How to Choose a Loudspeaker -- What the Science Shows - Page 120 - AVS Forum | Home Theater Discussions And Reviews
Forum Jump: 
 6384Likes
Reply
 
Thread Tools
post #3571 of 5320 Old 07-10-2019, 02:06 PM
AVS Forum Special Member
 
Scotth3886's Avatar
 
Join Date: Jan 2016
Location: New Albany, OH
Posts: 8,229
Mentioned: 35 Post(s)
Tagged: 0 Thread(s)
Quoted: 3997 Post(s)
Liked: 2464
Quote:
Originally Posted by Red MC View Post
Yep. Ever listen to a binaural recording made with a dummy head using good headphones? They can give you the illusion of 3D imaging in a space way beyond the confines of your head in all directions. And some QSound albums also do a relatively decent job of it with stereo speakers, placing sound sources well off to the side of the listening position, or above your head, and even behind you to a limited extent. QSound and similar 3D post-processing effects don't rely on room reflections, and in my experience, they are better heard in a dry acoustic environment where early reflections are absorbed.



Some speakers, when set up properly in the right room, are able to produce a reasonably decent soundstage even when you're sitting outside of the sweet spot, well off-axis. The best example I've heard was an MBL show demo where the little 121 stand-mounts were set up in a large L-shaped hotel suite with plenty of room around them. They generated a large soundstage, and you could get up and walk around and the position of the performers in the soundstage wouldn't shift around all that much. Omnis in general seem to be good at this. Most small stand-mounted speakers are good at it too, if they have enough room around them.

On the other hand, there are speakers with a very narrow sweet spot, outside of which there is little or no soundstage. Probably the best examples are large flat electrostats e.g. Sanders and King Sound. They image great when you're right on the centerline, but if you move off of it the soundstage is gone. Martin Logans are a little bit like this, and most horn speakers too.
So based on my own anecdotal experience, there seems to be an inverse correlation between the size of the sweet spot (over which a reasonable soundstage is maintained) and the directivity of the loudspeaker.



I agree you can't infer anything about time domain behavior from a spinorama.

But I don't follow your point about time differentials. The time of arrival of the direct wave is a function of placement only. Differences in the arrival times from multiple speakers can be calculated from their distances to the listening position, and they won't change if you swap different speakers into the same locations. Assuming pair matching is good, and it better be, then measuring two speakers doesn't give you any more information than measuring one.

I personally haven't experienced a correlation between time domain response and imaging. At one point, I was attracted by the idea of time coincident and phase coherent speakers, because some people swear they image better. But I haven't experienced that. For example, I haven't heard a Thiel or Vandersteen speaker image quite as well as the current KEFs which put the midrange and tweeter in opposite polarity.
"The best example I've heard was an MBL show demo where the little 121 stand-mounts were set up in a large L-shaped hotel suite with plenty of room around them."

Jolida room at AXPONA? If so, yes, really fantastic imagining, and IMO, much better than the far larger Extremes. The 121s didn't do 'scale' well at all, but the rest, great.

Last edited by Scotth3886; 07-10-2019 at 02:31 PM.
Scotth3886 is offline  
Sponsored Links
Advertisement
 
post #3572 of 5320 Old 07-10-2019, 02:08 PM
AVS Forum Special Member
 
Soulburner's Avatar
 
Join Date: Oct 2007
Location: Lincoln
Posts: 4,892
Mentioned: 46 Post(s)
Tagged: 0 Thread(s)
Quoted: 194 Post(s)
Liked: 1629
How about this. We choose a room to measure as a baseline. The room must resemble that of a typical home. You measure various models of speakers that have spinorama data using a two-mic setup on an artificial human head, changing only the speakers themselves. Everything else remains the same. Each speaker also gets ratings from listeners regarding imaging.

You have now recorded the speakers and the room, and human experiences. Now, subtract the data that is consistent in all the measurements (the room effects), and you're left with only that which is different between the speakers. You should then be able to correlate aspects of the speakers with what causes imaging differences.

I'm not sure we need to go to these lengths, but it would be interesting if we did.
Scotth3886 likes this.

HT: Samsung PN64H5000 (recommended settings) | NAD T758 V3 | Buchardt S400 (2) | Emotiva E2 (2) | Rythmik Audio F12 (2)
Soulburner is offline  
post #3573 of 5320 Old 07-10-2019, 02:23 PM
Member
 
Floyd Toole's Avatar
 
Join Date: Jun 2017
Location: California
Posts: 836
Mentioned: 89 Post(s)
Tagged: 0 Thread(s)
Quoted: 875 Post(s)
Liked: 3078
Quote:
Originally Posted by SouthernCA View Post
Very interesting. What were the results/findings of this survey?
Ah-h-h-h, so you expect me to put 11 pages of my book (Section 7.4.2) or 57 pages (Chapter 7) into this forum . . . that is about 1/10 of the nearly 500 page book.

Sorry, but it just might be time for you to spend some money. If it is any interest to you, writing technical books is a serious money loser. My earnings buy me some decent wine now and then - that's all - and each book took more than a year to write, not counting years of accumulating and reading papers by other researchers. Fortunately, I consider all of that more than acceptable as a summary of a lifetime of audio research.

Cheers
Floyd Toole is offline  
Sponsored Links
Advertisement
 
post #3574 of 5320 Old 07-10-2019, 02:30 PM
AVS Forum Special Member
 
Join Date: Dec 2016
Location: Las Vegas, NV
Posts: 1,605
Mentioned: 41 Post(s)
Tagged: 0 Thread(s)
Quoted: 1048 Post(s)
Liked: 1244
Telarc Engineer Jack Renner dead at 84

Sad news. Interesting interview in the article.

https://www.prosoundnetwork.com/reco...T=905118524236
DonH50 likes this.
Rex Anderson is offline  
post #3575 of 5320 Old 07-10-2019, 02:38 PM
AVS Forum Special Member
 
Scotth3886's Avatar
 
Join Date: Jan 2016
Location: New Albany, OH
Posts: 8,229
Mentioned: 35 Post(s)
Tagged: 0 Thread(s)
Quoted: 3997 Post(s)
Liked: 2464
Quote:
Originally Posted by Rex Anderson View Post
Sad news. Interesting interview in the article.

https://www.prosoundnetwork.com/reco...T=905118524236

From the same little tiny SE Ohio town that my dad was born in.

Scotth3886 is offline  
post #3576 of 5320 Old 07-10-2019, 02:45 PM
AVS Forum Club Gold
 
craig john's Avatar
 
Join Date: Oct 2003
Location: Lancaster, PA
Posts: 11,799
Mentioned: 65 Post(s)
Tagged: 0 Thread(s)
Quoted: 1339 Post(s)
Liked: 1577
Quote:
Originally Posted by sdurani View Post
What would you measure? Not asking for a technical answer, just a description of what aspects you think contribute to imaging. As I mentioned in my earlier post, manufacturers can measure consistency to see HOW identical two speakers are. What else do you think should be measured?

I don't know how to measure something that only exists in the listener's brain, and not as a hard physical entity. Nonetheless, I don't think it has ever been shown that having two speakers be identical in the frequency domain necessarily guarantees they will image correctly. At least I'm not aware of any evidence the shows a causal relationship. Are you aware of any such studies? While it seems logical that it would take two identical signals, (dual-mono and identical in terms of arrival times and levels), to ensure that the imaging is as precise as possible, I think there is more to it than that. The dispersion characteristics of the speakers will impact how much reflected sound impinges on the first arriving sound, which can have a detrimental impact on precise imaging. I think there's a possibility that cabinet diffractions could cause some arrival time disparities that could affect imaging. The alignment and placement of the drivers can also impact arrival times. The crossover could have some latency that affects arrival times. There's likely many other things that could affect this, many of which would not have an impact on the Spinorama measurements.



It's should be possible to measure how differently speakers image by using "hard" speakers to produce sounds in different locations, followed by simulated phantom images that are designed to image from the same places as the hard speakers were occupying. Then one would need to develop a measurement system that could simulate the way the human brain integrates the sonic information it receives to form the psycho-acoustic image. As I described in an earlier post, I think it has to entail 3 dimensional measurements taken in the time domain with dual mics, possibly like those in a binaural dummy head. If the speaker pair gets the placement of the psychoacoustic image in the same 3D location as the hard speaker, it could be said to image well. If the phantom image doesn't come close to the hard speaker, whether that be in size, location, scale or whatever, it could be said to image poorly. (Alternatively, one could use subjective listening tests, but I think that would be less reliable and repeatable than a 3-D measurement system. I realize that training the listeners can improve the consistency of the results, at least according to Harman, but I have yet to see that research repeated and verified by outside researchers. Until that happens, I'm not a believer... and I have taken the training.) Once speakers are identified as good or bad imagers, then maybe one could look for patterns in the types of speakers that image well, and also in types of speakers that don't image well. That could lead to investigations of what actually CAUSES speaker pairs to image well or not well. And THAT is what we really want to know.



Let me reverse the question to you. If you were designing an experiment to determine what causes speakers to image, and then testing speakers for imaging preference, how would you design such an experiment?


Craig

Lombardi said it:
"Perfection is not attainable, but if we chase perfection we can catch excellence."

My System
craig john is offline  
post #3577 of 5320 Old 07-10-2019, 02:47 PM
AVS Forum Special Member
 
krabapple's Avatar
 
Join Date: Apr 2003
Location: in a state bordered by Kentucky and Maine
Posts: 5,779
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 485 Post(s)
Liked: 469
Quote:
Originally Posted by Rex Anderson View Post
Very interesting. I just set up 4 Revel F208's and a C208 using a Lexicon MC-10. I'm also listening to all 2 channel music in surround using DTS Neural:x. It sounds the best of the upmix options to me. I agree with your description and analysis. I was at Harman last week and got to hear a lot of systems there, at Kevin Voeck's home and Floyd Toole's home. They were using Auro 3D in the SDP-75. Unfortunately, the Lexicon does not have that. I was able to try some options in the Auro 3D control software, it is very powerful and gives you many adjustable parameters. It would take some time to dial it in and I think you would probably find different settings work better for different songs or styles of music (i.e. classical vs rock).
In my room* and to my ears**, Dolby Pro Logic II (music mode) upmixing has typically sounded better than DTS Neural X (or any DTS upmixing flavor).

Unfortunately, Dolby has all but retired DPLII and replaced it with Dolby Surround Upmixer (DSU), which is, to my ears, less good. Would love to try a DBT to test this preference, though!

And I wish I could find Auro3D on a non budget-busting device.


(*Audyssey MultiXT32 correction below 350Hz, dual subs, treated room, matched loudspeakers all around, blah blah)

(**rather worn with age )
krabapple is offline  
post #3578 of 5320 Old 07-10-2019, 02:57 PM
AVS Forum Special Member
 
krabapple's Avatar
 
Join Date: Apr 2003
Location: in a state bordered by Kentucky and Maine
Posts: 5,779
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 485 Post(s)
Liked: 469
Quote:
Originally Posted by highmr View Post
I agree with that and that is how I listen to most music. The setting is a little bit obscure in Dolby Surround, so not all users may be aware of it.
DSU only has one user-configurable setting, AFAICT: 'Center Spread', with two options -- on and off.
krabapple is offline  
post #3579 of 5320 Old 07-10-2019, 02:58 PM
AVS Forum Special Member
 
Join Date: Mar 2017
Posts: 1,401
Mentioned: 15 Post(s)
Tagged: 0 Thread(s)
Quoted: 956 Post(s)
Liked: 288
Quote:
Originally Posted by Floyd Toole View Post
Ah-h-h-h, so you expect me to put 11 pages of my book (Section 7.4.2) or 57 pages (Chapter 7) into this forum . . . that is about 1/10 of the nearly 500 page book.



Sorry, but it just might be time for you to spend some money. If it is any interest to you, writing technical books is a serious money loser. My earnings buy me some decent wine now and then - that's all - and each book took more than a year to write, not counting years of accumulating and reading papers by other researchers. Fortunately, I consider all of that more than acceptable as a summary of a lifetime of audio research.



Cheers
I have your book. Did not realize that it was in the book. I will go look. Thanks.
motrek likes this.
SouthernCA is offline  
post #3580 of 5320 Old 07-10-2019, 03:04 PM
Member
 
Join Date: Sep 2018
Location: Central, PA
Posts: 170
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 98 Post(s)
Liked: 196
Quote:
Originally Posted by Floyd Toole View Post
Ah-h-h-h, so you expect me to put 11 pages of my book (Section 7.4.2) or 57 pages (Chapter 7) into this forum . . . that is about 1/10 of the nearly 500 page book.

Sorry, but it just might be time for you to spend some money. If it is any interest to you, writing technical books is a serious money loser. My earnings buy me some decent wine now and then - that's all - and each book took more than a year to write, not counting years of accumulating and reading papers by other researchers. Fortunately, I consider all of that more than acceptable as a summary of a lifetime of audio research.

Cheers

It's a conspiracy. We all got together and decided that if we ask enough questions, you'll eventually have provided your entire life's work to us across hundreds of posts, which we can then copy and paste into a free copy of your book. You caught on. We've been exposed. You'd think that a bunch of people that spend way to much time and money on stereo equipment would pony up for your book, but dilithium crystal coated speaker cables are pricey. Have to pinch pennies where we can. LOL
DonH50, motrek, duckymomo and 4 others like this.
garygreyh is offline  
post #3581 of 5320 Old 07-10-2019, 03:13 PM
Advanced Member
 
motrek's Avatar
 
Join Date: Dec 2013
Posts: 653
Mentioned: 7 Post(s)
Tagged: 0 Thread(s)
Quoted: 518 Post(s)
Liked: 191
Quote:
Originally Posted by craig john View Post
I don't know how to measure something that only exists in the listener's brain, ...
Please stop with this "listener's brain" nonsense.

Yes, sure, a human brain can process auditory cues to localize sounds.

What if the human then describes or points to where he thinks the sounds are occurring? Then that imaging (or at least part of it) exists "outside of his brain" as you would say.

Also, there's nothing that prevents a computer program from doing the same processing. Would you then say that imaging exists in the brain of the computer program? What does that even mean?
motrek is offline  
post #3582 of 5320 Old 07-10-2019, 03:13 PM
AVS Forum Special Member
 
krabapple's Avatar
 
Join Date: Apr 2003
Location: in a state bordered by Kentucky and Maine
Posts: 5,779
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 485 Post(s)
Liked: 469
Re: imaging and soundstage, I find that the amount of toe-in (which aims the axis of direct sound) makes a large difference. Unsurprisingly. Point being, comparing speakers for their ability to 'image' would require removing that as a variable.
krabapple is offline  
post #3583 of 5320 Old 07-10-2019, 03:19 PM
AVS Forum Special Member
 
torii's Avatar
 
Join Date: Mar 2015
Posts: 7,242
Mentioned: 54 Post(s)
Tagged: 0 Thread(s)
Quoted: 3592 Post(s)
Liked: 2027
Im sure the science can explain it all, unfortunately speakers mainly sound and image different. This leads to people buying speakers they like. science can say this speaker measures poorly, but that doesnt mean people wont love them.

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 online now  
post #3584 of 5320 Old 07-10-2019, 03:20 PM
Senior Member
 
Red MC's Avatar
 
Join Date: May 2016
Posts: 374
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 273 Post(s)
Liked: 219
Quote:
Originally Posted by Scotth3886 View Post
"The best example I've heard was an MBL show demo where the little 121 stand-mounts were set up in a large L-shaped hotel suite with plenty of room around them."

Jolida room at AXPONA? If so, yes, really fantastic imagining, and IMO, much better than the far larger Extremes. The 121s didn't do 'scale' well at all, but the rest, great.
No, it was Sydney 2012. They were sharing a larger suite with a Linn system, IIRC. On the first day, they were playing them in a cramped near field setup while the Linn system got top billing. They couldn't have been more than 2m apart and were practically up against a wall. Nevertheless, they caught my ear. When I came back the next day, they had swapped the systems and the MBLs were out in the room where they really impressed me. A bit bright though.

HT: Dynaudio C2, Contour S CX, 2x BM14S, Aperion surrounds, Simaudio Titan, Marantz AV8801, Oppo 103, Linn Majik DS, and a Pioneer Kuro
Stereo: Dynaudio Focus 160, Simaudio W-5 LE or Luxman M-600A, Linn Akurate DSM, 2x Rythmik F12G
Other interests: motorcycling, skiing, being active
Red MC is offline  
post #3585 of 5320 Old 07-10-2019, 03:28 PM
Oppo Beta Group
 
RichB's Avatar
 
Join Date: Nov 1999
Location: Massachusetts
Posts: 11,186
Mentioned: 33 Post(s)
Tagged: 0 Thread(s)
Quoted: 1967 Post(s)
Liked: 1431
Quote:
Originally Posted by SouthernCA View Post
"You cannot blame a scientific method and statistical model for failing to measure up to your subjective "imaging" and "soundstage" evaluation."

You are making it sound like subjective assessment of imaging and soundstage (quotes removed) is any less than subjective assessment of sound quality itself in double blind testing.
I quoted these terms because they lack a precise definition, we all think we know what they mean but may not be on the same page.

@craig john discusses the phantom image where instruments and vocals appear to have precise placement.
Others, may be referring to the apparent size of the soundstage, which may be an opposing concept.

My definition is in line with @R Harkness and @craig john where the instruments and vocals appear to have a position in the room.

When I want to listen to the phantom image, I sit on my coffee table about 6 feet from each Salon2.
With the proper source material, the sound appear to have depth and often from above.
This effect diminishes with distance.

Overall, the Salon2's generate a large sound that is not pinpointed to each driver, which is my preference.

Spinorama dispersion characteristics should have some correlation to phantom image and large soundstage but may be inversely related.

- Rich

Oppo UPD-205 x 2 | UPD-203 | Sonica DAC | Emotiva XMC-1 (v3) | Revel Salon2s, Voice2, Studio2s | Benchmark AHB2 x 4 | ATI AT522NC | Velodyne HGS-15 | LG 77C9 | Lumagen 2020 | HDFury Vertex x 2

Last edited by RichB; 07-10-2019 at 03:32 PM.
RichB is offline  
post #3586 of 5320 Old 07-10-2019, 03:30 PM
AVS Forum Special Member
 
Join Date: Mar 2017
Posts: 1,401
Mentioned: 15 Post(s)
Tagged: 0 Thread(s)
Quoted: 956 Post(s)
Liked: 288
Quote:
Originally Posted by RichB View Post
I quoted these terms because they lack a precise definition, we all think we know what they mean but may not be on the same page.


@craig john discusses the phantom image where instruments and vocals appear to have precise placement.
Others, may be referring to the apparent size of the soundstage, which may be an opposing concept.


My definition is in line with @R Harkness ; and @craig john where the instruments and vocals appear to have a position in the room.


When I want to listen to the phantom image, I sit on my coffee table about 6 feet from each Salon2.
With the proper source material, the sound appear to have depth and often from above.
This effect diminishes with distance.


Overall, the Salon2's generate a large sound that is not pinpointed to each driver, which is my preference.


I would think that the dispersion characteristics should have some correlation to phantom image and large soundstage which may be inversely related.


- Rich
I now understand what you meant and agree.
craig john likes this.
SouthernCA is offline  
post #3587 of 5320 Old 07-10-2019, 03:37 PM
AVS Forum Special Member
 
Scotth3886's Avatar
 
Join Date: Jan 2016
Location: New Albany, OH
Posts: 8,229
Mentioned: 35 Post(s)
Tagged: 0 Thread(s)
Quoted: 3997 Post(s)
Liked: 2464
Quote:
Originally Posted by craig john View Post
I don't know how to measure something that only exists in the listener's brain, and not as a hard physical entity. Nonetheless, I don't think it has ever been shown that having two speakers be identical in the frequency domain necessarily guarantees they will image correctly. At least I'm not aware of any evidence the shows a causal relationship. Are you aware of any such studies? While it seems logical that it would take two identical signals, (dual-mono and identical in terms of arrival times and levels), to ensure that the imaging is as precise as possible, I think there is more to it than that. The dispersion characteristics of the speakers will impact how much reflected sound impinges on the first arriving sound, which can have a detrimental impact on precise imaging. I think there's a possibility that cabinet diffractions could cause some arrival time disparities that could affect imaging. The alignment and placement of the drivers can also impact arrival times. The crossover could have some latency that affects arrival times. There's likely many other things that could affect this, many of which would not have an impact on the Spinorama measurements.



It's should be possible to measure how differently speakers image by using "hard" speakers to produce sounds in different locations, followed by simulated phantom images that are designed to image from the same places as the hard speakers were occupying. Then one would need to develop a measurement system that could simulate the way the human brain integrates the sonic information it receives to form the psycho-acoustic image. As I described in an earlier post, I think it has to entail 3 dimensional measurements taken in the time domain with dual mics, possibly like those in a binaural dummy head. If the speaker pair gets the placement of the psychoacoustic image in the same 3D location as the hard speaker, it could be said to image well. If the phantom image doesn't come close to the hard speaker, whether that be in size, location, scale or whatever, it could be said to image poorly. (Alternatively, one could use subjective listening tests, but I think that would be less reliable and repeatable than a 3-D measurement system. I realize that training the listeners can improve the consistency of the results, at least according to Harman, but I have yet to see that research repeated and verified by outside researchers. Until that happens, I'm not a believer... and I have taken the training.) Once speakers are identified as good or bad imagers, then maybe one could look for patterns in the types of speakers that image well, and also in types of speakers that don't image well. That could lead to investigations of what actually CAUSES speaker pairs to image well or not well. And THAT is what we really want to know.



Let me reverse the question to you. If you were designing an experiment to determine what causes speakers to image, and then testing speakers for imaging preference, how would you design such an experiment?


Craig

And why a given manufacture's top of the line big speakers generally image much worse their smaller speakers although their overall soundstage is larger? Many examples of this from the hot mess the Everest 67000s are versus even the lowly, although not cheap, new L100s which I thought were quite good. The original L100s replaced my Quads back in 70 or 71 and didn't image at all. The new revised ones are in the same shape and size box, but are far far better. What changed? The big MBL Extremes versus the much smaller 121s. The Von Schweikert Ultra 11s versus the 9s. The Fartin Logan Neoliths, which I've heard many many times and only heard them do their stuff once, versus the smaller narrower panel models. The little super cheap new Magnepan LRSs versus the 3.7i, which I passed buying because they had such a struggle with a bloated center image on Jennifer Warnes, Hunter, among others. I screwed with those for several hours trying to get them placed correctly and I couldn't do it. So that's just off of the top of my head, but there are many others if I sat around and thought about it for a while. You think its the same design team so what gives that they, IMO (of course), blow it the big ones? Or maybe I'm just brain damaged (yeah yeah, some truths are self-evident, got it) from my very early exposure (1947) to so much live unamplified music? It doesn't hold for everything though. Get me on the topic of pitch and I'm lost.
Scotth3886 is offline  
post #3588 of 5320 Old 07-10-2019, 03:43 PM
Member
 
Jon AA's Avatar
 
Join Date: Apr 2019
Location: WA State
Posts: 195
Mentioned: 9 Post(s)
Tagged: 0 Thread(s)
Quoted: 119 Post(s)
Liked: 128
Quote:
Originally Posted by craig john View Post
Imaging doesn't actually exist as a physical phenomenon in space and time. Imaging only exists in your brain.....
Thanks, but I would hope anybody following the conversation this far already knows this stuff.

Quote:
Now look at the Spinorama. I don't see ANY information about arrival times contained in the FR measurements. Frequency Response measurements, (also known as Magnitude Response measurements), only describe intensity differences of speakers at different frequencies. There is no time response information in the measurements. More importantly, there is definitely no time differential response information between the arrival times of TWO or MORE speakers... in the FR measurements of a single speaker.
Of course not. The Spinoramas are for evaluating speakers, not speaker setup and electronics. What part of what you described above has anything to do with speaker design (assuming a reasonably well designed speakers with good time alignment on the drivers, etc)?

Quote:
Setup the system for ideal central imaging, using a recording that has a strong central image, (female voice recordings are good for this.) Listen to the voice and ensure it is phantom imaged from directly in front of you. Then change the distance setting of one speaker vs. the other. The central image will move side to side based on whichever speaker has an earlier arrival time.)
Yes, been doing that for years. I currently have two calibrations for my setup--one calibrated to a central listening position for multiple listeners, one calibrated to my seat, which is several feet laterally offset. While it could be done manually, in my experience Audyssey has an uncanny ability to "get it right" for such a setup--measurements appear accurate within a fraction of an inch to properly set the time delay and intensities (channel level) match as accurately as I can measure.



Just put the 1st mic measurement location exactly where your head will be and bam--a perfectly centered, crisp, phantom image is the result. So much so you need to walk up to the center speaker and put your ear to it in order to verify it is indeed, shut off because you'd swear that's where the vocals are coming from when in stereo mode. All when sitting several feet off center.



But none of any of that has anything to do with speaker design so I don't understand why you're harping on speaker measurements not incorporating something that has nothing to do with speaker measurements.



Yes, different speakers will sound different and at first I thought you were trying to hone in on the minutia of the differences between different designs...but then you describe things that have nothing to do with speaker design--hopefully none of us are buying speaker sets that have different volumes and delays built into the speakers themselves. That's all controlled by setup/electronics.
motrek likes this.
Jon AA is offline  
post #3589 of 5320 Old 07-10-2019, 03:51 PM
AVS Forum Special Member
 
Scotth3886's Avatar
 
Join Date: Jan 2016
Location: New Albany, OH
Posts: 8,229
Mentioned: 35 Post(s)
Tagged: 0 Thread(s)
Quoted: 3997 Post(s)
Liked: 2464
Quote:
Originally Posted by RichB View Post
I quoted these terms because they lack a precise definition, we all think we know what they mean but may not be on the same page.

@craig john discusses the phantom image where instruments and vocals appear to have precise placement.
Others, may be referring to the apparent size of the soundstage, which may be an opposing concept.

My definition is in line with @R Harkness and @craig john where the instruments and vocals appear to have a position in the room.

When I want to listen to the phantom image, I sit on my coffee table about 6 feet from each Salon2.
With the proper source material, the sound appear to have depth and often from above.
This effect diminishes with distance.

Overall, the Salon2's generate a large sound that is not pinpointed to each driver, which is my preference.

Spinorama dispersion characteristics should have some correlation to phantom image and large soundstage but may be inversely related.

- Rich

I look at it more as the image within the soundstage / soundscape and the hope that objects with the image will scale appropriately, i.e., vocalist's mouth versus the size of the double bass or the Imperial Grand, etc. Conflict is that to get an appropriately similar 'surface loudness' as the real life Imperial Grand or double bass requires a large radiating area, which can crap up other aspects. The super huge, full range, crossoverless Soundlabs A1s come to mind, BUT due to the size of the panel, they were in the way of the image. They were/are a 'lights off' speaker or the panel is going to block that bassist, percussionist, etc that is more left but 15' behind the speaker. The illusion is sorta lost if you can see the speaker.
Scotth3886 is offline  
post #3590 of 5320 Old 07-10-2019, 04:07 PM
AVS Forum Club Gold
 
craig john's Avatar
 
Join Date: Oct 2003
Location: Lancaster, PA
Posts: 11,799
Mentioned: 65 Post(s)
Tagged: 0 Thread(s)
Quoted: 1339 Post(s)
Liked: 1577
Quote:
Originally Posted by motrek View Post
Please stop with this "listener's brain" nonsense.

Yes, sure, a human brain can process auditory cues to localize sounds.

What if the human then describes or points to where he thinks the sounds are occurring? Then that imaging (or at least part of it) exists "outside of his brain" as you would say.
When the listener points to the spot he perceives the sound to be originating from, what exists at that point in space? Not a sound source or a wavefront.


Quote:
Originally Posted by motrek View Post
Also, there's nothing that prevents a computer program from doing the same processing. Would you then say that imaging exists in the brain of the computer program? What does that even mean?
I suggested exactly that in the post you quoted.
Scotth3886 and garygreyh like this.

Lombardi said it:
"Perfection is not attainable, but if we chase perfection we can catch excellence."

My System
craig john is offline  
post #3591 of 5320 Old 07-10-2019, 04:16 PM
Advanced Member
 
motrek's Avatar
 
Join Date: Dec 2013
Posts: 653
Mentioned: 7 Post(s)
Tagged: 0 Thread(s)
Quoted: 518 Post(s)
Liked: 191
Quote:
Originally Posted by craig john View Post
When the listener points to the spot he perceives the sound to be originating from, what exists at that point in space? Not a sound source or a wavefront.
Right. There probably also isn't a puppy dog or a fire truck there, either. What does that matter? The idea is that a listener can convey his perception of where a sound is coming from, i.e., imaging. You've said half a dozen times that imaging can only possibly exist in a listener's brain, but if a listener can convey his sense of imaging (via talking, writing, pointing, etc.) then you're very clearly wrong.

Quote:
I suggested exactly that in the post you quoted.
I stopped reading your post after the first sentence about this listener's brain stuff.
motrek is offline  
post #3592 of 5320 Old 07-10-2019, 04:26 PM
AVS Forum Special Member
 
Soulburner's Avatar
 
Join Date: Oct 2007
Location: Lincoln
Posts: 4,892
Mentioned: 46 Post(s)
Tagged: 0 Thread(s)
Quoted: 194 Post(s)
Liked: 1629
The perception of the effect is processed in the brain.

The physics that cause that perception exist in the room. Therefore, it can be measured.
motrek likes this.

HT: Samsung PN64H5000 (recommended settings) | NAD T758 V3 | Buchardt S400 (2) | Emotiva E2 (2) | Rythmik Audio F12 (2)
Soulburner is offline  
post #3593 of 5320 Old 07-10-2019, 04:32 PM
AVS Forum Special Member
 
QueueCumber's Avatar
 
Join Date: Sep 2005
Location: Farmer McGregor's Garden
Posts: 6,318
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 165 Post(s)
Liked: 86
Quote:
Originally Posted by Soulburner View Post
The perception of the effect is processed in the brain.

The physics that cause that perception exist in the room. Therefore, it can be measured.
What is much more difficult to measure are differences in how the same effects are processed/percepted by different people.

Last edited by QueueCumber; 07-10-2019 at 04:36 PM.
QueueCumber is offline  
post #3594 of 5320 Old 07-10-2019, 04:48 PM
AVS Forum Special Member
 
Soulburner's Avatar
 
Join Date: Oct 2007
Location: Lincoln
Posts: 4,892
Mentioned: 46 Post(s)
Tagged: 0 Thread(s)
Quoted: 194 Post(s)
Liked: 1629
Quote:
Originally Posted by QueueCumber View Post
What is much more difficult to measure are differences in how the same effects are processed/percepted by different people.
Among those with normal audio processing abilities, meaning to exclude those with hearing problems, people generally process this stimuli the same way. Some may be more "experienced" listeners but that is where some training would get everyone on the same page or reasonably close to it. As was mentioned before, determining the direction of a sound is an evolved ability and those within the bell curve of the population would be recruited. Correlating perceptions with measurements so that measurements can tell us something about imaging - the same as Harman's methodology.
motrek likes this.

HT: Samsung PN64H5000 (recommended settings) | NAD T758 V3 | Buchardt S400 (2) | Emotiva E2 (2) | Rythmik Audio F12 (2)
Soulburner is offline  
post #3595 of 5320 Old 07-10-2019, 04:51 PM
AVS Forum Club Gold
 
craig john's Avatar
 
Join Date: Oct 2003
Location: Lancaster, PA
Posts: 11,799
Mentioned: 65 Post(s)
Tagged: 0 Thread(s)
Quoted: 1339 Post(s)
Liked: 1577
Quote:
Originally Posted by motrek View Post
Right. There probably also isn't a puppy dog or a fire truck there, either. What does that matter? The idea is that a listener can convey his perception of where a sound is coming from, i.e., imaging. You've said half a dozen times that imaging can only possibly exist in a listener's brain, but if a listener can convey his sense of imaging (via talking, writing, pointing, etc.) then you're very clearly wrong.
I get it... it's difficult to understand the concept of something existing only inside your brain.


Quote:
Originally Posted by motrek View Post
I stopped reading your post after the first sentence about this listener's brain stuff.
No comment.
Scotth3886 likes this.

Lombardi said it:
"Perfection is not attainable, but if we chase perfection we can catch excellence."

My System

Last edited by craig john; 07-10-2019 at 04:56 PM.
craig john is offline  
post #3596 of 5320 Old 07-10-2019, 04:54 PM
AVS Forum Addicted Member
 
sdurani's Avatar
 
Join Date: Oct 2001
Location: Monterey Park, CA
Posts: 28,004
Mentioned: 211 Post(s)
Tagged: 0 Thread(s)
Quoted: 7259 Post(s)
Liked: 6294
Quote:
Originally Posted by craig john View Post
I don't know how to measure something that only exists in the listener's brain, and not as a hard physical entity.
The same way they measure listener preference: do listening tests and then measure speakers to see if there is a correlation. Look at the questionnaire that Toole posted for his experiments in the mid-80s. The testing was done in stereo and mono. Yes, there can be phantom imaging with a single speaker (I've solo'd centre speakers that imaged forward of the speaker itself; i.e., floated between the centre speaker and me). The results from the sound quality part of the questionnaire tracked with the results from the directional/spatial quality part (good sounding speakers tended to be good imaging speakers). Mono listening results tracked with stereo listening results, except (as discussed repeatedly in this thread) the stereo results had less separation between them, making it easier/quicker to see preference rankings in the mono results.

So, if the results of the experiments to be believed, directional/spatial qualities of pairs of speakers track with sound quality of single speakers. And spinorama measurements are a pretty good indicator of sound quality preference. Does that mean spins can predict relative ranking of speakers when it comes to imaging and soundstage (directional/spatial qualities)? Like I said earlier, maybe there is research out there somewhere that shows significant changes in preference rankings between stereo versus mono tests (when it comes to directional/spatial qualities), but I haven't seen it.
Quote:
I don't think it has ever been shown that having two speakers be identical in the frequency domain necessarily guarantees they will image correctly. At least I'm not aware of any evidence the shows a causal relationship. Are you aware of any such studies?
No, but I haven't looked either. I was (maybe naively) accepting it as common knowledge. Speakers out of phase don't image. Speakers in phase do. Mis-matched speaker levels mess with directionality. Matched speaker levels image more accurately. Sooooo, the more identical the speakers (level, phase, etc), the better the imaging. Do you really need to see published research before believing that?
Quote:
I think there is more to it than that. The dispersion characteristics of the speakers will impact how much reflected sound impinges on the first arriving sound, which can have a detrimental impact on precise imaging. I think there's a possibility that cabinet diffractions could cause some arrival time disparities that could affect imaging. The alignment and placement of the drivers can also impact arrival times. The crossover could have some latency that affects arrival times. There's likely many other things that could affect this, many of which would not have an impact on the Spinorama measurements.
Other way 'round: while those factors might affect directional/spatial qualities, they'll also have audible impact on sound quality (those qualities are linked), which spinorama measurements can gauge quite well.
Quote:
Alternatively, one could use subjective listening tests, but I think that would be less reliable and repeatable than a 3-D measurement system. I realize that training the listeners can improve the consistency of the results, at least according to Harman, but I have yet to see that research repeated and verified by outside researchers. Until that happens, I'm not a believer... and I have taken the training.
Again, I think it's the other way around. For a phenomenon that occurs in your mind, two ears and a brain will yield better data than a pair of mics and a dummy head. If you're not a believer in trained listeners, then you can use untrained listeners. All that will accomplish is less repeatable, less consistent data to sift through.
Quote:
If you were designing an experiment to determine what causes speakers to image, and then testing speakers for imaging preference, how would you design such an experiment?
Like the one that was already done in the mid-80s.

Sanjay
sdurani is offline  
post #3597 of 5320 Old 07-10-2019, 05:09 PM
AVS Forum Special Member
 
QueueCumber's Avatar
 
Join Date: Sep 2005
Location: Farmer McGregor's Garden
Posts: 6,318
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 165 Post(s)
Liked: 86
Quote:
Originally Posted by Soulburner View Post
Among those with normal audio processing abilities, meaning to exclude those with hearing problems, people generally process this stimuli the same way.
That’s an assumption on your part. Perhaps certain aspects of perception are more universal, like spatial cues, because of their link to survival, but there may be other nuances that have yet to be identified that come into play. For example, similar or different nuances to those in this link: http://deutsch.ucsd.edu/psychology/pages.php?i=201
QueueCumber is offline  
post #3598 of 5320 Old 07-10-2019, 05:20 PM
AVS Forum Special Member
 
torii's Avatar
 
Join Date: Mar 2015
Posts: 7,242
Mentioned: 54 Post(s)
Tagged: 0 Thread(s)
Quoted: 3592 Post(s)
Liked: 2027
I think I remember reading that all the scientific instruments/gear used in all these tests were DOS based from 40 yrs ago...has new testing gear come out and does it help with anything?

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 online now  
post #3599 of 5320 Old 07-10-2019, 05:32 PM
Member
 
Floyd Toole's Avatar
 
Join Date: Jun 2017
Location: California
Posts: 836
Mentioned: 89 Post(s)
Tagged: 0 Thread(s)
Quoted: 875 Post(s)
Liked: 3078
Quote:
Originally Posted by QueueCumber View Post
What is much more difficult to measure are differences in how the same effects are processed/percepted by different people.
Indeed this is so. Humans learn and adapt, and it different people do it differently.

In my PhD research in the early '60s I was able to train listeners to believe that 0 interaural time difference (ITD) was not "center". All that was required was exposure to a laterally displaced abstract visual image on a CRT that supposedly tracked the auditory image - the visual image dominated the perception and subjects simply did not notice the discrepancy. After less than an hour of this "contradictory" experience a purely auditory localization exercise showed a displaced perception of "center". The aftereffect could last for more than an hour but retraining to zero ITD was possible in less time. At the time, psychoacoustic experimenters assumed that zero ITD was a definitive perception of center localizations.

In a much more complex way we adapt to rooms and to an impressive extent are able to "listen through" them to discern important qualities of voices and instruments in live experiences and loudspeakers in reproduced experiences.

Some of the discussions that I am reading assume that this is a "linear" cause and effect process. To a computer it is, but we are humans. We are barely able to model very basic perceptual dimensions with simple signals, much less interpretations of complex three-dimensional sound fields created artificially by two stereo speakers and a band or orchestra.

Behind it all there is an "Elephant in the Room" issue that has not, by my observation, been discussed. Strongly worded opinions about this or that in audio generally ignore the effects of hearing degradation. “I heard it, therefore it is real and true” is often assumed. It has been known for many decades that exposure to loud sounds causes temporary and then permanent hearing loss – the lost ability to hear low-level sounds. We also lose the ability to hear high frequencies. So, the combined effects of noise exposure at work and play, and age (presbycusis) reduces the information reaching our brains, and therefore our abilities to discern timbral and spatial (the ears may be different) subtleties and differences. When we go to our neighborhood audiologist for a “hearing test” it is based only on our ability to understand speech. It has nothing to do with discerning the small differences among pieces of high-performance audio gear. As discussed in Section 3.2 in the 3rd edition, within the audiologist’s interpretation of “normal” hearing it is possible to observe substantially degraded ability to judge sound quality. The opinions are real but not consistent in repeated exposures to the same sounds.

But then there is the recent discovery of something called “hidden hearing loss”. This can happen at any age, with or without accompanying audiometric threshold shift. The effect is binaural, affecting one’s ability to discern direction and to separate multiple sounds that coexist. The most common real-life example is when we have difficulty carrying on a conversation in a restaurant – the cocktail party effect, which takes me back to my advisor when I was a beginning PhD student. Chapter 17 describes some of the effects, and all of them bear on our perceptions of stereo soundstage properties. So, what any one of us perceives, may or may not be relevant to others.

I stopped participating in listening tests around age 60 because at Harman we not only track the subjective evaluations of loudspeakers, we also track the statistical performance of the subjects. I was not performing as well as I used to, so I retired from the “trained listener” population, and now any opinions I have are my own. But, all is not lost. Having spent 50 years at subjective/objective relations in loudspeaker performance, I now put more trust in a good set of anechoic measurements than in my own on-the-spot subjective evaluations.

Meanwhile, all that said, the 9.4.6 channel system I now have is sufficient to send chills down my spine when it is fed a good recording or movie. And the soundstage and imaging I hear are up to my highest expectations - when it is fed a good recording or movie. But that is me . . . in my room.
Floyd Toole is offline  
post #3600 of 5320 Old 07-10-2019, 06:11 PM
AVS Forum Special Member
 
Soulburner's Avatar
 
Join Date: Oct 2007
Location: Lincoln
Posts: 4,892
Mentioned: 46 Post(s)
Tagged: 0 Thread(s)
Quoted: 194 Post(s)
Liked: 1629
Quote:
Originally Posted by Floyd Toole View Post
But then there is the recent discovery of something called “hidden hearing loss”. This can happen at any age, with or without accompanying audiometric threshold shift. The effect is binaural, affecting one’s ability to discern direction and to separate multiple sounds that coexist. The most common real-life example is when we have difficulty carrying on a conversation in a restaurant – the cocktail party effect, which takes me back to my advisor when I was a beginning PhD student. Chapter 17 describes some of the effects, and all of them bear on our perceptions of stereo soundstage properties. So, what any one of us perceives, may or may not be relevant to others.

I stopped participating in listening tests around age 60 because at Harman we not only track the subjective evaluations of loudspeakers, we also track the statistical performance of the subjects. I was not performing as well as I used to, so I retired from the “trained listener” population, and now any opinions I have are my own. But, all is not lost. Having spent 50 years at subjective/objective relations in loudspeaker performance, I now put more trust in a good set of anechoic measurements than in my own on-the-spot subjective evaluations.
Yes, we should definitely recruit within the bell curve of the population and ensure participants are up to snuff in any future studies. That is good methodology for any study, really.

HT: Samsung PN64H5000 (recommended settings) | NAD T758 V3 | Buchardt S400 (2) | Emotiva E2 (2) | Rythmik Audio F12 (2)
Soulburner 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