How to Choose a Loudspeaker -- What the Science Shows - Page 17 - AVS Forum | Home Theater Discussions And Reviews
Forum Jump: 
 5575Likes
Reply
 
Thread Tools
post #481 of 4354 Old 01-05-2019, 08:39 AM
Senior Member
 
youngho's Avatar
 
Join Date: Feb 2003
Posts: 421
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 17 Post(s)
Liked: 30
Quote:
Originally Posted by Hifisound View Post
I wish one would be able round up the best options in various types of speakers : 2-way waveguide ( M2, NS15, modded 4722N), multiways ( salon and maybe one more competitor), coax( Kef blade/reference), cardiod ( Dutch 8c), OB (Orion), maybe a Danley too !!! and a very cheap option like LSR series . And do DBT in Harman lab with AVS members. 😀
Thats some wish isn’t it 😀
It seems like some of these do occupy multiple categories, as the Dutch & Dutch 8C is also a waveguide design. I once thought it would be interesting to own a "stable" of different loudspeaker designs ("horses for courses"), maybe some classics and/or unique applications:
1. Freestanding, floor-standing cone/dome with narrower baffle and smaller waveguide: Revel,
2. Freestanding, floor-standing speaker with larger waveguide/horn: JBL, Geddes
3. "Bookshelf" (requiring stands) speaker with wider baffle: Stirling LS3/6 or JBL Pro LSR6332
4. Passive boundary-friendly (can positioned very near or up against wall) speakers: Roy Allison, Gradient
5. Active boundary-friendly speakers: Kii, Dutch & Dutch
6. Electrostatic: Quad, Magnepan
7. Dynamic OB: Orion or Spatial?
8. Omnipolar or at least extremely wide dispersion: Linkwitz Pluto, Epique CBT24
youngho is offline  
Sponsored Links
Advertisement
 
post #482 of 4354 Old 01-05-2019, 09:44 AM
AVS Forum Addicted Member
 
Ericglo's Avatar
 
Join Date: Oct 2004
Location: Just below the US in South Florida
Posts: 11,552
Mentioned: 19 Post(s)
Tagged: 0 Thread(s)
Quoted: 3444 Post(s)
Liked: 1894
Quote:
Originally Posted by audioguy View Post
Most of those folks moved to another forum (www/WhatsBestForum.com). And they still get in urinating contests with one another.
I see Sean Olive and Todd Welti have forums over there. That must be like yelling "bomb" in an airplane.

Having fun playing the new mobile game Volley Village
Ericglo is offline  
post #483 of 4354 Old 01-05-2019, 09:44 AM
AVS Forum Special Member
 
chikoo's Avatar
 
Join Date: Dec 2011
Posts: 1,213
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 87 Post(s)
Liked: 40
Quote:
Originally Posted by Hifisound View Post
I wish one would be able round up the best options in various types of speakers : 2-way waveguide ( M2, NS15, modded 4722N), multiways ( salon and maybe one more competitor), coax( Kef blade/reference), cardiod ( Dutch 8c), OB (Orion), maybe a Danley too !!! and a very cheap option like LSR series . And do DBT in Harman lab with AVS members. 😀

Thats some wish isn’t it 😀
I believe we are looking for the best sounding speaker, regardless of what drivers, or how it looks. Horn, kevlar, titanium, Aluminium, Ceramic, Exotic sandwiches should not matter.

Once the best speakers are determined, then it would be OK to reveal these artifacts.
chikoo is offline  
Sponsored Links
Advertisement
 
post #484 of 4354 Old 01-05-2019, 09:49 AM
AVS Forum Special Member
 
chikoo's Avatar
 
Join Date: Dec 2011
Posts: 1,213
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 87 Post(s)
Liked: 40
Quote:
Originally Posted by avkv View Post
Choosing a loudspeaker may be the biggest challenge for music and home theater lovers. There are countless brands from which to choose, and even more claims and counter-claims. Since the room has such a profound impact on the sound of a loudspeaker at lower frequencies, and it is impossible to listen in a blind test at an audio store, if they can find one, there is little that an audiophile can do to make a rational decision. Fortunately, science has come to the rescue with a set of measurements that have been proven to demonstrate an extremely close correlation with sound quality, as based on carefully controlled double-blind listening tests. This group of measurements have been adopted as the industry standard for measuring loudspeakers, as ANSI/CEA-2034-A. https://standards.cta.tech/apps/grou...project_id=165

Contradicting the oft-repeated claim that choosing a loudspeaker is a very personal choice, research has proven that regardless of age, culture, or listening experience, all people with nominally normal hearing generally agree on which speakers sound better than others. Indeed, there is a universal definition of what sounds good. http://www.aes.org/e-lib/browse.cfm?elib=12794 and https://secure.aes.org/forum/pubs/co...ns/?elib=12847

In this thread, we will publish the results of these measurements. In addition, we will discuss their correlation to double-blind listening tests, http://seanolive.blogspot.com/2008/1...udspeaker.html as well as publishing the results of formal listening tests, when available. We will add measurement results as they become available. The intention of this thread is for it to be reality-based, and to inform and discuss loudspeaker measurements and listening tests. The papers that really started it all are now available for free from the Audio Engineering Society here: http://www.aes.org/e-lib/browse.cfm?elib=5276 and here: http://www.aes.org/e-lib/browse.cfm?elib=5270
Is there a test where all the tones are fed all at once to the speaker? How did the speaker behave? Did it respond same as pushing in one tone at a time?

To me that is more important for when i am listening to music, the speaker is attempting to play multiple frequencies at the same time. I believe that understanding how it can handle more than a single frequency at any given point in time is where the secret sauce lies, and the reason why many not so good speakers are actually more pleasant to hear in real life.

Last edited by chikoo; 01-05-2019 at 09:52 AM.
chikoo is offline  
post #485 of 4354 Old 01-05-2019, 10:06 AM
AVS Forum Special Member
 
audioguy's Avatar
 
Join Date: Jun 2001
Location: Not far from Atlanta - but far enough!
Posts: 9,256
Mentioned: 92 Post(s)
Tagged: 0 Thread(s)
Quoted: 4718 Post(s)
Liked: 3500
Quote:
Originally Posted by Ericglo View Post
I see Sean Olive and Todd Welti have forums over there. That must be like yelling "bomb" in an airplane.
Except they bailed. Got treated poorly!! Science vs subjectivism anti-science !!
avkv and mik2h like this.
audioguy is offline  
post #486 of 4354 Old 01-05-2019, 10:17 AM
Senior Member
 
Hifisound's Avatar
 
Join Date: Nov 2012
Location: Bangalore, India
Posts: 383
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 124 Post(s)
Liked: 32
Quote:
Originally Posted by chikoo View Post
I believe we are looking for the best sounding speaker, regardless of what drivers, or how it looks. Horn, kevlar, titanium, Aluminium, Ceramic, Exotic sandwiches should not matter.

Once the best speakers are determined, then it would be OK to reveal these artifacts.
Hence the DBT. The candidates are very well measuring speakers using various approaches.

Hifisound is offline  
post #487 of 4354 Old 01-05-2019, 11:12 AM
Member
 
Floyd Toole's Avatar
 
Join Date: Jun 2017
Location: California
Posts: 755
Mentioned: 80 Post(s)
Tagged: 0 Thread(s)
Quoted: 781 Post(s)
Liked: 2853
Quote:
Originally Posted by chikoo View Post
Is there a test where all the tones are fed all at once to the speaker? How did the speaker behave? Did it respond same as pushing in one tone at a time?

To me that is more important for when i am listening to music, the speaker is attempting to play multiple frequencies at the same time. I believe that understanding how it can handle more than a single frequency at any given point in time is where the secret sauce lies, and the reason why many not so good speakers are actually more pleasant to hear in real life.
I see that you have asked this more than once, so let us get it behind us all. There is nothing complicated or mysterious here.

Loudspeakers can chew gum and walk at the same time - in other words it doesn't matter how many frequencies are in the signal. Music and voices are complex signals (many frequencies, ever changing), and all of the information is in the waveform (ever changing). The job of the loudspeaker is to reproduce the waveform - it is basically a linear input-output device reproducing the voltage waveform that is delivered by the power amplifier.

The shape of the waveform is completely described by the measured "transfer function": amplitude and phase vs. frequency (equivalent to the time-domain impulse response). Amplitude response is the dominant information because humans are almost completely unresponsive to phase shift.

As for testing with "all tones" at once, one of the old original methods of measurement was to drive the speaker with pink noise - i.e. all frequencies - and analyze the result with a real-time analyzer. Sweeping tones or quasi-random noise signals deliver the same results, but faster and with higher accuracy and frequency resolution.
sdurani, avkv, krabapple and 13 others like this.
Floyd Toole is offline  
post #488 of 4354 Old 01-05-2019, 05:53 PM
Newbie
 
Join Date: Aug 2018
Posts: 5
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 4 Post(s)
Liked: 1
I’ve enjoyed the science aspects of this thread. Thanks to Kevin and Dr. Toole for posting.

One question that has been asked in several round about ways, but not answered is: When is the response flat/accurate enough such that other aspects become the deciding factor between speakers?

It makes sense that flat and even off axis response is desirable. Probably any of the serious speaker manufactures could make something near perfect in this regard with an unlimited budget. Paul Barton has said as much. You can see the approaches and trade offs made by various companies. Revel uses the waveguidec, Kef the coaxial drivers, Vivid a 2” mid-tweet etc. If this is so important, why doesn’t Revel use the mid/tweeter in all speakers? Probably because of a cost trade off and at some cost point it is good enough.

Why is the Salon2 preferred over the appreatantly better measuring f228be? Distortion, bass extension, baffle diffraction? When do these become important over a sufficiently flat/even response?

PS. I’de like to seen the Spins of the f226be. Kevin, Do you have one?
SirPostaLittle is offline  
post #489 of 4354 Old 01-05-2019, 06:28 PM
Senior Member
 
youngho's Avatar
 
Join Date: Feb 2003
Posts: 421
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 17 Post(s)
Liked: 30
Quote:
Originally Posted by SirPostaLittle View Post
I’ve enjoyed the science aspects of this thread. Thanks to Kevin and Dr. Toole for posting.

One question that has been asked in several round about ways, but not answered is: When is the response flat/accurate enough such that other aspects become the deciding factor between speakers?
https://www.avsforum.com/forum/89-sp...l#post57363236

Quote:
Why is the Salon2 preferred over the appreatantly better measuring f228be? Distortion, bass extension, baffle diffraction? When do these become important over a sufficiently flat/even response?
https://www.avsforum.com/forum/89-sp...l#post57359530, also briefly discussed in Floyd Toole's book in section 12.1
youngho is offline  
post #490 of 4354 Old 01-05-2019, 06:43 PM
AVS Forum Special Member
 
chikoo's Avatar
 
Join Date: Dec 2011
Posts: 1,213
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 87 Post(s)
Liked: 40
Quote:
Originally Posted by Floyd Toole View Post
I see that you have asked this more than once, so let us get it behind us all. There is nothing complicated or mysterious here.

Loudspeakers can chew gum and walk at the same time - in other words it doesn't matter how many frequencies are in the signal. Music and voices are complex signals (many frequencies, ever changing), and all of the information is in the waveform (ever changing). The job of the loudspeaker is to reproduce the waveform - it is basically a linear input-output device reproducing the voltage waveform that is delivered by the power amplifier.

The shape of the waveform is completely described by the measured "transfer function": amplitude and phase vs. frequency (equivalent to the time-domain impulse response). Amplitude response is the dominant information because humans are almost completely unresponsive to phase shift.

As for testing with "all tones" at once, one of the old original methods of measurement was to drive the speaker with pink noise - i.e. all frequencies - and analyze the result with a real-time analyzer. Sweeping tones or quasi-random noise signals deliver the same results, but faster and with higher accuracy and frequency resolution.
Thank you for responding Dr Toole. Much appreciated!
chikoo is offline  
post #491 of 4354 Old 01-05-2019, 06:57 PM
AVS Forum Special Member
 
torii's Avatar
 
Join Date: Mar 2015
Posts: 6,873
Mentioned: 48 Post(s)
Tagged: 0 Thread(s)
Quoted: 3420 Post(s)
Liked: 1902
one aspect of the science that I have been appreciative of lately has been my study of my headphone collection of over 30 years...which equals about 30 pairs of headphones. with all the different response graphs posted by different reviewers/testors...I have heard some of the differences and can understand why flat can be desirable. I have 4 different sennheiser hd models where some might say they all sound the same as the graphs are only a db or 2 different...but listening close with headphones you can hear the differences really well. I think with loudspeakers most probably couldnt hear 1 or 2 db differences unless very experienced or trained.

Power: Marantz sr7008, NAD C 275Bee x 2, Video: Oppo 103, Samsung 75un6300
Speakers: Focal aria 948, Focal cc900, Klipsch synergy KSF 10.5, Magnepan LRS
Subs: Rythmik FV25HP, Rythmik FV15HP
torii is offline  
post #492 of 4354 Old 01-06-2019, 04:40 AM
AVS Forum Special Member
 
Join Date: Mar 2017
Posts: 1,340
Mentioned: 15 Post(s)
Tagged: 0 Thread(s)
Quoted: 912 Post(s)
Liked: 276
Quote:
Originally Posted by Floyd Toole View Post
I see that you have asked this more than once, so let us get it behind us all. There is nothing complicated or mysterious here.

Loudspeakers can chew gum and walk at the same time - in other words it doesn't matter how many frequencies are in the signal. Music and voices are complex signals (many frequencies, ever changing), and all of the information is in the waveform (ever changing). The job of the loudspeaker is to reproduce the waveform - it is basically a linear input-output device reproducing the voltage waveform that is delivered by the power amplifier.

The shape of the waveform is completely described by the measured "transfer function": amplitude and phase vs. frequency (equivalent to the time-domain impulse response). Amplitude response is the dominant information because humans are almost completely unresponsive to phase shift.

As for testing with "all tones" at once, one of the old original methods of measurement was to drive the speaker with pink noise - i.e. all frequencies - and analyze the result with a real-time analyzer. Sweeping tones or quasi-random noise signals deliver the same results, but faster and with higher accuracy and frequency resolution.
Thanks Dr. Toole. I was wondering about the same for quite some time.

Sent from my SAMSUNG-SM-G930A using Tapatalk

Last edited by SouthernCA; 01-06-2019 at 04:43 AM.
SouthernCA is offline  
post #493 of 4354 Old 01-06-2019, 09:22 AM
Member
 
Floyd Toole's Avatar
 
Join Date: Jun 2017
Location: California
Posts: 755
Mentioned: 80 Post(s)
Tagged: 0 Thread(s)
Quoted: 781 Post(s)
Liked: 2853
Quote:
Originally Posted by torii View Post
one aspect of the science that I have been appreciative of lately has been my study of my headphone collection of over 30 years...which equals about 30 pairs of headphones. with all the different response graphs posted by different reviewers/testors...I have heard some of the differences and can understand why flat can be desirable. I have 4 different sennheiser hd models where some might say they all sound the same as the graphs are only a db or 2 different...but listening close with headphones you can hear the differences really well. I think with loudspeakers most probably couldnt hear 1 or 2 db differences unless very experienced or trained.
Two interesting facts:
1. The highest rated headphones in double-blind tests deliver a frequency response to the eardrum that is very similar to that delivered by highly rated loudspeakers in a normally reflective room. There is an optimum target. Check out papers by Dr. Sean Olive and colleagues in the AES.
2. One would think (as I once did) that listening through headphones or in an anechoic chamber would be the most sensitive situation for hearing resonances and low-Q spectrum variations. Wrong! It turns out that humans use reflections (repetitions of the radiated sound) to accumulate more information about the sound and the audible thresholds for these spectral variations are significantly lower.
Toole, F. E. and Olive, S.E. (1988). “The modification of timbre by resonances: perception and measurement”, J. Audio Eng. Soc., 36, pp. 122-142.
A cynical part of my brain thinks that this is why we have tolerated so many crummy headphones over the years, while loudspeakers have made progress towards a describable goal.

The real life demonstration of this is to think about how a "concert in the park" sounds compared to the same concert in a venue - any venue. If it is an acoustical (unamplified) concert, one hopes for rain, to move it indoors. The timbre is much enriched inside because we hear more resonances - all voices and musical instruments are collections of resonances.
Vergiliusm, Muza, Gooddoc and 10 others like this.
Floyd Toole is offline  
post #494 of 4354 Old 01-06-2019, 11:38 AM
Advanced Member
 
Tomas2's Avatar
 
Join Date: Jan 2017
Location: New Orleans
Posts: 740
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 459 Post(s)
Liked: 716
Regarding evaluating a given speaker in mono is not optimal IMO, almost 100% of stereo productions have considerable amount of simple amplitude panning and XY micing techniques (exception being Mid-Side which has definite mono compatibility) and summing cancellations are unavoidable. Although i do understand the practicality of testing this way (mono) since all monitor pairs image differently ...thus making it practically impossible for controlled A/B comparisons.

Presently with speaker performance of the caliber of the Revel Salon 2 and JBL M2 i think it would be revealing to subjectively compare these two with professional grade ATC or Dynaudio monitors. Many of the current studios are now being designed utilizing ATC monitors ...integrated into clever front-to-back design philosophy:

http://www.northwardacoustics.com/about/

”FTB designs are very controlled acoustical environments, creating a quasi anechoic response speaker-to-room and speaker-to-engineer, yet maintaining a natural environment for the engineer to be within the room, doing so without compromising the speaker-to-engineer response. Thanks to the use of self-noises (noises produced by people moving and talking in the room) to trigger room environmental feedback created by surfaces invisible to the speakers, we can fully avoid having to use speaker-to-room feedback and Haas Kicker terminations like in older designs. In short, there are two rooms in one.

Lastly i thought might be worth mentioning the recent Dynaudio approach to speaker measurements here:

https://www.dynaudio.com/dynaudio-ac...demark-quality

SAMSUNG QLED | ROTEL | MOREL | M&K | HAFLER | TECHNICS SP-25
Tomas2 is offline  
post #495 of 4354 Old 01-06-2019, 11:40 AM
AVS Forum Special Member
 
torii's Avatar
 
Join Date: Mar 2015
Posts: 6,873
Mentioned: 48 Post(s)
Tagged: 0 Thread(s)
Quoted: 3420 Post(s)
Liked: 1902
science wise I think in ear monitors should be what everyone in the field should be working on. their ability to provide amazing sound, their portability and small size, and the sonics they can provide is just amazing....some of them have 10 freakin drivers inside just one.... they seal up your ear canal(like a room) and those tiny drivers bounce sound everywhere inside your ear. its hard spending 2 grand plus for some tiny iem's but its the future.


https://www.64audio.com/product/1964...r-Monitor___62

Power: Marantz sr7008, NAD C 275Bee x 2, Video: Oppo 103, Samsung 75un6300
Speakers: Focal aria 948, Focal cc900, Klipsch synergy KSF 10.5, Magnepan LRS
Subs: Rythmik FV25HP, Rythmik FV15HP

Last edited by torii; 01-06-2019 at 12:06 PM.
torii is offline  
post #496 of 4354 Old 01-06-2019, 12:58 PM
Member
 
Floyd Toole's Avatar
 
Join Date: Jun 2017
Location: California
Posts: 755
Mentioned: 80 Post(s)
Tagged: 0 Thread(s)
Quoted: 781 Post(s)
Liked: 2853
Quote:
Originally Posted by torii View Post
science wise I think in ear monitors should be what everyone in the field should be working on. their ability to provide amazing sound, their portability and small size, and the sonics they can provide is just amazing....some of them have 10 freakin drivers inside just one.... they seal up your ear canal(like a room) and those tiny drivers bounce sound everywhere inside your ear. its hard spending 2 grand plus for some tiny iem's but its the future.


https://www.64audio.com/product/1964...r-Monitor___62
In-ear monitors are really insert earphones ruggedized to play loud enough to overcome leakage from on-stage sounds. That explains the multiple drivers - it has nothing to do with bouncing sound around the earcanals which cannot happen because of the size of the earcanal compared to the wavelengths of the sounds. There is no requirement for them to be "accurate" because they are usually equalized to deliver the spectrum individual musicians need.
avkv, SouthernCA and mik2h like this.
Floyd Toole is offline  
post #497 of 4354 Old 01-06-2019, 01:23 PM
Member
 
Join Date: Sep 2018
Location: Central, PA
Posts: 160
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 90 Post(s)
Liked: 172
[QUOTE=torii;57387986]science wise I think in ear monitors should be what everyone in the field should be working on. their ability to provide amazing sound, their portability and small size, and the sonics they can provide is just amazing....some of them have 10 freakin drivers inside just one.... they seal up your ear canal(like a room) and those tiny drivers bounce sound everywhere inside your ear. its hard spending 2 grand plus for some tiny iem's but its the future.

As a professional audio engineer and former touring musician, I have used numerous IEM (in ear monitors) over the years. More is not necessarily better. The "ears" I've settled on use a single dynamic driver as opposed to multiple balanced armature style drivers. The work extremely well for my live performances, and I do reference my mixes on them occasionally, but there a are a variety of reasons that they won't (and shouldn't) replace speakers as a primary monitoring system for mixing / mastering work. That's a topic for a different thread as it's outside the scope of this one, which has frequently strayed off topic already.



Additionally, kudos to Dr. Floyd for participating in these forums. The information he presented here is not only compelling, but has practical applications for those purchasing speakers, as well as those interested in learning more about the scientific aspects of the audio field. There are likely not many in similar positions that would do so.
CruelInventions likes this.
garygreyh is online now  
post #498 of 4354 Old 01-06-2019, 01:23 PM
AVS Forum Special Member
 
torii's Avatar
 
Join Date: Mar 2015
Posts: 6,873
Mentioned: 48 Post(s)
Tagged: 0 Thread(s)
Quoted: 3420 Post(s)
Liked: 1902
I guess bouncing around the wrong word. all the science probably already done by the hearing aid makers...at least insurance might pay for those

Power: Marantz sr7008, NAD C 275Bee x 2, Video: Oppo 103, Samsung 75un6300
Speakers: Focal aria 948, Focal cc900, Klipsch synergy KSF 10.5, Magnepan LRS
Subs: Rythmik FV25HP, Rythmik FV15HP

Last edited by torii; 01-06-2019 at 01:43 PM.
torii is offline  
post #499 of 4354 Old 01-06-2019, 01:59 PM
AVS Forum Addicted Member
 
Ericglo's Avatar
 
Join Date: Oct 2004
Location: Just below the US in South Florida
Posts: 11,552
Mentioned: 19 Post(s)
Tagged: 0 Thread(s)
Quoted: 3444 Post(s)
Liked: 1894
Quote:
Originally Posted by Tomas2 View Post
Regarding evaluating a given speaker in mono is not optimal IMO, almost 100% of stereo productions have considerable amount of simple amplitude panning and XY micing techniques (exception being Mid-Side which has definite mono compatibility) and summing cancellations are unavoidable. Although i do understand the practicality of testing this way (mono) since all monitor pairs image differently ...thus making it practically impossible for controlled A/B comparisons.

Presently with speaker performance of the caliber of the Revel Salon 2 and JBL M2 i think it would be revealing to subjectively compare these two with professional grade ATC or Dynaudio monitors. Many of the current studios are now being designed utilizing ATC monitors ...integrated into clever front-to-back design philosophy:

http://www.northwardacoustics.com/about/

”FTB designs are very controlled acoustical environments, creating a quasi anechoic response speaker-to-room and speaker-to-engineer, yet maintaining a natural environment for the engineer to be within the room, doing so without compromising the speaker-to-engineer response. Thanks to the use of self-noises (noises produced by people moving and talking in the room) to trigger room environmental feedback created by surfaces invisible to the speakers, we can fully avoid having to use speaker-to-room feedback and Haas Kicker terminations like in older designs. In short, there are two rooms in one.

Lastly i thought might be worth mentioning the recent Dynaudio approach to speaker measurements here:

https://www.dynaudio.com/dynaudio-ac...demark-quality

How is having a big room any different than doing the measurements outside?

Quote:


Instead, we’re able to simulate the good effects of an anechoic chamber inside Jupiter without having to deal with the undesirable ones thanks to its sheer size. Simply put, we shut off the microphones between their measuring the impulse sound (an instantaneous ‘click’ that contains all frequencies) and the reflection coming back off the walls. It’s split-second stuff, but it effectively takes the room out of the equation.

Having fun playing the new mobile game Volley Village
Ericglo is offline  
post #500 of 4354 Old 01-06-2019, 03:24 PM
Member
 
Floyd Toole's Avatar
 
Join Date: Jun 2017
Location: California
Posts: 755
Mentioned: 80 Post(s)
Tagged: 0 Thread(s)
Quoted: 781 Post(s)
Liked: 2853
Quote:
Originally Posted by Ericglo View Post
How is having a big room any different than doing the measurements outside?
Interesting. This is the technique pioneered by KEF in the 80's when they used their innovative digital impulse testing - a large room with the reflections gated out. A large space is necessary to get reasonable frequency resolution down to a usefully low frequency, especially if the measurement distance is the necessary 2 m or more. The lowest frequency and the frequency resolution are the inverse of the reflection-free time window. The time-windowed FFT has been around for decades, even among DIYers and useful results at mid and high frequencies are possible in domestic sized spaces. Pro audio people have long used large hangers, gymnasiums, etc., and outdoors, for measurements of large systems needing measuring distances of 7m or more.

Anechoic chambers can be calibrated to low frequencies, offering high resolution at low frequencies, are very quiet, and permit measurement of non-linear distortion. The big room time gated method is different, not inherently better.
avkv, Ericglo, unretarded and 1 others like this.
Floyd Toole is offline  
post #501 of 4354 Old 01-07-2019, 12:38 AM
Advanced Member
 
TimVG's Avatar
 
Join Date: Apr 2011
Posts: 544
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 263 Post(s)
Liked: 399
Quote:
Originally Posted by Tomas2 View Post
Regarding evaluating a given speaker in mono is not optimal IMO, almost 100% of stereo productions have considerable amount of simple amplitude panning and XY micing techniques (exception being Mid-Side which has definite mono compatibility) and summing cancellations are unavoidable. Although i do understand the practicality of testing this way (mono) since all monitor pairs image differently ...thus making it practically impossible for controlled A/B comparisons.
I believe the tests are performed with one of the L/R channels to avoid summing cancellations. Not being familiar with the program material used in the MLL, I wonder what would happen if one would compare two neutral reproducers (let's say the Salon2, and perhaps the JBL 4365) in mono, with close mic'ed recordings featuring material close(ish) to the loudspeakers' own directivity index (e.g. a human voice would be closer to the Salon2, and a trombone would be closer to the 4365). Also see chapter 10.8 in Sound Reproduction 'the directivities of common sound sources'. I recall reading some comments from John Schuermann how the M2 excelled on close mic'ed recordings, and am wondering if there is a correlation. Naturally program material will vary a lot, and most, if not all, program material will have some 'ambiance' added (natural or artificial) - which is perhaps why in blind tests the cone/dome systems seem preferred. My preference appears to be narrower dispersion systems - perhaps as an active classical musician I am simply used to hearing many of these instruments from a much closer perspective than most.

I would expect differences in 'imaging' to be directly related to the DI as well, since imaging, apart from what is in the recording, would be dependent on the reflected sounds.

Last edited by TimVG; 01-07-2019 at 12:45 AM.
TimVG is offline  
post #502 of 4354 Old 01-07-2019, 11:15 AM
AVS Forum Addicted Member
 
Ericglo's Avatar
 
Join Date: Oct 2004
Location: Just below the US in South Florida
Posts: 11,552
Mentioned: 19 Post(s)
Tagged: 0 Thread(s)
Quoted: 3444 Post(s)
Liked: 1894
Quote:
Originally Posted by Floyd Toole View Post
The article says: "D.E.L. Shorter of the BBC was the first to accurately measure and identify sources of driver and cabinet resonances in the late 1950’s".

For me this whole piece is a reminder of times past. Shorter was the principal peer reviewer of my 1985 & 1986 JAES papers. The journal editors had trouble finding someone who had the background. He was a nice, smart, man and we carried on a mail correspondence for a time afterwards. The BBC efforts in measuring the time domain performance were indeed pioneering, but much of it unnecessary. As we now know, transducers are minimum-phase devices over their operating frequency ranges. This simply means that the time domain information is in the amplitude response. Waterfalls add no new information. My papers showed that when the anechoic amplitude response is flat and smooth there are no resonances - which, not surprisingly, is why listeners prefer such loudspeakers in double-blind comparison tests. If there are residual resonances, we now know the audible thresholds so they can be attenuated by the necessary amounts, allowing more cost effective designs. We also know that, because they are minimum-phase phenomena they are treatable by minimum-phase equalization - a huge advantage of active loudspeakers.

I visited the BBC labs and Harwood toured me around. I was surprised to see and hear the typical broadcast control rooms that their monitor loudspeakers were designed to perform in. Many were small, acoustically very dead spaces. Direct sound would be dominant; off-axis performance cannot have mattered much. Figure 18.5(f) shows the BBC LS5/8 which was not bad on axis, but not great off axis. It did not perform well in normally reflective rooms. This trait found its way into some other British loudspeakers, such as the KEF 105.2 shown in Figure 5.4. It was interesting that about the time the BBC acoustical research activities were fading out, my own work was nicely underway.

Thanks for the memory . . .

Did you read the whole article or just the clip? I haven't read this site in probably 15 years or more and I decided to reread some of the info.

It brought to mind some questions. What do you consider the most significant advancement(s) in speakers over the last twenty years? What research has been most interesting, significant, impactful?

Also, what areas of research do you consider important, interesting, etc going forward?

Having fun playing the new mobile game Volley Village
Ericglo is offline  
post #503 of 4354 Old 01-07-2019, 01:36 PM
Member
 
Floyd Toole's Avatar
 
Join Date: Jun 2017
Location: California
Posts: 755
Mentioned: 80 Post(s)
Tagged: 0 Thread(s)
Quoted: 781 Post(s)
Liked: 2853
Ericglo said: "Did you read the whole article or just the clip?"

What a lot of words! An interesting survey of technologies over the years, with lots of opinions on the audible consequences but a serious lack of (a) trustworthy measurements and (b) unbiased listening evaluations. He would learn from the measurements in my book, especially Chapter 18. He seems to think we hear waveforms.

"What do you consider the most significant advancement(s) in speakers over the last twenty years?"

Check out the figure attached to my post #50 and I think it is obvious that the 30 year progress has been in the refinement of transducer performance combined with better, computer aided, system design. Computer modeling figures powerfully in all of this. Good results require a faith in the existing science, not golden ears.

"What research has been most interesting, significant, impactful?"

At the risk of tooting my own horn, I would like to think that the most consequential research has involved meaningful double-blind listening tests, combined with accurate and comprehensive anechoic data on loudspeakers. The correlation between the two domains is strong. Few - almost nobody - in the industry has had the facilities, time, money, etc. to do the necessary research that has now nailed down the basic performance targets for loudspeakers that listeners reward with high sound quality scores. The Canadian government and then Harman International invested in the problem and the results are published for all to see The problem is that such measurements data are so scarce. See the Figure attached to my post #220 .

"what areas of research do you consider important, interesting, etc going forward?"

The source of greatest variation in what we hear from the best loudspeakers is in program material, the circle of confusion is real. We need psychological research into why so many people - professionals and consumers - are science deniers. Too many choose instead to follow their own sighted, biased, opinions of the moment, or the stylish writings of subjective reviewers who do nothing defensible in their subjective commentary and, for the most part, do no useful measurements. If we made absolutely no progress in loudspeaker design beyond where we are now, that alone would elevate the standard of sound quality everywhere.

A second line of research would be to explore methods of delivering more credible spatial illusions from two channels, which it seems we are stuck with for life. Flexible multichannel upmixing directed by metadata in recordings may be a practical route.
RichB, avkv, Muza and 3 others like this.
Floyd Toole is offline  
post #504 of 4354 Old 01-07-2019, 02:42 PM
AVS Forum Special Member
 
Bill-99's Avatar
 
Join Date: Oct 2003
Location: Columbus, OH
Posts: 1,511
Mentioned: 30 Post(s)
Tagged: 0 Thread(s)
Quoted: 831 Post(s)
Liked: 517
"We need psychological research into why so many people - professionals and consumers - are science deniers."

Spot on and not limited to just loudspeakers!

Just one more upgrade and things will be perfect.
Bill-99 is offline  
post #505 of 4354 Old 01-07-2019, 03:11 PM
AVS Forum Special Member
 
Steve Goff's Avatar
 
Join Date: Aug 1999
Location: Olympia, WA, USA
Posts: 2,034
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 448 Post(s)
Liked: 260
Quote:
Originally Posted by avkv View Post
Jakeshields,

Both KEF and TAD speakers suffer from Intermodulation Distortion since the woofer or mid cones are the waveguides. While it is true that their off-axis radiation it symmetrical, the design also limits dispersion at the highest frequencies, which is heard as a lack of "air."
Comparing the off-axis response of the Salon 2 and the KEF Reference 5 as shown by Stereophile’s measurements, I’m not sure that I would agree the KEF limits dispersion more than the Revel at high frequencies. I might note, though, that the KEF speaker has a smoother transition off axis. Maybe the KEF would lack “air” in a direct comparison, but I have not had a chance to compare the two.

But my question is about intermodulation distortion. Have you performed any tests showing that the intermodulation distortion on the current crop of KEF speakers (or earlier ones with similar drivers and crossover regions) is audible to such an extent that it compromises the accuracy of the speakers? Most of these speakers cross over between the woofer and midrange at about 400 Hz. I understand and agree with your conclusion that the spinorama measurements correlate well with listening tests. Have you performed listening tests that show a similar relationship between listener preference and intermodulation distortion?




Sent from my iPad using Tapatalk

Steve Goff
Steve Goff is offline  
post #506 of 4354 Old 01-07-2019, 04:30 PM
Member
 
Floyd Toole's Avatar
 
Join Date: Jun 2017
Location: California
Posts: 755
Mentioned: 80 Post(s)
Tagged: 0 Thread(s)
Quoted: 781 Post(s)
Liked: 2853
Quote:
Originally Posted by Steve Goff View Post
Comparing the off-axis response of the Salon 2 and the KEF Reference 5 as shown by Stereophile’s measurements, I’m not sure that I would agree the KEF limits dispersion more than the Revel at high frequencies. I might note, though, that the KEF speaker has a smoother transition off axis. Maybe the KEF would lack “air” in a direct comparison, but I have not had a chance to compare the two.

But my question is about intermodulation distortion. Have you performed any tests showing that the intermodulation distortion on the current crop of KEF speakers (or earlier ones with similar drivers and crossover regions) is audible to such an extent that it compromises the accuracy of the speakers? Most of these speakers cross over between the woofer and midrange at about 400 Hz. I understand and agree with your conclusion that the spinorama measurements correlate well with listening tests. Have you performed listening tests that show a similar relationship between listener preference and intermodulation distortion?




Sent from my iPad using Tapatalk
To see a spinorama on a KEF Reference 5 go to: http://www.kef.com/uploads/files/THE...ath_200514.pdf
Look at section 4.1.5 to see that the spinorama curves pivot around 80 Hz to show a progressive off-axis falloff. Why they modified the display to pivot at 80 Hz is puzzling because they should all be the same below about 100 Hz, but that would make the forward directional bias look even worse. Perhaps someone can explain that. The curves are smooth, though, which is good.

Both harmonic and IM distortion show up occasionally - very rarely - in double-blind listening tests. It did with a concentric model, as presented in a paper to the Acoustical Society of America LA Chapter by Mark Glazer. When linear data cannot explain a result, one looks deeper into the non-linear realm.

As I point out in my book, and in these forums, conventional harmonic and intermodulation distortion measurements do not correlate well with audibility in music. But when high enough, it is heard.
avkv and mik2h like this.
Floyd Toole is offline  
post #507 of 4354 Old 01-07-2019, 04:47 PM
AVS Forum Special Member
 
aarons915's Avatar
 
Join Date: Oct 2012
Location: Cincinnati, OH
Posts: 1,047
Mentioned: 10 Post(s)
Tagged: 0 Thread(s)
Quoted: 668 Post(s)
Liked: 599
Quote:
Originally Posted by Steve Goff View Post
Comparing the off-axis response of the Salon 2 and the KEF Reference 5 as shown by Stereophile’s measurements, I’m not sure that I would agree the KEF limits dispersion more than the Revel at high frequencies. I might note, though, that the KEF speaker has a smoother transition off axis. Maybe the KEF would lack “air” in a direct comparison, but I have not had a chance to compare the two.

But my question is about intermodulation distortion. Have you performed any tests showing that the intermodulation distortion on the current crop of KEF speakers (or earlier ones with similar drivers and crossover regions) is audible to such an extent that it compromises the accuracy of the speakers? Most of these speakers cross over between the woofer and midrange at about 400 Hz. I understand and agree with your conclusion that the spinorama measurements correlate well with listening tests. Have you performed listening tests that show a similar relationship between listener preference and intermodulation distortion?




Sent from my iPad using Tapatalk
If you search the Revel thread for intermodulation distortion or IMD you'll find some discussion on this. I challenged someone awhile back who brought it up and took my own measurements in room with REW where I had it simulate 2 tones, 1 in the bass and 1 at the crossover frequency to see how much was detected in either the tweeter or the midwoofer. I had my volume 20 decibels louder than I would ever have it in practice and the most I could get was less than 4% IMD. I'm sure I didn't do the test properly but I wasn't able to find the standard IMD test, I'm not sure there is one. I have 2 way KEFs but to your point, none of the 3 ways have that problem because they are crossed over from 300-500Hz, depending on the model. Also, I would go as far as to say as long as 2 way Kefs are used with a high pass filter(I use a 90Hz 4th order) and not played at ridiculous volumes, IMD isn't an issue either.
sdurani likes this.
aarons915 is offline  
post #508 of 4354 Old 01-07-2019, 05:27 PM
Member
 
Floyd Toole's Avatar
 
Join Date: Jun 2017
Location: California
Posts: 755
Mentioned: 80 Post(s)
Tagged: 0 Thread(s)
Quoted: 781 Post(s)
Liked: 2853
Quote:
Originally Posted by aarons915 View Post
If you search the Revel thread for intermodulation distortion or IMD you'll find some discussion on this. I challenged someone awhile back who brought it up and took my own measurements in room with REW where I had it simulate 2 tones, 1 in the bass and 1 at the crossover frequency to see how much was detected in either the tweeter or the midwoofer. I had my volume 20 decibels louder than I would ever have it in practice and the most I could get was less than 4% IMD. I'm sure I didn't do the test properly but I wasn't able to find the standard IMD test, I'm not sure there is one. I have 2 way KEFs but to your point, none of the 3 ways have that problem because they are crossed over from 300-500Hz, depending on the model. Also, I would go as far as to say as long as 2 way Kefs are used with a high pass filter(I use a 90Hz 4th order) and not played at ridiculous volumes, IMD isn't an issue either.
Yes a great many two-way speakers would have been really great if they were 3-ways . If the woofer/mid is small enough to transition half decently to a tweeter, it will have inadequate low bass and/or cannot play loud - add a few subs. But as sound level needs rise, one quickly enters 3-way territory, which can deliver more sound and with better directivity characteristics. Salon2s are 4 ways. Not cheap, but they sound good and can play indecently loud without audible (to my ears) stress. IMHO there needs to be a progressive reduction in diaphragm size as frequency increases. That costs money.
avkv and aarons915 like this.
Floyd Toole is offline  
post #509 of 4354 Old 01-07-2019, 05:47 PM
AVS Forum Special Member
 
unretarded's Avatar
 
Join Date: Nov 2016
Location: Ventura Ca
Posts: 4,077
Mentioned: 70 Post(s)
Tagged: 0 Thread(s)
Quoted: 1687 Post(s)
Liked: 2074
Quote:
Originally Posted by Floyd Toole View Post
IMHO there needs to be a progressive reduction in diaphragm size as frequency increases. That costs money.

That's been my humble DIY findings as well on the lower end of the spectrum is where I encountered the problem.



Crossing from a 18 inch sub to 5.25 main...…..that's a bridge too far... Crossing from the 18 inch sub to a 15 inch main was a much better match in ways I can not explain properly and I would be sure to mess up the terminology only making it more confusion.


I think the proper terms to describe some of it are particle velocity and tactile energy...…..


Any input on that would be greatly appreciated, I know there is something there, but the science and terminology eludes me....thanks.

Link to Stereo Integrity SI HT 18 sub build......https://www.avsforum.com/forum/155-di...-pedestal.html
Speakers and subs for sale...https://www.avsforum.com/forum/209-au...kers-subs.html
unretarded is offline  
post #510 of 4354 Old 01-07-2019, 06:35 PM
Member
 
Floyd Toole's Avatar
 
Join Date: Jun 2017
Location: California
Posts: 755
Mentioned: 80 Post(s)
Tagged: 0 Thread(s)
Quoted: 781 Post(s)
Liked: 2853
Quote:
Originally Posted by unretarded View Post
That's been my humble DIY findings as well on the lower end of the spectrum is where I encountered the problem.



Crossing from a 18 inch sub to 5.25 main...…..that's a bridge too far... Crossing from the 18 inch sub to a 15 inch main was a much better match in ways I can not explain properly and I would be sure to mess up the terminology only making it more confusion.


I think the proper terms to describe some of it are particle velocity and tactile energy...…..


Any input on that would be greatly appreciated, I know there is something there, but the science and terminology eludes me....thanks.
At low frequencies it is volume velocity not particle velocity. What it the volume of air displaced by a diaphragm per unit movement. 18" woofers are great for cinemas, but are overkill for homes - as I describe in Chapter 8 multiple smaller subs are more efficient and sound better for more listeners. Nevertheless, transitioning any sub at 80 Hz or so to a 5.25 inch main is definitely a stretch - at low sound levels maybe, but not for movies or lots of modern music.
avkv and unretarded like this.
Floyd Toole is offline  
Sponsored Links
Advertisement
 
Reply Speakers

Tags
cea 2034 , double-blind , listening tests , loudspeaker measurements , spinorama

Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off