manual suggested that set up.  I think I tried it, but it looked absolutely horrible.  I guess I didn't think it was anything special, as I never tried it again.  But I guess it's worth trying to see if it works for you.  
I wonder if some of the very high frequencies can end up phased out unintentionally . Different source material that depends on slight delay and effects could end up cancelling out before it reaches you . The highest frequencies are short enough to basically key together and also create peaks and harshness . Just a thought . Also i set my speakers straight when this topic came up just to try it . Sweet spot is kind of everywhere now . Way more enjoyable especially if listening from various spots within the room . The sound stage is enormous from my chair   
I agree that proper toe-in is something determined by experimentation.  This is the case with ALL aspects of proper speaker placement in a room.  

As for a reason for extreme toe-in, this can be useful if you are attempting to widen the area that one can sit in and still have some stereo imaging.  With extreme toe-in, the listener who is pushed closer to the left speaker will now be on-axis with the right speaker and off-axis with respect to the left speaker.  That means that the closer proximity to the left speaker is somewhat compensated for by the more direct sound of the right speaker.  The opposite holds true for the listener closer to the right speaker.
Toe-in is essentially an old school left-over idea from the 70s and 80s, prior to the whole concept of tweaks, especially room treatment. You know, before Tube Traps, Corner Tunes, tiny little bowl acoustic resonators, etc. and before speaker set-up tracks on Test CDs. When your room is firing on all cylinders the best toe-in is no toe-in. Also, speakers are generally placed too far apart. People! Hel-loo! That’s why a lot of folks must use toe-in - because the center of the stage disappears when the speakers Re too far apart.
ganash, you bet. I always listen with my eyes closed when I am listening seriously. Visual cues can easily distort your audio perception. Our eyes and ears are wired together. As an example if one inner ear malfunctions your eyes will start to drift to that side then your brain says "look ahead you idiot." the result is called nystagmus and what you get is vertigo. The room spins. 
Put on Herbie Hancock's Sextant album and play rain dance with the lights out or Weather Report's Mysterious Traveller and Nubian Sundance. 
When your room is firing on all cylinders the best toe-in is no toe-in.

This is so incredibly speaker dependent.
As for imaging, I find that toe-in affects whether musical instruments appear in a continuous range from speaker to speaker. If there is a missing center, or there is a L, C and R but nothing in between, toe-in can be the game changer.

And yes, too often speakers are too far apart for the room they are in.

Not just toe-in. Tilt-back.

My speaker soul-brother, Troels Gravesen, and I have both found that sometimes when we design a speaker to be listened with the ear at tweeter height the best place to listen ends up being a little between it and the mid-woofer. 

The point is, of course, to have the most enjoyable listening experience for you.

No reason not to experiment even when all the tools and tech says to listen another way.

I've been experimenting to dial in my Harbeth Super HL5 Plus speakers now that I have my custom stands.  Sitting flat on the stands, they lose some of the midrange and treble.  Tilting them back a bit really opened them up.  They sound best to me either facing straight ahead or with just a hair of toe-in.  I tried some more aggressive toe-in and didn't like it.
wow ¡!¡! i tried the more extreme toe-in position with the intersection about a foot in front of me face... i am liking it. deeper soundstage and the AirPulse Model 1 always had invisible midrange.. but now it is ghostly. soundstage width had decreased but I am slowly inching the speakers wider.. instrument positioning in the soundstage is improved as well...

update #1.. the monitors do not like to be too close to sidewalls.. 
It is simply wrong to state categorically that any particular degree of toe-in is optimal and anything else indicates some kind of problem.  All adjustments--placement, toe-in, rake angle (tilt back) are specific to a particular speaker model, room acoustics, placement of the speaker and listener, listener taste, etc.  

With most speakers and situations, the ideal degree of toe-in involves trade-offs.  With little or no toe-in, the soundstage will tend to be wide and open sounding, but, the center image might not be as strong and pin-point tight.  By increasing toe-in that center image will tend to get more pronounced, but, usually the trade-off is a woundstage that does not sound as expansive, or the listener does not feel quite as immersed in the soundfield.  The "right" toe-in is where the listener is most satisfied with the balance of trade-offs.