Frontal Horizontal Directivity Measurements of a Speaker ?..

0 Members and 1 Guest are viewing this topic. Read 4911 times.

Crossoverless

  • Restricted
  • Posts: 128
I believe that the better the frontal horizontal directivity measurements are , the closer your get to hearing LIVE MUSIC in your HOME! :violin: :drums:  :guitar:

 What do you all think:?:

 http://www.princeton.edu/3D3A/Directivity.html

rajacat

  • Full Member
  • Posts: 3239
  • Washington State
Re: Frontal Horizontal Directivity Measurements of a Speaker ?..
« Reply #1 on: 25 Jul 2015, 07:41 pm »
Here's the horizontal polar measurement of the SEOS 18 waveguide. They look very good to me. :)


steve f

  • Full Member
  • Posts: 682
Re: Frontal Horizontal Directivity Measurements of a Speaker ?..
« Reply #2 on: 26 Jul 2015, 03:42 am »
Not a bad plot. Some waistbanding.

rajacat

  • Full Member
  • Posts: 3239
  • Washington State
Re: Frontal Horizontal Directivity Measurements of a Speaker ?..
« Reply #3 on: 26 Jul 2015, 02:23 pm »

. Here are polars of the SEOS18, using a DE250 (driven with a series 100uF cap)
Horizontal:


rajacat

  • Full Member
  • Posts: 3239
  • Washington State
Re: Frontal Horizontal Directivity Measurements of a Speaker ?..
« Reply #4 on: 26 Jul 2015, 02:27 pm »
I believe that the better the frontal horizontal directivity measurements are , the closer your get to hearing LIVE MUSIC in your HOME! :violin: :drums:  :guitar:

 What do you all think:?:

 http://www.princeton.edu/3D3A/Directivity.html
I wonder how much smoothing was applied to those Princeton polars?

steve f

  • Full Member
  • Posts: 682
Re: Frontal Horizontal Directivity Measurements of a Speaker ?..
« Reply #5 on: 26 Jul 2015, 03:11 pm »
That's the trouble with charts. Unless you have all the details, you might be comparing apples to oranges.

A few comments. There is no question that planer type ESLs have good plots.

Horn/Lens type drivers can do pretty well themselves. The best I've seen are from H290C (Wayne Parham designed) there is a lot to be said for controlled directivity.

And a question. Has anyone seen similar plots from open baffle speakers? Just curious.

steve

Crossoverless

  • Restricted
  • Posts: 128
Re: Frontal Horizontal Directivity Measurements of a Speaker ?..
« Reply #6 on: 26 Jul 2015, 10:25 pm »
That's the trouble with charts. Unless you have all the details, you might be comparing apples to oranges.

A few comments. There is no question that planer type ESLs have good plots.

Horn/Lens type drivers can do pretty well themselves. The best I've seen are from H290C (Wayne Parham designed) there is a lot to be said for controlled directivity.

And a question. Has anyone seen similar plots from open baffle speakers? Just curious.

steve
Here's a Directivity plot from a open baffle speaker. And they also show what a ideal Directivity plot looks like too.

http://www.redspade-audio.blogspot.com/2011/02/understanding-directivity-plots.html

Crossoverless

  • Restricted
  • Posts: 128
Re: Frontal Horizontal Directivity Measurements of a Speaker ?..
« Reply #7 on: 27 Jul 2015, 05:40 pm »

JohnR

Re: Frontal Horizontal Directivity Measurements of a Speaker ?..
« Reply #8 on: 27 Jul 2015, 06:11 pm »
Um, that's your own thread....  :scratch:

AJinFLA

  • Industry Participant
  • Posts: 1114
  • Soundfield Audio Loudspeakers
    • Soundfield Audio
Re: Frontal Horizontal Directivity Measurements of a Speaker ?..
« Reply #9 on: 27 Jul 2015, 06:46 pm »
I believe that the better the frontal horizontal directivity measurements are , the closer your get to hearing LIVE MUSIC in your HOME!
You mean the about 10% of the live soundfield captured by stereo that you're reproducing?
Umm, I guess getting that frontal radiation part smooth, doesn't hurt.

cheers,

AJ

rajacat

  • Full Member
  • Posts: 3239
  • Washington State
Re: Frontal Horizontal Directivity Measurements of a Speaker ?..
« Reply #10 on: 27 Jul 2015, 09:01 pm »
That's the trouble with charts. Unless you have all the details, you might be comparing apples to oranges.

A few comments. There is no question that planer type ESLs have good plots.

Horn/Lens type drivers can do pretty well themselves. The best I've seen are from H290C (Wayne Parham designed) there is a lot to be said for controlled directivity.

And a question. Has anyone seen similar plots from open baffle speakers? Just curious.

steve
Steve, I agree, it's tough to compare unless you test at the same time under identical conditions.

To my eye the SEOS18 has better directivity than the H290c. Here's a H290c plot.
http://www.pispeakers.com/Measurements/H290C_Horizontal_Contour_Normalized_6dB_strata.jpg

If you look at the plots closely, the SEOS18 holds its pattern down to ~ 700hz whereas the H290c holds to ~1700hz. Also the H290c chart should be larger in order to examine more closely the detail such as waistbanding.
Here's another SEOS18 plot. This time it was measured under different conditions and the Denovo BA750 cd was used which is capable of going down to ~650hz.
Very good pattern control to ~650. It's virtually flat to ~1200 and then exhibits some minor wastebanding but still good down to ~650. The CD matches it well.


IMO the H290c is more equivalent to the SEOS12 but not quite as good. The H290 chart should be larger in order to show waistbanding and other issues more clearly. Here's a SEOS12 plot.



It holds pattern to `1100hz with some minor waistbanding and within 90 degrees.


steve f

  • Full Member
  • Posts: 682
Re: Frontal Horizontal Directivity Measurements of a Speaker ?..
« Reply #11 on: 28 Jul 2015, 12:15 am »
The SEOS does have good pattern. I still find the H290C superior. Especially for my application. Your mileage may vary.

JoshK

Re: Frontal Horizontal Directivity Measurements of a Speaker ?..
« Reply #12 on: 28 Jul 2015, 02:26 pm »
I still find the H290C superior.

In what way?

JoshK

Re: Frontal Horizontal Directivity Measurements of a Speaker ?..
« Reply #13 on: 28 Jul 2015, 02:31 pm »
That site in the first post is great.   I hadn't seen that one before.   I was a little surprised by the Gedlee Nathan.  IMO it doesn't look great.

The ESL plots are interesting and correlate with my experience as well.  They have the head in the vise narrow directivity but there is no denying they have smooth directivity.  I prefer controlled by modestly wider directivity such as with waveguides. 

poseidonsvoice

  • Full Member
  • Posts: 4016
  • Science is not a democracy - Earl Geddes
    • 2 channel/7 channel setup
Re: Frontal Horizontal Directivity Measurements of a Speaker ?..
« Reply #14 on: 28 Jul 2015, 02:37 pm »
That site in the first post is great.   I hadn't seen that one before.   I was a little surprised by the Gedlee Nathan.  IMO it doesn't look great.

The ESL plots are interesting and correlate with my experience as well.  They have the head in the vise narrow directivity but there is no denying they have smooth directivity.  I prefer controlled by modestly wider directivity such as with waveguides.

That Nathan was an older design and not refined, but in any case, Earl removed it from the line up because compared to the Abbey(c), NA12 and NS15, it was no contest. But do look at the power response of the Nathan, not bad.

The ESL plots clearly show a very narrow directivity like you said, you've got one spot and that's it; move your head, it's over. Also the pattern control is not as low as a waveguide at all, so the narrow directivity is only at high frequencies, I would want something from slightly less than 1000Hz up to 10khz of a flat high directivity index, *not* a rising directivity index. You'll see all the ESL's are rising.

Different strokes for different folks I guess.

Best,
Anand.

rajacat

  • Full Member
  • Posts: 3239
  • Washington State
Re: Frontal Horizontal Directivity Measurements of a Speaker ?..
« Reply #15 on: 28 Jul 2015, 05:00 pm »
You mean the about 10% of the live soundfield captured by stereo that you're reproducing?
Umm, I guess getting that frontal radiation part smooth, doesn't hurt.

cheers,

AJ
On the contrary, all of the information via the source is captured by the waveguide. Where else would it go?
Depth of soundfield, layering and ambience are included.

steve f

  • Full Member
  • Posts: 682
Re: Frontal Horizontal Directivity Measurements of a Speaker ?..
« Reply #16 on: 28 Jul 2015, 08:44 pm »
@JoshK,

The H290c has about 3-4 DB less ripple than the SEOS 12. It's also a physically deeper horn. I need that. The SEOS 12 is good, but the H290C is better. If I was going to design a controlled directivity speaker tomorrow, that's what I'd choose.

steve

DaveC113

  • Industry Contributor
  • Posts: 4341
  • ZenWaveAudio.com
Re: Frontal Horizontal Directivity Measurements of a Speaker ?..
« Reply #17 on: 28 Jul 2015, 08:59 pm »
I'll just re-post my response at WBF, it's a good topic...


Also take into account the CRC/Harman preference testing that shows a preference for smooth off axis frequency response, then look at the speakers tested in the Princeton study... some are very rough! Also, some assume from the Harman test data that a perfectly even frequency response is key but I don't believe this is the case and that preference is shown for a smooth response rather than a flat response. Not that a flat response isn't desirable, it's just not at the top of the priority list imo. I've heard amazing systems that don't have full extension at the frequency extremes and also some that do not have even FR but are very clear, coherent and sound amazing overall. Many people can't recognize deviations from flat FR without training.

As Tom said there is also a correlation between directivity and optimal room treatment, something that is rarely mentioned. For example Sanders shows his system at RMAF without room treatments but from a look at the measurements it's obvious how he gets decent results in a bare hotel room while other systems cannot. Personally, I think he's gone too far and the pattern is too narrow but there's plenty of room for personal preference. People who are used to hearing early 1st reflections might actually miss them despite the improvement it makes to get rid of them...


Quote
In commercial, sound polar information is much more common and required for room /sound system design and it is known that directivity is tied to intelligibility, the understanding random words.

In the home, reflected sound particularly from side wall reflections harms the delivery of a stereo image and so polar information can be an indicator of the ability to produce a mono phantom / solid stereo image as opposed to a "wall of sound" or in the worst case, obvious right and left sources.

The issue in each case is a preservation of time information and the tie in can be seen in the measurement which independent of language predicts intelligibility, the STIpa measurement.
That measurement is similar to that used in optics to show resolution and uses the audio equivalent of the optical Modulation Transfer Function, in fact the STIpa measurement uses 7 voice range MTF measurements. The optical version is easy to understand;

http://photo.net/learn/optics/mtf/

As one can see there, it is the degradation that happens when the "on and off" contrast is reduced and in the audio version it is when the off period is "filled in" with late arriving sound which can be reflected sound, noise or other signals which are not tied coherently to the direct sound.
That is why it is important for interaural crosstalk as well (the concern in the link), that late information contaminates the desired result.

But, for normal stereo, it is also very important as it effects how solid a mono phantom image is. In many recording studio's small monitors on the meter bridge are used, this is because the mix engineer now sits in the near field where the direct sound is much stronger than the wall reflections which arrive much later.

To be clear, there are other things which also effect the preservation of the stereo image like time coherency which most multi-way speakers do not do (all frequencies produced and arrive at the same time) but the directivty is the one that governs how much the room effects the results.

The cumulative effect can be easily heard playing (the same signal) a soft voice through both the right and left speakers. In a perfect system, the voice floats in front of you solidly as if a person were standing there speaking and you are not aware of the right and left sources.

At the opposite extreme where one has a large amount of incoherent sound arriving at ones ears, one plays the same signal and hears the right and left speakers as the obvious sources with a vague center phantom image which in some cases may extend the entire space between the right and left sources..

An extreme example of loss of time information is from the line arrays used in most concert venues, these are essentially incapable of producing a mono phantom image (or stereo image) because there is an arrival from each and every source arriving at ones ears but separated in time according to the path lengths for each. This is also why the voice intelligibility is so low with these systems, the same contamination of the "off" periods of the MTF is why.

On the other hand, a true coherent stereo image is so rare that most people only have a vague idea what it can be and think it's only the effect of sound coming from between the two speakers and have never heard a system that can produce a solid sense of voice location anywhere between the two.
This is why adding absorption on the walls to reduce the specular reflection is often judged to be a major improvement.

Directivity on the other hand reduces or eliminates the need for that absorption as less or much less destructive sound is projected to the sides etc.
Hope that helps.
Tom Danley
Director of R&D
Danley Sound Labs

rajacat

  • Full Member
  • Posts: 3239
  • Washington State
Re: Frontal Horizontal Directivity Measurements of a Speaker ?..
« Reply #18 on: 28 Jul 2015, 09:16 pm »
@JoshK,

The H290c has about 3-4 DB less ripple than the SEOS 12. It's also a physically deeper horn. I need that. The SEOS 12 is good, but the H290C is better. If I was going to design a controlled directivity speaker tomorrow, that's what I'd choose.

steve
This looks to be a rehashing of the SEOS12 v. H290c debate. :duh:  The only way this could be resolved is with an extended A/B listening sessions with neutral participants.
BTW AudioKinesis is now using the SEOS12 as its standard waveguide.

steve f

  • Full Member
  • Posts: 682
Re: Frontal Horizontal Directivity Measurements of a Speaker ?..
« Reply #19 on: 28 Jul 2015, 09:27 pm »
No not rehashing. I just like one better than another, was asked why, and answered the question. Nobody is obligated to go with my choice. As I stated before, your mileage may vary.

BTW I auditioned the AK at AXPONA, and posted that I liked it.

I have no financial or other interest in either product. Apparently my choice isn't acceptable to some people.
I resent the accusation.

steve