why not the perfect reproduction?

[Mike Lavigne]

All that requires is every stage to be correctly done and faithful, adding / changing as little as possible all the way to your ears.
The difference between headphones (that aren’t perfect either but at least don’t act on your room) and loudspeaker imaging, ought to be a big red flag for people looking at “where the problem really is” in addition to the normal culprits.
Best,
Tom Danley

headphones are certainly not any more linear than loudspeakers; i would agree that their output is more predictable. unfortunately; the environment of the 'near ear' is just as challenging as the room, just not quite as variable.

as far as imaging; headphones are either a disaster, or total disaster (relative to imaging in my room).

better headphones can allow a listener to more easily hear detail than in most systems, but not all systems; but have lots of other issues.

recently i purchased 2 sets of headphones (HD-800 and Stax O2 Mk1) and 2 headphone amps (Woo 6SE and Stax SRM-717) to investigate for myself whether they can get me closer to the music than my 'system'. both of these are pretty far up the food chain of headphones....if not quite to the very top.

neither betters my system in any way. both are quite good and i can see their attraction in bang-for-the-buck terms.

 

[Tom Danley]

Mike

It is my contention that part of the problem is both in not approaching the capture and launching of the sound field properly.
What your ears do (pina response etc) is not part of the electronic capture, like the “thin film analogy” above the goal is to capture, preserve, transfer reproduce / re-create something like the real thing would have right where you are.

I don’t want to repeat post but since you DO have some very good head phones, I would be interested to see if you think these captures a more real stereo image.
The loudspeakers we make at work are my ideas about launching / reproducing the wavefront , the recordings are an experiment in capturing that.
This and the other recordings are a different approach to capturing the sound field, an invention in progress.
In this case, it is designed to be the “forward view” with a width about 120 degrees (about like your vision) which is all I can get in two channels (takes more channels and speakers to “go all the way around and overhead” .
Best,
Tom

Try these
http://www.danleysoundlabs.com/Donny's Harley.wav

http://www.danleysoundlabs.com/TrainStart.wav

 

[Bruce B]

I don’t want to repeat post but since you DO have some very good head phones, I would be interested to see if you think these captures a more real stereo image.
The loudspeakers we make at work are my ideas about launching / reproducing the wavefront , the recordings are an experiment in capturing that.

We have the same headphones here and I will say that NO heaphone captures the spatial cues of depth and horizontal imaging. None..... sure it will capture the size of the room, but you can not tell if one instrument is in front of another and how far they are apart, or even if they are side by side. That's why you cannot use headphones in mastering or mixing. The only time I listen on headphones in the studio is to pick out artifacts and such.

 

[Phelonious Ponk]

Mike, those are very good headphone systems. Of course they will never image the way speakers can in a room, but if neither of those systems present detail better than your speaker system, congratulations are in order. You have, evidently, addressed a long list of problems common to even very good speaker systems, at a level that is unheard of. And you have utterly conquered your room's acoustics. I've heard a lot of systems operating on a lot of levels. I've never heard a domestic hifi system that could come close to the detail you should be hearing from those headphone systems. I've heard a few studio control rooms that come close, but even there, as Bruce said, they pick up the cans (and often not of the quality you're talking about) to listen in for the small stuff.

Tim

 

[Mike Lavigne]

Mike, those are very good headphone systems. Of course they will never image the way speakers can in a room, but if neither of those systems present detail better than your speaker system, congratulations are in order. You have, evidently, addressed a long list of problems common to even very good speaker systems, at a level that is unheard of. And you have utterly conquered your room's acoustics. I've heard a lot of systems operating on a lot of levels. I've never heard a domestic hifi system that could come close to the detail you should be hearing from those headphone systems. I've heard a few studio control rooms that come close, but even there, as Bruce said, they pick up the cans (and often not of the quality you're talking about) to listen in for the small stuff.

Tim

of course, my perceptions are subjective and therefore not definitive.

before i purchased these 2 particular headphone systems i researched what might be representative of a very good dynamic set-up and electrostatic setup. i assumed that headphones could allow me to determine what i was not hearing in my system. i was surprised that i could not hear more detail from either headphone setup than my 'system'. which was the area where i had expected to find a difference in favor of the headphones.

both headphone amps are connected directly to my single ended 'line out' on my darTZeel preamp, so they are not attenuated by my preamp. i also listened with the headphone amps connected directly with my sources, but heard no difference. the darTZeel is very quiet and 'transparent' apparently. the headphone amps are both plugged into the Equi=tech power grid using one of Gary's Absolute Fidelity power cords. so they are optimized with my sources and the power grid beyond what these headphone amps might ordinarily see.

i spent about a week going back and forth trying to find the upside of the headphones. i'll continue to investigate.

my explanation of why i'm hearing 'at least equal levels of detail' thru my system compared to these formidable headphone systems i think has to do with how quiet my system is compared to most loudspeaker systems, but also the resolution of the darTZeel amplifier and the accuton ceramic mid-range and arum cantus ribbon tweeter in my Evolution Accoustics MM3's. everything matters. everything matters. low noise floor is the overwhelming advantage a headphone system generally has. in my case; it does not have this advantage. the room/system acoustics are certainly a part of what works well in my loudspeaker system but noise floor is a larger part of the question of detail.

Bruce, in his post above, does feel he gets additional 'clues' from his HD 800's.

i would expect that Bruce's system would be very similar to mine in these areas. same speakers. we use different preamps and different amps. our cables are different. he lives more in the city and i live in the boonies on 4+ acres; but he has a room inside a room and it's dead quiet. we both use the Equi=tech. not sure he has a separate headphone amp which should give the headphones an advantage in my system.

he is better looking than me and has a motorcycle.

 

[Tom Danley]

I am not sure if you were following the posts here but I suspect we are approaching the elephant from different sides..

“We have the same headphones here and I will say that NO headphone captures the spatial cues of depth and horizontal imaging. None..”

I agree 100% however but would offer headphones are not involved in the capture process.

For me, approaching this from a transducer / acoustic point of view, I believe the reproduction system is a series of links in a chain. While it is easiest to measure think about and address the middle of the chain where it is electronic, it is at either end where it falls down the hardest.

The reason is once the signal is “electronic” it is one dimensional. I don’t mean to ignore frequency time etc but In other words, like when you add two signals with a resistor, one has complete vector summation. Take two simple sound sources and once they are more than about 1 /3 wavelength apart, they DO NOT add like signals in one dimension like resistors. Reverse one of the two sources at a larger spacing and one finds the total power remains unchanged and only the shape of the interference pattern changes. The ends of the chain pick up spatial information and re-create the spatial signature …or not.

As a loudspeaker / transducer person it seemed to me that part of the key to how speakers worked in larger spaces was governed in part by how the sources behaved in 3 dimensions, not the one dimension I could measure with one microphone. By eliminating the self interference essentially all multi-way speakers produce spatially, we were able to make powerful speakers that even in large scale use also sound like or better than many hifi speakers. It has been the fact that regular people can even hear the difference that has allowed us a small company to bump out the big name line arrays that are so popular in larger scale sound. Do an internet search on our speakers to see how they are thought of.

One side effect of the one source / no self interference result was that they also produced a very strong mono phantom image. While that is nearly irrelevant in most commercial sound it is important in the home or studio.

The microphone experiment is the current state of my attempt to “capture” the directional information without the interference pattern that is encoded when live recordings are made normally.
In the 80’s I heard a recording technique at Don Davis’s Synaudcon farm, it was called “In the ear recording” where a tiny tube was placed next to a person’s ear drums and the actual sound was recorded.
This recording included a generalized pina response and so when played back properly, gave a very vivid and REAL stereo image. In one of the recordings Don played was someone walking around the pits at Indy 500, most cool.

It struck me since that ones ears already took care of that, what I should do is capture and then reproduce the vector motion of the air molecules, your ears / brain will know the rest

With the right loudspeakers, one can produce a continuous fully believable mono phantom or panorama between the two sources, by exploiting head response tricks (fakery) one can extend that.
With three speakers in a triangle, one can fill in the triangular space, with four one has the “front wall” where an image can appear at any height or position and so on. To go all the way around the same way requires 5 or 6 channels.

We supply many SH100 loudspeakers for artificial reverberation systems like Vras etc, but it appears that the actual overhead image can be captured with just two or three more channels.

What is on the recordings are the two channels that describe the forward facing image. I am not saying it is perfect if you saw how ugly it is you would laugh and there is Zero post processing or sweetening, this is the raw output not a finished record product.

I am saying I think this gives a very different image than most have heard even with headphones and if it does sound like that, then that should say there ARE still other ways to capture which give audibly different results and this is part of the issue overall.. If you use SH-50’s at a typical stereo spacing, what you hear with good headphones happens in front of you in the horizontal plane albeit somewhat narrower.

If you didn’t try the recordings, please do, these are not your regular cup of earwax..
Best,
Tom



I am not sure if you listened to the recordings though, if you didn’t I would be curious what you thought, if so, I would still be curious what you thought.

 

[Mike Lavigne]

I am not sure if you were following the posts here but I suspect we are approaching the elephant from different sides..

“We have the same headphones here and I will say that NO headphone captures the spatial cues of depth and horizontal imaging. None..”

I agree 100% however but would offer headphones are not involved in the capture process.

For me, approaching this from a transducer / acoustic point of view, I believe the reproduction system is a series of links in a chain. While it is easiest to measure think about and address the middle of the chain where it is electronic, it is at either end where it falls down the hardest.

The reason is once the signal is “electronic” it is one dimensional. I don’t mean to ignore frequency time etc but In other words, like when you add two signals with a resistor, one has complete vector summation. Take two simple sound sources and once they are more than about 1 /3 wavelength apart, they DO NOT add like signals in one dimension like resistors. Reverse one of the two sources at a larger spacing and one finds the total power remains unchanged and only the shape of the interference pattern changes. The ends of the chain pick up spatial information and re-create the spatial signature …or not.

As a loudspeaker / transducer person it seemed to me that part of the key to how speakers worked in larger spaces was governed in part by how the sources behaved in 3 dimensions, not the one dimension I could measure with one microphone. By eliminating the self interference essentially all multi-way speakers produce spatially, we were able to make powerful speakers that even in large scale use also sound like or better than many hifi speakers. It has been the fact that regular people can even hear the difference that has allowed us a small company to bump out the big name line arrays that are so popular in larger scale sound. Do an internet search on our speakers to see how they are thought of.

One side effect of the one source / no self interference result was that they also produced a very strong mono phantom image. While that is nearly irrelevant in most commercial sound it is important in the home or studio.

The microphone experiment is the current state of my attempt to “capture” the directional information without the interference pattern that is encoded when live recordings are made normally.
In the 80’s I heard a recording technique at Don Davis’s Synaudcon farm, it was called “In the ear recording” where a tiny tube was placed next to a person’s ear drums and the actual sound was recorded.
This recording included a generalized pina response and so when played back properly, gave a very vivid and REAL stereo image. In one of the recordings Don played was someone walking around the pits at Indy 500, most cool.

It struck me since that ones ears already took care of that, what I should do is capture and then reproduce the vector motion of the air molecules, your ears / brain will know the rest

With the right loudspeakers, one can produce a continuous fully believable mono phantom or panorama between the two sources, by exploiting head response tricks (fakery) one can extend that.
With three speakers in a triangle, one can fill in the triangular space, with four one has the “front wall” where an image can appear at any height or position and so on. To go all the way around the same way requires 5 or 6 channels.

We supply many SH100 loudspeakers for artificial reverberation systems like Vras etc, but it appears that the actual overhead image can be captured with just two or three more channels.

What is on the recordings are the two channels that describe the forward facing image. I am not saying it is perfect if you saw how ugly it is you would laugh and there is Zero post processing or sweetening, this is the raw output not a finished record product.

I am saying I think this gives a very different image than most have heard even with headphones and if it does sound like that, then that should say there ARE still other ways to capture which give audibly different results and this is part of the issue overall.. If you use SH-50’s at a typical stereo spacing, what you hear with good headphones happens in front of you in the horizontal plane albeit somewhat narrower.

If you didn’t try the recordings, please do, these are not your regular cup of earwax..
Best,
Tom



I am not sure if you listened to the recordings though, if you didn’t I would be curious what you thought, if so, I would still be curious what you thought.

i did listen to the 2 recordings, but only thru my laptop speakers.....as i'm a bit 'slow' about capturing them onto my server so i could listen thru my system or headphones.

both recordings did allow for a wonderful rendering of space and motion as far as i could tell as i don't typically listen to my laptop. i don't have the appropriate plug to use my headphones with my laptop.

later today i'll get my son to install those files on my server and then i'll listen properly.

sorry about the 'headphones' detour from the topic.

 

[microstrip]

http://www.danleysoundlabs.com/pdf/danley_tapped.pdf
The staggered “progressively assembled wavefront” design also compensates for the phase shift in the crossovers such that a speaker like the SH-50 can (like an ESL-63) re-produce a square wave, over a broad band. In the SH-50 doing it from fair to near perfect from about 240Hz spanning the low / mid and Mid / high crossover to about 2700Hz and not just in one position in front.
It radiates the same way as the sheet of film excited by a source at the apex, a segment of a point source.
Best,
Tom Danley

The polar graphs of the SH-50 really look impressive - if it sounds like a 100dB/W ESL63 it sould be an impressive speaker!

Although my current main speakers are the Soundlabs I still keep the ESL63 - it has been a constant in my system for 30 years, sometimes in the primary system, others just as a secondary. Used within their limitations they are the most lifelike speaker I have owned. But they really need good electronics and sources to show their best. One of my dreams is just to carry the modifications that are fitted to the two pairs in the SME music room - mostly added mass to the structure.

 

[Tom Danley]

Hi microstrip
“The polar graphs of the SH-50 really look impressive - if it sounds like a 100dB/W ESL63 it sould be an impressive speaker!”

Thanks, I would like to think they do share a similar sound with the dynamics of a horn, the result is similar although the method is different.
The 63 has a particular resonance, a place in my heart for me, it did open my eyes to a different intention, a different way of thinking than I was aware of.
That was the first time I had ever thought about radiating a sound is a three dimensional process and at best we measure one location, here a speaker was deliberately producing a part of a simple sphere. If you did that, then the signal IS identical or close everywhere on the front surface. What I call spatial distortion is the degree that surface is not uniform, not identical.

One way a loudspeakers spatial coloration can be “heard” is if one listens to the signal an omni mic picks up vs the signal signal.
One of the ways we have refined the speakers (more of a sanity check of the right direction) is to make generation loss recordings where music is passed through the speaker and then recorded with generation one colorations. That is repeated each time multiplying the colorations and errors (with no errors one has no generation loss). By two or three generations most speakers are unlistenable. While this was fun because you can hear what’s wrong very clearly, the generation loss test do not give one any guidance so far as what to fix and how.
Anyway, you can’t tell much but this video might be fun, it is a much larger speaker used in sports stadium sized listening areas (2nd link) .
It is also a coherent single source and so at least that part can be heard in a recording of it.
At the farthest distance here (they measured later) camera was 700 feet, the camera a Cannon t2i, audio is straight from the camera.

http://www.facebook.com/video/video.php?v=1717077460850&oid=126113687424773&comments

http://www.lightingandsoundamerica.com/news/story.asp?ID=-4B8JO6

“Although my current main speakers are the Soundlabs I still keep the ESL63 - it has been a constant in my system for 30 years, sometimes in the primary system, others just as a secondary.
Used within their limitations they are the most lifelike speaker I have owned.”

I can imagine the SL’s (the only ones I have heard were very large driven by huge OTL’s) would be as hard or harder to drive than my home made panels (which were harder than the 63’s and Accustats). That size would provide a LARGE amount of directivity compared to most speakers and so the near field is much larger than usual (cones and domes), normally that would mean your imaging is likely very good from much less room sound like the setup I heard. Fwiw (producing the lowest level of room sound) is the reason directivity is so important in larger rooms, the larger the room, the smaller the absorption is relative to cubic volume. IE; put the sound only where the people are.

I had built a direct drive tube amplifier once but used a pair of old Mcintosh M-60’s most of the time.
Funny, I worked at a TV store in High School in the early 70’s, people were trading in the Mac stuff for the new Panasonic hifi gear (pre Technics).
I used to buy the traded in mac stuff from the store for myself (which was as close as I could ever come to buying anything cool).
To go back in time and buy a dusty mac 240 and mx 110 for $20 again ahh.

The issue was /is ESS speakers are a parallel capacitive load coupled through a transformer who’s high end response is governed by series inductance. Like tube output transformers, the best step up transformers are not simple. Back then driving my home built panels, SS amplifiers tended to sound bright or brittle or even icky while the Mac sounded fine. What amplifiers do you use?
Best,
Tom Danley

 

[microstrip]

(...)
I had built a direct drive tube amplifier once but used a pair of old Mcintosh M-60’s most of the time.
Funny, I worked at a TV store in High School in the early 70’s, people were trading in the Mac stuff for the new Panasonic hifi gear (pre Technics).
I used to buy the traded in mac stuff from the store for myself (which was as close as I could ever come to buying anything cool).
To go back in time and buy a dusty mac 240 and mx 110 for $20 again ahh.

The issue was /is ESS speakers are a parallel capacitive load coupled through a transformer who’s high end response is governed by series inductance. Like tube output transformers, the best step up transformers are not simple. Back then driving my home built panels, SS amplifiers tended to sound bright or brittle or even icky while the Mac sounded fine. What amplifiers do you use?
Best,
Tom Danley

Tom,
Thanks for such a detailed answer - I never got such a good explanation about the Sound Labs interaction with rooms!

I also found Soundlabs to sound brittle (not bright as they have a treble level adjustment) with most soild state amplifiers. I am using them with with VTL MB750s, but also listened to them with the Atmasphere MA2's OTLs . Another amplifier that sounded wonderful with them was the Dartzeel stereo, but it did not have power enough for them. An extreme case was the Pathos TT amplifier - it could not play loud, but detail, musicality and presence were extreme. Both these amplifiers should sound great with your SH-50.