10 Audio Myths Debunked For Better Sound

[VoidX]

There is no research that I am aware of that shows a preference for toe-in.

I just did one quickly in free space, and the direct waves clearly show a better soundstage for toe-in. The speaker is an in-wall one (Aurea In-Wall 6+6), not made for being toed-in. The colors are: reddish - antinode, blue-ish - node, green - nominal gain. The greener the picture's tint, the better.

Simulation based on recordings in the room, parallel speakers on the left, toed-in speakers on the right:

The right image is way greener, this means the sound is much more uniform than without being toed-in. The corners are excluded from the simulation, only the +/-90 degrees from axis are processed, thus toe-in removes those parts from being simulated.

For correct time alignment, here's the same after Dirac Live:

Results are exactly the same, but perfectly symmetrical.

Personally, I find the changes made to the sound are usually subtle, and I have experimented with toe-in using many different speakers.

I also did, but the changes in balance were great, even in well-treated rooms. When the speakers were toed-in, the virtual center moved along with me. This happens when the angle is set so the gains on the line you walk perfectly mimic a fade - this is actually a THX studio requirement.

 

[3db]

Voidx maybe onto something here. Maybe toe in effects arent very audible for speakers with uniform off axes response but speakers with non uniform off axes response may see an improvement with some toe in. It would be a worthwhile endevour to study this. I wonder if Dr Toole is interested in shedding some light.

 

[shadyJ]

Sure, not everyone can be perfectly on axis across a couch. But if we're taking all these off axis measurements, and we know the SPL drops the further off your ears/mic are, isn't that a good reason to point the speakers toward the listeners?

Depends on the speakers' dispersion pattern, depends on the room, depends on the desired tonality that you want. There is no right answer here, just a preference.

 

[shadyJ]

I just did one quickly in free space, and the direct waves clearly show a better soundstage for toe-in. The speaker is an in-wall one (Aurea In-Wall 6+6), not made for being toed-in. The colors are: reddish - antinode, blue-ish - node, green - nominal gain. The greener the picture's tint, the better.

Simulation based on recordings in the room, parallel speakers on the left, toed-in speakers on the right:

View attachment 52745
The right image is way greener, this means the sound is much more uniform than without being toed-in. The corners are excluded from the simulation, only the +/-90 degrees from axis are processed, thus toe-in removes those parts from being simulated.

For correct time alignment, here's the same after Dirac Live:
View attachment 52746View attachment 52747
Results are exactly the same, but perfectly symmetrical.


I also did, but the changes in balance were great, even in well-treated rooms. When the speakers were toed-in, the virtual center moved along with me. This happens when the angle is set so the gains on the line you walk perfectly mimic a fade - this is actually a THX studio requirement.

I appreciate the effort here, but it seems too much of a simplification for real world applications. This looks like it is assuming a mono signal at a single frequency in a room that is perfectly symmetrical as well as perfectly reflective. Of course, that doesn't occur in natural listening situation. The nodes and antinodes would also shift per frequency, and the result would be some kind of comb-filterng, but the thing is comb-filtering isn't all that audible.

I would guess what matters more for toe-in is the tonal shift that occurs at that angle of the speaker's response, as well as the attenuation of the level of reflections that comes off the sidewalls.

 

[ryanosaur]

^^ Sometimes we see where Speakers have a slightly more even FR being even as little as 15º off-axis.

Personally, I had an absolutely abhorrent experience sitting directly on-axis to KEF Q and R series speakers.

So here we have an interesting example about sitting off-axis rather than on. And perhaps using Toe-In to "tune" the speakers so the FR can be at the best for the greatest number of seats.

There is always reason to experiment and be open to ALL options. In the end, we need to have the general understanding that what works in one room may not work in another... but that it's OK!
Even though my Speakers are recommended to be perpendicular to the Front Wall, they are also recommended to be ~15º off-axis at the LP.

How do we reconcile this? Flexibility!

 

[shadyJ]

Voidx maybe onto something here. Maybe toe in effects arent very audible for speakers with uniform off axes response but speakers with non uniform off axes response may see an improvement with some toe in. It would be a worthwhile endevour to study this. I wonder if Dr Toole is interested in shedding some light.

Off the top of my head, I don't think the dispersion pattern would matter much for the idea that he is trying to convey. Since the speakers would presumably have identical behavior, the level of comb-filtering that occur would have the same level in that scenario. The signals will always intersect with the same amount of amplitude.

 

[VoidX]

This looks like it is assuming a mono signal at a single frequency in a room that is perfectly symmetrical as well as perfectly reflective.

You're mostly right, but there is a but. Every signal can be decomposed into frequencies, and all theories/proofs on a single frequency can be expanded to all others. In this case, the signal was a single frequency, but actually not mono, this is why Dirac could correct it. Gain and phase differences produce different patterns, but with the same observation. The reason I choose a single frequency is not just what I wrote, but also because the local sum of multiple frequencies converge to a constant sound field as it approaches the Dirac-delta, which is the sum of all frequencies.

This theory is not that far in the scientific method, but seems promising. I would work on this simulator further, but I barely have the time in this exam season. I'll open source it as soon as I can, and someone can pick up where I left. Only walls and uniform particle propagation should be added, it's more than enough for a definitive proof. This should fix that both of us work from hearing only, which is not that reliable.

thing is comb-filtering isn't all that audible

It has a very easy proof: no object resonates on a single frequency, their frequency response is a bell. This is also true for the bones in your cochlea, and adding their response together will result in a smoothing. This is why we smooth curves after measurements, otherwise it's a huge comb we wouldn't even hear. But this is unrelated to the toe-in problem, as unlike the comb filter, it's not constant in macro scale.

I would guess what matters more for toe-in is the tonal shift that occurs at that angle of the speaker's response, as well as the attenuation of the level of reflections that comes off the sidewalls.

Tonal shift is a huge change, but I was talking about the case where both of them are EQd correctly. If we remove this detail, your point is perfectly valid, some speakers have better off-axis response. But correct EQ removes this difference, or could even transfer the off-axis response to on-axis if you'd like. The virtual center when you don't exactly sit between the speakers is what's destroyed without toe-in.

 

[VoidX]

The signals will always intersect with the same amount of amplitude.

Intersections can't be removed, but they can be faded. This is shown on the simulations, and when you add reflections, it would fade even more.

 

[lhunka]

lOVE jAMES REVIEWS. Love the debunking of audio myths from Audioholics. Best audio info around. I love James reviews as he talks about speakers. For me this is 90% of the equation. Mechanical transducers are impossible to get right - they are so totally flawed - poor frequency response, high distortion - the absolute bottom end of the audio chain (I will ignore turntables and cartridges for the purpose of this observation) Electronics - DACS, preamps, amps are so far superior in terms of frequency response, distortion compared to mechanical transducers. Can you really hear the difference of a cable or electronics when it is feeding such a terrible source? Don't get me wrong - I love speakers in spite of their flaws as I love listening to LP's as technically crappy as they may be.. In my younger days the goal was straight wire with gain. If that was the goal why would there be any preference? That obviously has never happened.
I am not meaning to denigrate electronics. I have a personal love for electrostatic speakers and I can assure you that electrostatics demand extremely capable amps. Gene reviewed his speakers with Pass and Krells. I understand that with his system but the Pass amps are some of the best on the planet and I am sure Gene would not disagree. However they were not the best for his system. Having said that few have that level of system for the vast majority of commercial speakers this is really a non issue.
So to open some other perhaps unpopular issues. The end stage reference is your personal hearing. Have you ever had a hearing test- are your ears any good? If you haven't done this the perhaps you are deluding yourself. I say this with mixed emotions. I have done multiple hearing tests and my hearing measures terrible. Is this the end of the world - no. I can readily hear the difference in my Sanders electrostatics and my Yamaha studio monitors It is humbling but .adds perspective.
The last observation is you have no control over the recording process. You have to listen to what the orignals provided. Some recordings are great (I listen to a lot of old jazz where recordings are terrible) and some not. It may seem counterintuitive but these bad recordings actually sound better over less revealing systems
Just random thoughts about listening and audio bs.

 

[buckchester]

I was thinking the exact same thing when I read that part of the article...

Me too...

 

[shadyJ]

Me too...

This article was completed long before Theo's review of the IsoAcoustics feet. It is not a response to that article. I appreciate the editorial freedom which Gene allows for both articles to be published at the same site. Different views are presented, and the reader is free to decide what the reality is for themselves.

 

[buckchester]

This article was completed long before Theo's review of the IsoAcoustics feet. It is not a response to that article. I appreciate the editorial freedom which Gene allows for both articles to be published at the same site. Different views are presented, and the reader is free to decide what the reality is for themselves.

And your thoughts on the product in question...?

 

[shadyJ]

And your thoughts on the product in question...?

My thoughts are stated in the article that is the subject of discussion of this thread.

 

[buckchester]

My thoughts are stated in the article that is the subject of discussion of this thread.

Noted. I'm not surprised you don't think they would contribute much, if any, improved sound to a quality loudspeaker. It would behoove that reviewer to conduct a follow-up blind test to see if his results will differ. I suspect they would.

 

[Ted R]

Thanks James and Gene for publishing this. It all seems to make sense and even though some of your content is pushing my limits of audio technical knowelege, I can still tell exactly where you are going with all of your comments which is awesome!