Is Linkwitz Wrong? Phychoacoustics in Stereo Triangle.

[gainphile]

When was stereo ever stated as being anything more than an illusion?

" I came across a website with completely different conclusions. These conclusions are marketed primarily for the creation of music, not reproduction, but can still be applied."

 

[LineArray]

Assuming Linkwitz assumption is true that aslong as CD speakers are used and the reflected sound are attenuated copies of the direct sound, the first reflection points would still benefit from phase-coherent diffusion, otherwise untreated first reflection points will cause comb filtering. The diffusion will not selectively attenuate frequencies, like absorbers do.
...

Hi together,

i am "not happy" with that "copy notion", being it willingly introduced by SL or not.


It is known, that accepted "room curves" are falling with frequency:
Reflected sound may not represent "a copy" of the direct sound to achieve that.

And where diffusivity is applied in rooms (which i regard beneficial), we have single reflections broken up in a multitude of reflections, that may have a smooth "overall" spectrum, but the single reflection will have altered frequency response and phase. The reflected sound is then decorrelated from the (loudspeaker's) direct sound, which also reduces interference effects that come with strong and "coherent" (single) reflections.

The "copy" notion does neither go together (work) with "falling room curves" nor with "diffusivity" of the room ...

Since both IMHO are beneficial in loudspeaker-/room interaction, there must be something "plain wrong" with that "copy" notion, being it intended "literally" by SL or not ...


Reflections as "copies" do not work well and are also misleading IMO.

Also "true CD" in loudspeakers does not work in such rooms, having reverberation time fairly constant with frequency.

 

[LineArray]

Supplement:

Even the evaluations and results in "loudspeaker preference" as conducted at Harman MLL (Multichannel Listening Lab) by Floyd Toole and Sean Olive, are based on a highly diffusive Environment ...

From there we know "something" about preferred loudspeaker/room interaction, but that knowledge is not based on reflected sound in the listening room being "copies" of the direct sound:

It is IMO intentionally based on "something completely different" (to remember Monty Python ...).


https://www.harman.com/sites/defaul...:02/files/HarmanWhitePaperMLLListeningLab.pdf

See e.g. Page 17

 

[63127]

Hi cormeister,

If there are "naked", flat and close side walls, there will be a huge difference (dipole vs. conventional) in that a "toed in" dipole will excite less ipsilateral reflections than the conventional monopole speaker and thus the balance between ipsilateral and contralateral reflections will move to the contralateral, when using a dipole.

If the side walls - at relevant reflection areas - are absorptive or even highly diffusive, the perceivable difference (dipole vs. conventional) will be smaller.





IMO you are making a common mistake here in assuming the rearward radiated sound from a dipole speaker would be aiming at "simulating/mimicking the reflections of a music hall" (or original venue in general): This is a common misunderstanding but nevertheless plain wrong.

Hi,
Well I'm not making the assumptions you think. I know the room is not involved in completely mimicking the spatial reverberation I was simply stating the same info is fired forwards and back. Since most mics are not truly omnidirectional its still an approximation. The psychoacoustics part of speaker design could be argued till the cows come home, I like Jazz in a small club where the piano and trumpet and drums are 10' from my ears, very direct and and swamps reverent energy.

Also it's interesting you thought I was a reviewer, quite the opposite. I have been quietly building speakers here in Colorado for three decades mostly for myself but have built many speakers for people who gave up throwing money at the problem, including a previous Orion owner.

Do you happen to own the LX or Orion? Would be interesting to see the FR and radiation pattern at your listening position. When I get the chance I will post the FR and polar pattern at my chair.

 

[LineArray]

Hi,

Hi,
Well I'm not making the assumptions you think.

OK.

The psychoacoustics part of speaker design could be argued till the cows come home ...

i have to remember that saying/idiom, should even work in translation to my region

Also it's interesting you thought I was a reviewer, quite the opposite. I have been quietly building speakers here in Colorado for three decades mostly for myself but have built many speakers for people who gave up throwing money at the problem, including a previous Orion owner.

i see, quite comparable to my story.

Do you happen to own the LX or Orion? Would be interesting to see the FR and radiation pattern at your listening position. When I get the chance I will post the FR and polar pattern at my chair.

No, i am sorry. But i have some experience with my own dipole designs.

I am currently shifting more towards cardioids (still working with separate dipole subs in the room) regarding main speakers.

Dipoles as main speakers are not applicable to so many rooms, e.g. due to needed front wall distance. We usually do not have too large living rooms here in Europe and such distances often place a conflict with the use of the listening room as a living room (and WAF).

Cardioids perform pretty robust due to different front wall distances possible ...

 

[soundbloke]

I came across a website with completely different conclusions. These conclusions are marketed primarily for the creation of music, not reproduction, but can still be applied

Not quite sure what this is getting at...

But if we are talking about recorded music - that is real performers in an acoustic space then stereo cannot be more than an illusion (except as previously stated).

If we are instead talking about music created electronically that has never been convoluted with any real acoustic or microphone then the creation of an intended sound direction is plausible. What constitutes fidelity or illusion here is somewhat ambiguous, however.

 

[Pano]

This eventually brought me a way more convincing result. I couldn't figure out why at first. After Jim1961 showed his measurements (mimicking a measurement at the ear with head shading) things fell into place why this was different, better even.

Interesting results and graphs, thanks for posting. I can understand how this might work like our phase shuffler or large amplitude reflections. It fills in the gaps caused by head shading. Now you are making me regret not experimenting more with ambient channels. There is so much to do and learn.

 

[stvnharr]

I heard the Orion once at RMAF, maybe its first year. I was also underwhelmed. "What's the big deal?" about sums it up. Nice enough, but not really out of the ordinary, not a game changer. I've spoken to a few people who have heard it in a domestic setting and really like it, so maybe it was the room?

FWIW, I'm an open baffle fan, and also love classical music. So it should be right up my alley.

I also heard the Orion the first time SL showed at RMAF and had the same reaction. You had to sit in the middle seat, and even then.......... My own speakers use Seas Mags, W18's, and have passive filter networks, and sound much the same.
I then heard the Orions at a later RMAF and the sound was better, but only because SL began using Nelson's amps and ditched the home theater ATI's.

For best sound and speaker setup it's best to go a couple blocks from RMAF to Soundings and listen there. Sometimes they have a room at RMAF, sometimes not. If you understand what they do, you'll never go back to anything else.

 

[Pano]

It is known, that accepted "room curves" are falling with frequency:
Reflected sound may not represent "a copy" of the direct sound to achieve that.

I had forgotten, but in the highly diffuse acoustics of the lava cave, I did not like the classic falling response room curve. If it didn't measure flat, it sounded too dull. In domestic listening rooms the falling response usually sounded most natural to me.

Does an even, diffuse reflected sound field negate the need for the room curve? I wonder.

 

[LineArray]

...
Does an even, diffuse reflected sound field negate the need for the room curve? I wonder.

You mean the need especially for a room curve falling with frequency, like the curve from Bruel&Kjaer or from Harman ?

Usually i would not think so ...

Did you listen in a usual stereo setup in that lava cave, having "flat" on axis response of the speakers pointed at the listening seat ?

I think it may also be possible to have some variations of the preferred "target curve" when comparing different programmes like speech, classic music, ...




2 Links that come to my mind regarding inroom target curve:

An Acoustic Basis for the Harman Listener Target Curve | InnerFidelity

http://www.2pi-online.de/diffuse_attenuation_color.png

_________

Would be interesting - nevertheless - to have a reverberation curve (reverberation time versus frequency) of the lava cave.

 

[LineArray]

Supplement:

"Effect of acoustic environment on the sensitivity of
speech transmission index to source directivity", Ken Stewart and Densil Cabrera

https://www.acoustics.asn.au/conference_proceedings/AAS2008/papers/p43.pdf

For directivity of human talkers cf. Page 6, Figure 5


My interpretation of the diagram:

In long term spectra of "conversational speech" a human talker seems to resemble a "constant directivity" sound source above approx. 1.5 Khz rather closely.

 

[milkshake]

milkshake,

I'd be interested in hearing your recordings. But I'd like to request a few additional/differerent contents --
1) delays on the order of 20msec or so (not just a few --- Haas relevant time delays)
2) delays that can be made to come from a different direction (so, multichannel, maybe 4.0?). I don't think there is any question that the directions that sound reflections come from is a relevant variable, is there? In my personal (and obviously, subjective for now) experience, this makes the difference between being in a large doorway to a room where music is playing; or being in the room where music is playing.

First things first.
Here are some files with various single delay's mixed at the same level as original and reverb added to the original sound at various levels.
https://www.dropbox.com/s/un74oy0yucak225/Delay Reverb.rar?dl=0

 

[Pano]

You mean the need especially for a room curve falling with frequency, like the curve from Bruel&Kjaer or from Harman ?

Yes, exactly

Did you listen in a usual stereo setup in that lava cave, having "flat" on axis response of the speakers pointed at the listening seat ?

The speakers were not pointed directly at the listening spot, almost flat to the "room". But when they measured close to flat at the listening chair is when the tonal balance was best. If I used the B-K curve it would sound a little dull. In most rooms I find the system sounds too bright unless the curve is used.

I don't know if it was the unusual acoustics that made the difference. There is some writing on the subject that seems to suggest this, but I'll have a hard time finding it again.

 

[LineArray]

...
The speakers were not pointed directly at the listening spot, almost flat to the "room". But when they measured close to flat at the listening chair is when the tonal balance was best.

Hi Pano,

under these conditions - lava cave or not - IMO not much else is shown than "direct sound has to be flat".

Since there must have been a significant drop in highs due to speakers not pointed at the listener ...

 

[Pano]

No, you have to read that again. I measured at the listening position.

 

[LineArray]

Hi Pano,

yes maybe i am misunderstanding the setting.


I will explain my line of thought step by step, so we can see where i might be wrong.

- if your speakers were aligned to be "flat on axis" (freefield response), there should be a drop towards upper highs when listening (say) 20..30 degrees off axis.

- in that setting under question you did not toe in your speakers

- then you equalized - as i understood - for roomcurve (inroom response) at listening seat to be almost flat and this sounded more balanced to you in that very setting than the usual falling roomcurve.


Hopefully, there is no misunderstanding up to here ... ?


Now you placed the question, whether this "exception" was due to the (highly diffusive) lava cave setting, which may not need a falling room curve.

My "objection" here is, that maybe true, but first of all we have to consider also, that in this setting without toe in - in order to achieve flat freefield response in mids and highs at the listening seat - we have to lift the highs, to get the direct sound close to flat anyhow (*): This could bee seen as a goal of high priority or preference.

After having achieved that, getting a "close to flat" roomcurve (too) is more like "ineviteable consequence" (in an environment having only little absorbtion...) because the non toed in speakers are firing most of their uppper highs energy into the room and "beside/off the listener" but not "at the listener".

_______________

(*) Due to rising DI of almost all speakers and off axis drop in highs coming with that.

 

[ua100k]

Recordings with and without reflections

A good demonstration of reflections on a recording.
Track 2: Cello recording of Justin Pearson with 3 mics.
Album: Mike Valentine, Chasing the dragon

Both identical setups (as claimed) done with 3 mics.
One inside the church and the other, ironically in a graveyard.

If this is played without initial description, Audiophiles are mixed as to which they like more. The music engineers and musicians tend to like the outdoor mix more.

The amount of absorption is very speaker dependent, i.e. dependent on how much the speaker throws out in different directions from 500 hz up. Below that is vibrations more then reflections and is less directional.

If the graveyard recording is played in an untreated room, it may start to sound like the recording from inside and then make it more to ones taste but not true to the recording.

If the speaker designer is expecting reflections to boost the sound/power of the speaker, then absorbtion is counter to the speakers performance.

 

[silverprout]

Sorry i can't resist.
My 2cts...

...in your normal listening room...

You should try to record your own sounds with the appropriate technique and high quality mics and preamplifiers (you must be very carefull of the sources distances with XY mics configurations) at the listening spot.

Then read the raw files... at the listening/recording spot and pay attention to the diffuse sound rendering of the imaging, you will notice someting if you are not deaf.

 

[Pano]

Now you placed the question, whether this "exception" was due to the (highly diffusive) lava cave setting, which may not need a falling room curve.

My basic answer to this is "I do not know".
The listening position was about 15 degs off axis on each speaker. Thus approximately 1/2 the normal stereo triangle, which should be 30 degs off is the speakers are flat to the room, right?. On axis, the tweeter had a slight rising response, but at about 15 degrees they were near flat - I don't remember exactly. If a descending room curve like the B-K classic room curve was applied, the tonal balance was too dull. It wasn't horrible, just a little dull. Flat off-axis (at the listening chair) sounded the most balanced. This is the only room that worked that way for me.

Why that is, I do not know. Could it be that the on-axis rising response was causing the off axis response to sound dull? Hard to say. Was it the lack of much reflected energy?

Finding some other evidence of room curves preferred in anechoic or hypo-echoic spaces would be interesting reading.

 

[Pano]

Back to the subject of Phychoacoustics and the Stereo Triangle, I was thinking about an accidental experiment from years ago.

Funny. This reminds me of once when I was running Altec 811 horns on top of open baffle woofers. Seems I had gotten the polarity reversed on the horns without knowing. Wow! What an effect.

Since the windows in house were always open the soundstage went right outside, there was a trumpet player hiding in the bougainvillea out the window on the right, a snare drum in the hall, guitars in the kitchen. But my neighbors in the house behind would not shut up. Kept mumbling, grumbling about something. Until I realized it was not the neighbors but the piano player! Probably Keith Jarrett . Really sounding like he was out the window 30' behind me at the neighbor's house.

As fun as that all was, it sure killed normal vocals recorded right in the center. So the wiring got put back "right".

Why does a polarity flip above ~900Hz throw sounds so far and wide? it's very definitely a stereo "trick", if not an accurate one.