Jump to content

The BIGGEST Change in Electrostat Speakers in years...


Recommended Posts



But if you are in the "sweet spot" why do you care if the directivity changes outside the sweet spot

 

In the sweet spot you still get all of the reflected sounds, these come from somewhere away from the direct axis and if they have uneven freq response at the different angles which are reflected (mostly the first order reflection points, side walls and rear wall are of interest)

Edited by hochopeper
Link to comment
Share on other sites

The goal is 75 deg wide dispersion @ -6 dB?

 

 

I also think that graph goes a bit too far on the LF side of things .... my understanding is that directivity gets less important once the ear+brain is using phase differences rather than level differences for sound localisation.

Link to comment
Share on other sites

But if you are in the "sweet spot" why do you care if the directivity changes outside the sweet spot

 

Firstly, the Sanders paper is wrong in almost every paragraph. It is a typical white paper.

 

Secondly, hochopeper beat me to my second point! Sanders tried to argue that ELS speakers don't produce much room interaction, so the direct sound in the centre is all that counts. Of course, cutting out room interaction brings us closer to anechoic sound, which is highly disliked and hence undesirable. Fact is, it's like Goldilocks' porridge: there can be too much or too little reflected sound. There is a 'just right' amount, from the 'just right' surfaces, in 'just right' delays, with 'just right' spectral balance. Omni's and flat panels both get it wrong.

  • Like 1
Link to comment
Share on other sites

But if you are in the "sweet spot" why do you care if the directivity changes outside the sweet spot

 

Reflections....  unless you are listening outside, sound bounces of the walls and gets in your ears.   The degree to which the reflected sound matches (in spectrum) the direct sound is particularly audible  (this is to do with how delayed sound is integrated by the ear when it is different in spectral balance)

 

 

Think of a sound occurring in a room  (eg.  smashing a glass with a hammer) ....  this radiates (relatively) equally in all directions

 

... now imagine the sound again, but in certain directions... or at certain frequencies ..... 4x more or less  (6dB) sound is radiated.     I'm sure you can easily imagine how this sounds "unnatural".

 

... but now if you look at an example panel speaker .... we can see that over an octave or so, the sound output at 45 degrees of axis  (prime reflection territory) than the sound balance differs by up to 12dB (or more?!) ... ie.   16 times less sound radiated from one octave to the next.

Link to comment
Share on other sites



  • Volunteer

Firstly, the Sanders paper is wrong in almost every paragraph. It is a typical white paper.

 

Secondly, hochopeper beat me to my second point! Sanders tried to argue that ELS speakers don't produce much room interaction, so the direct sound in the centre is all that counts. Of course, cutting out room interaction brings us closer to anechoic sound, which is highly disliked and hence undesirable. Fact is, it's like Goldilocks' porridge: there can be too much or too little reflected sound. There is a 'just right' amount, from the 'just right' surfaces, in 'just right' delays, with 'just right' spectral balance. Omni's and flat panels both get it wrong.

 

I'd be interested to know on which points he is wrong

 

As for room interactions, my understanding (and I may be wrong of course) is that you have them with all speakers. The benefit with very directional speakers is that the reflections are much lower in energy and therefore the brain is better at ignoring them

 

In any event, despite you guys telling me how bad my speakers (and other electrostats) must sound, in my opinion the best two systems I have ever heard at any price have been electrostats 

Edited by Sir Sanders Zingmore
  • Like 1
Link to comment
Share on other sites

I think the part that is often missed in this conversation about directivity goals is that the human ear has roughly three ways of dealing with reflections:

 

1st takes all early sounds and sort of combines them together ... so the first order reflections become part of what is perceived as part of the direct sound. (say up to 6-10ms I can't remember the exact numbers but 6 is generally the starting point as a goal for setting up a room)

 

2nd The slightly later reflections are perceived as the decay. (Dave/Newman do you have the numbers for this? I think it's roughly 80ms or so but that's from memory.)

 

3rd The really really late reflections are perceived as echos.

 

 

For an idea of the delay of reflections from each side wall, and front and rear walls have a look at this blog where the speaker/listener locations are drawn in and delay contours are drawn.

 

So Sanders is really only addressing point number one with his concerns around directivity and has missed the importance of the second and/or assumes relatively large rooms with optimal speaker/listener placement.

 

If you've got a room where the first order reflections are outside that ~6ms window you might avoid the 1st or at least limit it. For listener preference the second time period that I mention above should still be a relatively good match for the speaker's direct sound and preferably mostly side wall reflection (apparent source width). The 3rd really should be avoided (absorption).

Link to comment
Share on other sites

As for room interactions, my understanding (and I may be wrong of course) is that you have them with all speakers. The benefit with very directional speakers is that the reflections are much lower in energy and therefore the brain is better at ignoring them

 

The ear works with levels and time ... reducing the level may help in some situations, but the ability of the brain to filter that sound also depends on when it arrives. If you have some frequencies that are being rejected and others that are retained in the sound that you 'hear' then there will be some problem in the way the sound is percieved.

Link to comment
Share on other sites

I'd be interested to know on which points he is wrong

 

The generalisation that reflections are to be avoided via very narrow dispersion implies a direct sound to reflected sound ratio which is too low in the opinion of some.   The observation that a reflected sound to direct sound ratio which is >> 0 is preferred for good sound  (ie.  you want some reflections), implies that reflections shouldn't be considered as purely comb-filtering distortion.  

 

Also the frequency range where most panels 'beam' their sound  (ie.treble / 5khz +) is the area where it is found not to matter as much.    One reason for this is that sound reflects to poorly these frequencies, but there may be others.....   however, the frequencies where a typical panel might change it's directivity over  (say 500->3000)  are the ones where it's found to matter most (or at all).

 

If you think about a speaker playing a sound, VS that same sound naturally occurring ....  a major difference between the two events is that the speaker radiates unequally.   Different frequencies radiate more or less strongly in different directions....  whereas natural sounds are more omni-directional  (without us getting more complex and considering boundary effects).

 

 

 

Back to the theme.   Omni radiators can fix the changing directivity with frequency problem... but seem to have a too low overall directivity for most recording techniques and/or listener preference.

 

All speaker have compromises.  Most major.   So it seems mean to rip on panel speakers by holding them up to a fairly lofty goal.

Link to comment
Share on other sites



In any event, despite you guys telling me how bad my speakers (and other electrostats) must sound, in my opinion the best two systems I have ever heard at any price have been electrostats 

 

They have different compromises to most speakers.   They sound quite different.   They sound a lot closer in many ways to my ideal speakers (moderate, but controlled directivity) than most wide dispersion speaker  (especially ones with large lumps off axis, like many narrow box cone/dome)

Link to comment
Share on other sites

Firstly, the Sanders paper is wrong in almost every paragraph. It is a typical white paper.

 

Wow! Newman ... I didn't take you for a world-renowned hi-fi "expert"!  So where do you get any cred as a hifi 'oracle'?  :nana

 

Whereas Roger Sanders has made a career out of designing and producing products which punters buy.  That's his day job.  :P

 

 

Andy

Edited by andyr
Link to comment
Share on other sites



Gents

 

We have removed an image from this post, as it was "NSFW". The key is to link the images with a warning that it might be NSFW please. 

 

Alistair 

 

"NSFW"???

 

Edit:- or are you referring to that classic 'X' pic for, I think it was, Accustat?  'XXX' more like it!!  :lol:

 

Andy

Edited by andyr
Link to comment
Share on other sites

Gents

 

We have removed an image from this post, as it was "NSFW". The key is to link the images with a warning that it might be NSFW please. 

 

Alistair 

 

Please...SNA is not a work application...nor should it be used during work hours...to now have to put a warning on an image that is in open view in Facebook or in any newsagency...is political correctness gone mad & a good example of minority rule...heavy sigh...Rob

  • Like 1
Link to comment
Share on other sites



  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...
To Top