Roger Sanders' views on audio: The discussion thread

[Fitzcaraldo215]

How do you think this can be explained measurement wise?

Or simply the simplest one of the speakers are just tall and that's why they always sound large/tall or something else?

I hear this whenever I've heard Sound Labs and Martin Logans at shows/dealer as well. Also the one time heard Acoustats (2+2?). IME it's most unsettling with solo instrument music and lesser extent on baroque cantatas, where the choir is standing layered in rows.

I think this was present to a lesser extent on Magnepans, even the large ones, and I say this because I can't recall them doing this, with the electrostats mentioned it was something that stood out immediately. I also owned the Magnepan MMG at one point and they projected images that were "normal size" and these are not tall speakers.

Good question. But, as I hinted, it might at least partially be the result of the visual cues of the tall speakers affecting what we think we hear during sighted listening. Or, it might be something else, including other suggestions from our subconscious from prior experience. Audio measurements, important as they are, only deal with things outside the eardrums. Imaging deals with things in the brain, the only place where the sound impulses from both ears come together. Hence, imaging is highly subjective and personal and it may be heavily influenced by our brain, bias and prior suggestion, perhaps even more so than many other things in audio already are.

Possibly, objective, DBT listener experiments could be used to compare these subjective imaging properties across many human subjects. But, I expect that listening room acoustics and speaker placement, rather than just speakers alone, also play a major role. So, I think questions about whether a given speaker "images better" would be impossible to answer definitively in objective fashion.

By the way, I concur that tall dipoles do sound different in many ways from monopoles, love 'em or hate 'em. That includes some aspects of imaging.

But, as I said, I do not hear any apparent height distortion or vertical stretching in the image using solo or small ensemble classical recordings on my tall Martin Logans. Or, if it is indeed there, it nontheless sounds plausible to me and consistent with my live concert experience. Possibly, my consistent use of a horizontal center channel in Mch listening plays an important role. In general, I believe a discretely recorded center channel provides major audible benefits, including with imaging.

For my purposes, as I indicated earlier, little could demonstrate this more clearly than the comparison I described of 12 simple a capella singing voices of Stile Antico in an arced row across the stage, live as you see them represented in the video link below vs. recorded audio-only.

I have been buying and listening their recordings for 10 years. Imaging nut that I am, they have never sounded like anything but the 12 singers in a curved row as you see them in the video. And, the nature of their music frequently shifts from solo to multiple voices up to the full ensemble. Voices are rather good, if not perfect, examples of point sources. But, before attending the live concert last week and recently googling this video, I did not have any visual evidence of their actual placement in performance. The aural image I had in my mind was always just based purely on what I heard from the SACDs. That is now confirmed to my own personal satisfaction in my system to be fully consistent with their placement in performance, and with no vertical stretching of the image.

But, of course, no general, universal conclusions can be drawn from my own anecdotal, subjective experience.

 

[svart-hvitt]

Good question. But, as I hinted, it might at least partially be the result of the visual cues of the tall speakers affecting what we think we hear during sighted listening. Or, it might be something else, including other suggestions from our subconscious from prior experience. Audio measurements, important as they are, only deal with things outside the eardrums. Imaging deals with things in the brain, the only place where the sound impulses from both ears come together. Hence, imaging is highly subjective and personal and it may be heavily influenced by our brain, bias and prior suggestion, perhaps even more so than many other things in audio already are.

Possibly, objective, DBT listener experiments could be used to compare these subjective imaging properties across many human subjects. But, I expect that listening room acoustics and speaker placement, rather than just speakers alone, also play a major role. So, I think questions about whether a given speaker "images better" would be impossible to answer definitively in objective fashion.

By the way, I concur that tall dipoles do sound different in many ways from monopoles, love 'em or hate 'em. That includes some aspects of imaging.

But, as I said, I do not hear any apparent height distortion or vertical stretching in the image using solo or small ensemble classical recordings on my tall Martin Logans. Or, if it is indeed there, it nontheless sounds plausible to me and consistent with my live concert experience. Possibly, my consistent use of a horizontal center channel in Mch listening plays an important role. In general, I believe a discretely recorded center channel provides major audible benefits, including with imaging.

For my purposes, as I indicated earlier, little could demonstrate this more clearly than the comparison I described of 12 simple a capella singing voices of Stile Antico in an arced row across the stage, live as you see them represented in the video link below vs. recorded audio-only.

I have been buying and listening their recordings for 10 years. Imaging nut that I am, they have never sounded like anything but the 12 singers in a curved row as you see them in the video. And, the nature of their music frequently shifts from solo to multiple voices up to the full ensemble. Voices are rather good, if not perfect, examples of point sources. But, before attending the live concert last week and recently googling this video, I did not have any visual evidence of their actual placement in performance. The aural image I had in my mind was always just based purely on what I heard from the SACDs. That is now confirmed to my own personal satisfaction in my system to be fully consistent with their placement in performance, and with no vertical stretching of the image.

But, of course, no general, universal conclusions can be drawn from my own anecdotal, subjective experience.

On «imaging»:

Referring to @Floyd Toole , see BACCH thread and book, imaging - which is not a scientific term AFAIK - is a figment of our imagination, our sensory apparatus. Good vs bad speakers are revealed in mono setups. The way from mono to «imaging» (which I guess is stereo or multi-channel related) seems to be explained by how we place those speakers in the room or apply DSP for effect.

So if I understand Toole correctly, there is no such thing as «great imaging speakers». Speakers are either good or bad. The rest (for example «imaging») comes as part of the package that is the source.

 

[Fitzcaraldo215]

I cannot disagree in any way. However, seeking "better" imaging, a personized phantom in our heads, is still important to many audiophiles and it comes up time and again in audio discourse, though without any useful objective standards.. And, yes, objectively "better" speakers are potentially capable of delivering "better" imaging, but many other factors may also come into play, including also technologies such as BACCH or Mch.

 

[oivavoi]

I do think imaging is a real phenonemon that has objective and measurable correlates. There is no reason to think that low-fi technology has any advantages for imaging - like vinyl, distorting low-powered audiophile amps, bad passive crossovers, resonating cabinets, low dynamic capabilities, uneven dispersion, and so on.

But there has been very little research on it, unfortunately. What we don't know, for example, is whether particular types of speaker dispersion and room response is conducive to imaging or not. Does phase response in speakers matter? I don't think we know that either. There is a lot of anecdotal evidence on the audiophile internets that early lateral reflections are bad for imaging. This would make it reasonable to go for narrow dispersion speakers, like electrostats, dynamic dipoles, horns, etc. This is stated by BACCH in their FAQ for example. And it has been claimed by Geddes as a rationale for his designs.

But afaik, no psychoacoustic evidence for this exists. As linked in the BACCH thread, the one study I know of (Choisel 2005) didn't find any adverse effect on imaging from reflections that were 9.5 ms delayed.

There was also the Linkwitz challenge test by David Clark. The outcome variable was admittedly vague, and results were not significant. Still, among the 19 participants a pair of cheap conventional active monitors from Behringer and a DIY omni speaker - both with wide lateral dispersion - were perceived as creating an acoustic scene that was at least on a par with the dipole Linkwitz Orions (they actually scored better than the Orions, but again, not statistically significant).

I personally find that imaging becomes clearer, more defined and deeper the more I hear the direct sound, with my current active monitors at least. Which could imply that room reflections indeed mess up the imaging for me. But how would it be if I was listening to point source omnis with perfect 360 degree dispersion? (which I hopefully will start doing some months from now). Will I perceive the early reflections as detrimental to imaging, or will they be so similar to the direct sound that my brain won't care about them? I don't know. The creator of those speakers, Don Morrison, explicitly claims their ability to image well as the main rationale for his design.

Summed up: I'm fairly certain that the imaging capabilities of a speaker setup has objective correlates. It is reasonable that more fidelitous speakers will image better than less fidelitous speakers. What we don't know yet, I would say, is how imaging is affected by room acoustics and the dispersion pattern of the speakers.

 

[RayDunzl]

But there has been very little research on it (imaging)

My limited "research" says:

• I prefer less wall reflections.
• Phase alone can change the location of sound sources in the perceived stereo image, with no change in amplitude between the speakers.
• Wall Reflections mess with phase (and perception)

Nearfield, my little JBLs have a nice phase and tight unsmoothed frequency and phase response (similar to the dipoles at listening position).

Farfield, they're a mess compared to my dipoles.

Unsmoothed frequency response of JBL vs ML dipole in this (relatively untreated) room at the listening position:

I've been informed "You can't hear that", my response would be "I hear something", which manifests itself in the perceived imaging.

 

[oivavoi]

Interesting, Ray. These messurements are after the acourate-correction, right?

 

[RayDunzl]

These messurements are after the acourate-correction, right?

The "hashiness" is not related to the "room correction".

Uncorrected, unsmoothed, JBL (my room) vs ML dipole at the listening position:



DallasJustice M2 left speaker only, in his room, at the listening position:



JBL 308, more nearfield, from a meter, or maybe 50 inches (I forget, old measurement) instead of ten feet:



Amateur Conclusion: The "hash" is from interaction with the room, not the drivers themselves.

 

[RayDunzl]

Here's a pair of MartinLogan CLS dipoles in their relatively untreated room with some Rythmik subs (my audio buddy across town), green, vs mine, red, and my JBL for reference, gray.

His setup has the listening position at about 12~13 feet.

Similar reduced hashiness is evident.



It's too bad panels aren't taken more seriously, so we could see some additional professional opinions.

 

[Fitzcaraldo215]

I had the CLSs for many years with Janis subs. The mids, highs and deep bass were fine. It was the mid/upper bass from the panels that sucked (out). I don't see that in your graph, but we did not have decent measurement tools and EQ in those days to enable me to tweak it.

Thanks, Ray.

 

[DonH56]

Quick comments (gotta' run):

• Dipoles interact less with the room since they radiate little off the sides or top/bottom. Curved panels such as ML uses interact more with the side walls to provide a bit more "spaciousness" in the sound from the added reflections.
• Midbass suckout is usually the room and/or interaction with the back wave from the dipole.
• Without room interaction imaging tends to be more precise but any ambience is from the source and not the room. Some people prefer that, many do not, and like the sound with the added reflections.

YMMV - Don

 

[Cosmik]

If you want hash-free you can have it: just create an acoustically dead anechoic room; if you believe that 'hash' per se is bad, then eliminate it - it's easy to do. Your measurements will be conventionally perfect.

Or, you can pursue the paradoxical alternative: create highly directional 'phased arrays' that give you an approximation to anechoic measurements in a normal room. Large panel speakers are an example of this.

Or, you can embrace the hash, and accept that as long as the speaker itself is something like neutral, your ears and brain will probably cope.

 

[Fitzcaraldo215]

On «imaging»:

Referring to @Floyd Toole , see BACCH thread and book, imaging - which is not a scientific term AFAIK - is a figment of our imagination, our sensory apparatus. Good vs bad speakers are revealed in mono setups. The way from mono to «imaging» (which I guess is stereo or multi-channel related) seems to be explained by how we place those speakers in the room or apply DSP for effect.

So if I understand Toole correctly, there is no such thing as «great imaging speakers». Speakers are either good or bad. The rest (for example «imaging») comes as part of the package that is the source.

Yes, again, but one more thing I forgot to add re Toole. As I recall, he makes the point in his book with empirical verification that listeners become less "picky" about the quality differences between speakers as the playback channel count increases. The difference between their comparative preferences for A or B, C, or D is reduced as channels are increased from mono to stereo to Mch. So, the stronger preference for "good" speakers with smooth, relatively flat frequency response and a good directivity index becomes somewhat less significant in "normal" stereo or Mch listening. Also, overall, stereo is generally preferred to mono and Mch generally preferred to stereo using the same speakers.

There may be many reasons for this. But, one that I think might also be there is that the increasing presence of an aural, dimensional "image" vs. mono becomes important, and that somewhat reduces the dominance of just the speaker frequency and directivity response in listener preference.

However, I have absolutely no doubt that Toole still believes, as do I, that "good" speakers, as determined by mono testing, are still the way to go. My own Martin Logans famously did not fare well in his testing. But, I persist with them anyway with reasonably high satisfaction in a Mch setup playing mostly discretely recorded Mch classical recordings.

 

[Jakob1863]

I do think imaging is a real phenonemon that has objective and measurable correlates. There is no reason to think that low-fi technology has any advantages for imaging - like vinyl, distorting low-powered audiophile amps, bad passive crossovers, resonating cabinets, low dynamic capabilities, uneven dispersion, and so on.

It is as real as it can get for human listeners, provided that we keep in mind that "imaging" is mainly something that we imagine, means - as said before - we perceive something that isn´t real but sounds like it were.
What in the reproduction is deteriorating, depends imo - as usual - on the environmental conditions. Early or late reflections, sound distribution patterns, room sizes, listening distance and ... and....

But there has been very little research on it, unfortunately. What we don't know, for example, is whether particular types of speaker dispersion and room response is conducive to imaging or not. Does phase response in speakers matter?

it depends on the meaning of "conducive"; as we normally do not know which way the recorded content is intended to "image" we refer to our experience of real acoustical events, so often everything that approximate better that experience from the past could be considered as being more conducive.
Of course sometimes we can conclude from the content ("dry" recorded for example, with near microphone setup and little or no artificial reverb) which way it is intended. But if you don´t like that, because it´s far out compared to real events is it conducive? From a certain viewpoint of course but from the other (cited quite often) it wouldn´t be.

Interchannel phase response does matter - as we know from recording and listening - down to surprisingly small numbers; intrachannel isn´t so finely explored, group delay does play a role (Blauert & Law ?19xx?), Kunchur did some more experiments which might have shown that intrachannel phase response is more important than we thought before, but there is replication (and or additional experimentation needed) as some methodological questions/problems occured and conclusions were imo a bit to optimistic.

I don't think we know that either. There is a lot of anecdotal evidence on the audiophile internets that early lateral reflections are bad for imaging. This would make it reasonable to go for narrow dispersion speakers, like electrostats, dynamic dipoles, horns, etc. This is stated by BACCH in their FAQ for example. And it has been claimed by Geddes as a rationale for his designs.

But afaik, no psychoacoustic evidence for this exists. As linked in the BACCH thread, the one study I know of (Choisel 2005) didn't find any adverse effect on imaging from reflections that were 9.5 ms delayed.<snip>

I´ll cite some more in the next days; there is a lot of research going on wrt perception of spatial properties.

 

[oivavoi]

Thanks, Jakob, excellent stuff!

 

[beeface]

I don't have a technical background, and there's a couple of things I've always wondered about their Magtech amps.

"In short, the Magtech's power supply is unique and solves all the problems of other regulators that have prevented power amplifiers from being regulated -- something they badly need even more than other types of electronics. The Magtech regulator's circuit, and particularly the digital control technology involved is the subject of a patent, which currently is pending.

The Magtech amplifier modules are the same sophisticated ones used in the ESL amp that are capable of very high power, the ability to drive 1/3 Ohm loads, can handle the most difficult loads (as presented by electrostatic speakers), and need no protective circuitry that ruins the sound of many solid state amps.

When the ESL amplifier modules are combined with a practical voltage regulator, the result is an amplifier with seemingly unlimited power, virtually unmeasurable distortion, and the ability to drive even the most difficult loudspeakers with ease. The Magtech offers a truly new level of performance in amplifiers."​

Above quote is from this whitepaper.

More info here and here.

I note that the technology has a pending patent, but I do wonder a couple of things:

What's the catch (if there is one)? I know that the whitepaper acts as marketing, but I don't have enough technical knowledge to determine that they haven't come up with a solution without drawbacks.

And say if this is in fact a near-perfect amplifier, why haven't competitors produced something similar? Patents aside, I would have thought a savvy engineer would have produced something similar-but-different enough to market.

 

[Wombat]

I don't have a technical background, and there's a couple of things I've always wondered about their Magtech amps.

"In short, the Magtech's power supply is unique and solves all the problems of other regulators that have prevented power amplifiers from being regulated -- something they badly need even more than other types of electronics. The Magtech regulator's circuit, and particularly the digital control technology involved is the subject of a patent, which currently is pending.

The Magtech amplifier modules are the same sophisticated ones used in the ESL amp that are capable of very high power, the ability to drive 1/3 Ohm loads, can handle the most difficult loads (as presented by electrostatic speakers), and need no protective circuitry that ruins the sound of many solid state amps.

When the ESL amplifier modules are combined with a practical voltage regulator, the result is an amplifier with seemingly unlimited power, virtually unmeasurable distortion, and the ability to drive even the most difficult loudspeakers with ease. The Magtech offers a truly new level of performance in amplifiers."​

Above quote is from this whitepaper.

More info here and here.

I note that the technology has a pending patent, but I do wonder a couple of things:

What's the catch (if there is one)? I know that the whitepaper acts as marketing, but I don't have enough technical knowledge to determine that they haven't come up with a solution without drawbacks.

And say if this is in fact a near-perfect amplifier, why haven't competitors produced something similar? Patents aside, I would have thought a savvy engineer would have produced something similar-but-different enough to market.

Various Brands have implemented sophisticated PS voltage regulation including on-demand supply of peak demand voltage/current, well above the continuous RMS rating, for shorter periods of time. The NAD(Carver and Polk) system was called Power Envelope. Commutating power supplies that have more than one rail voltage that is/are brought into play as needed are not new.
Magtech's approach may be to do with a 'novel'(in patent terminology) method of control switching.

FYI:

Download PDF Proposal, here: https://www.hifiengine.com/manual_library/nad/power-envelope.shtml

A design paper: https://pdfs.semanticscholar.org/e9d1/33d8e51071b7e5ef3c9e3bf408bb0df2977d.pdf

 

[Burning Sounds]

I have one example:

JBL LSR 308 vs Martin Logan reQuest

View attachment 11997

The harmonic distortion can be easily heard.

And third harmonic on a sine log sweep:

View attachment 11998

A better cone/dome speaker will perform with less obvious difference. The JBL crosses from its 8" woofer and midrange to the tweeter at 1.7khz. Above that, the third harmonic is much less pronounced.

Oh, one more example. Infinity P-363 (three way), which has a better midrange, but maybe a lesser tweeter:

View attachment 12001

And one more... My buddy has a pair of Martin Logan CLS at his house (blue) vs my reQuests (green)

View attachment 12002

Thanks for this, Ray - where was your mic placed for these measurements?

 

[RayDunzl]

where was your mic placed for these measurements?

Always (unless not) at the point between my ears at the sweet spot listening position on the couch, 10 feet from speakers, in an 18 x 14 x 9 room, with the left rear corner wide open to the mess hall.

 

[Burning Sounds]

Then if your room is adding any distortion (which I imagine it is) those are very good measurements, indeed.

My LX521 measurements are not exactly comparable as I haven't done any averaging and the levels are slightly different, but here they are for comparison. Played through JRiver with Acourate switched in. The crossover between lower and upper mid drivers is at 1kHz - (L-R2).

Left speaker only - mic at 1 meter



And with mic at the listening position - about 8ft from speakers in a 20x12x8 room.



The 2nd harmonic is the dominant one in my case and seems to increase more than other harmonics with level and distance. I'm presuming this is primarily the room affecting things??

 

[RayDunzl]

The 2nd harmonic is the dominant one in my case and seems to increase more than other harmonics with level and distance. I'm presuming this is primarily the room affecting things??

I don't see how the room would add harmonics. (I get to be wrong)

Exception: Something physical in the room resonating with the fundamental and creating its own set of harmonics. I figure that is rather unlikely.