advertisement


The best 'speakers are the width of a human head - discuss

Robert

Robert

Tapehead
Something I came across recently in a John Atkinson interview with a 'speaker designer.
Specifically, the best sounding 'speakers - for human voice - are those where the front baffle matches that in width to the average human head.

The argument goes that your mouth sees your head as a baffle of around 7-8 inches, therefore a speaker system should approximate this in order to accurately reproduce a voice at the correct scale.

Interesting notion and not one I've seen before, so I'll just leave it there for now......
 
Alan Shaw* came out with this type of thing years ago, citing that the size of an LS3/5a was the size it was to be same size as a persons head.

(and same shape too if you’re a block head).

*Harbeth.
 
PhilofCas said:
Alan Shaw* came out with this type of thing years ago, citing that the size of an LS3/5a was the size it was to be same size as a persons head.
Click to expand...

I've had three pairs on Kans and now have Isobariks. They share exactly the same mid and treble driver but the presentation is quite different. Kans 'disappear' in a way Briks don't and vocals do sound more realistic and three-dimensional. In fact I would go as far as to say that apart from bass depth and overall scale, Kans are better than Isobariks in all other respects.

The cabs make a difference, you can't deny it.
 
Robert said:
Something I came across recently in a John Atkinson interview with a 'speaker designer.
Specifically, the best sounding 'speakers - for human voice - are those where the front baffle matches that in width to the average human head.
Click to expand...

I’m not buying that at all despite being a huge, huge fan of mini-monitors. Exhibit A: the Quad ESL. Once you’ve beaten that on voice we can discuss the idea!

PS I am still forming my crackpot theory as to why I like tiny speakers and point source speakers so much and I suspect it is phase/time coherence rather than anything else, or at least for the potential for it to be a hell of a lot better than big multi-driver speakers with drivers spaced apart on the vertical plane. I’m currently comparing LS3/5As and JR149s, two speakers that should be all but identical, but really aren’t, and my best guess is that comes down to phase too (I know the BBC spent an absolute fortune addressing it!).
 
Robert said:
Something I came across recently in a John Atkinson interview with a 'speaker designer.
Specifically, the best sounding 'speakers - for human voice - are those where the front baffle matches that in width to the average human head.

The argument goes that your mouth sees your head as a baffle of around 7-8 inches, therefore a speaker system should approximate this in order to accurately reproduce a voice at the correct scale.

Interesting notion and not one I've seen before, so I'll just leave it there for now......
Click to expand...

Then to reproduce a bass viol a speaker would have to be as wide as one? Or a piano? Or an orchestra?
 
Robert said:
Interesting notion and not one I've seen before, so I'll just leave it there for now......
Click to expand...

1unAGjs.jpg
 
Tony L said:
I’m currently comparing LS3/5As and JR149s, two speakers that should be all but identical, but really aren’t, and my best guess is that comes down to phase too (I know the BBC spent an absolute fortune addressing it!).
Click to expand...

Do you have a step response plot for either?

I think there are some more overt differences between LS3/5A and JR149 - the former has quite a bit less internal volume than the latter, and has more baffle-step compensation.
 
I would imagine that the best speaker for producing any band limited source is one that closely mimics surface/radiating area of the source. At least as far as scale goes. There's more to life than scale though.
 
Tony L said:
I’m not buying that at all despite being a huge, huge fan of mini-monitors. Exhibit A: the Quad ESL. Once you’ve beaten that on voice we can discuss the idea!
Click to expand...

They're not in this game as, despite having a large frontal area, they do not have a large reflective front cabinet area. Almost the entire frontal area of the Quads is very lightweight non-reflective driver.
 
Interesting observation here........
The speakers I heard and really disappear when music is playing are the LS3/5A and the likes as well as some very narrow Raidho speakers with multiple small white woofers..............all being about the width of my face !
I also had a similar sensation with some Sonus Faber floor standers and again, about the width of a human head.
I never had that feeling from a large 3 way speaker of any kind !
 
I once told a dealer, facetiously, that a speaker's mid-range driver needs to have a minimum diameter equivalent to the bell of a tenor saxophone. He took it seriously and ridiculed the idea. Henceforth I changed my mind; it was no longer a flippant comment but an absolute cornerstone of audio science and aesthetics.
 
  • Like
Reactions: awl
Robert said:
Something I came across recently in a John Atkinson interview with a 'speaker designer.
Specifically, the best sounding 'speakers - for human voice - are those where the front baffle matches that in width to the average human head.

The argument goes that your mouth sees your head as a baffle of around 7-8 inches, therefore a speaker system should approximate this in order to accurately reproduce a voice at the correct scale.

Interesting notion and not one I've seen before, so I'll just leave it there for now......
Click to expand...
I am not convinced by the simple argument. Producing sound is one thing but re-producing what you have captured seems to me to depend more on how you have captured the sound than how it was originally produced.
 
You must log in or register to reply here.


advertisement


Back
Top