How do we perceive “soundstage” and “imaging”?

[_thelaughingman]

How is it that some reviewers talk about Sound stage and imaging on IEM's?

 

[majingotan]

How is it that some reviewers talk about Sound stage and imaging on IEM's?

They just don't explain that the illusion they hear is due to specific dips in FR curve

 

[audio2design]

Mostly tweeter range

And yet ITD is 200-1.5KHz. Level based is higher frequencies, but tends to tail w.r.t. localization about 6KHz.

 

[audio2design]

How is it that some reviewers talk about Sound stage and imaging on IEM's?

At that point you are just rearranging deck chairs on the Titanic so may as well go all in!

 

[_thelaughingman]

They just don't explain that the illusion they hear is due to specific dips in FR curve

Whats even worse is that they try to make it seem tangible with subjective opinion only to sell their reviews.

 

[Thomas_A]

And yet ITD is 200-1.5KHz. Level based is higher frequencies, but tends to tail w.r.t. localization about 6KHz.

Yes, of course. But when my speaker is close to the wall without damping, it is mostly the tweeter that cause reflections and need damping behind (and with felt vertically). This is audible and decreases detail and imaging. The upper frequencies of the woofer are probably more directional.

 

[Tangband]

Yes, of course. But when my speaker is close to the wall without damping, it is mostly the tweeter that cause reflections and need damping behind (and with felt vertically). This is audible and decreases detail and imaging. The upper frequencies of the woofer are probably more directional.

Very true.
The ”inverse distance law ” also says that sounds and reflections decreases with the longer distance from soundsource to wall, making reflections more noticable if the speaker is very near the wall.
Using deep waveguides and conctant directivity horns can lessen the destructive effect and the need for damping high freq reflections within 2 ms from the source.

 

[audio2design]

Yes, of course. But when my speaker is close to the wall without damping, it is mostly the tweeter that cause reflections and need damping behind (and with felt vertically). This is audible and decreases detail and imaging. The upper frequencies of the woofer are probably more directional.

No, really they are not, not for any normal front firing dynamic speaker. Lots of plots right here on ASR reveal this as will almost any speaker review on stereophile (they typically show plots out to 90 degrees off axis. Most speakers have a fair amount of energy through the ITD range out to 90 degrees of axis rolling off at 90 degrees as the frequency increase.

 

[Thomas_A]

No, really they are not, not for any normal front firing dynamic speaker. Lots of plots right here on ASR reveal this as will almost any speaker review on stereophile (they typically show plots out to 90 degrees off axis. Most speakers have a fair amount of energy through the ITD range out to 90 degrees of axis rolling off at 90 degrees as the frequency increase.

Well my speakers are almnost 45 degree toe in and tweeter is also around 3 cm from top edge.

 

[audio2design]

Well my speakers are almnost 45 degree toe in and tweeter is also around 3 cm from top edge.

And there will still be more power at 45 degrees and sub tweeter frequencies. Either way reflective in the middle is in almost every case worse than diffused or mild broad band (not base) absorption

 

[Thomas_A]

And there will still be more power at 45 degrees and sub tweeter frequencies. Either way reflective in the middle is in almost every case worse than diffused or mild broad band (not base) absorption

So how would you show this ”bad” reflection? ETC from close distance say 50 cm? Or just the geometric first reflection point that reaches the LP?

 

[Thomas_A]

And there will still be more power at 45 degrees and sub tweeter frequencies. Either way reflective in the middle is in almost every case worse than diffused or mild broad band (not base) absorption

So from the pic, how does primary reflection from the centre arrive?

https://www.audioholics.com/room-acoustics/room-reflections-human-adaptation/copy_of_Figure7.12vs3channelfronts.jpg/image

 

[Joeshmoe]

Here is the reply from Moon, for those still interested. They replied right away, the delay in posting entirely mine.

Hey Joe,

Welcome to the world of high fidelity! It’s true that the finer details sometimes get lost in the shuffle, so I doubt you need me to confirm that you’re not just hearing what you want to believe. I have had this discussion with our engineers on numerous occasions, and a well-designed DAC can enable the listener to better perceive localisation cues in an audio playback system.

Whether hearing a genuine sonic event or the playback of a recorded sound, two of the most significant factors in situating sounds in space are their inter-aural stereophony and the delays between the direct and reflected sounds (off walls, floors, ceilings, and any other surface). The delays and intensities of the reflected sounds relative to the originals vary. Reflected sounds that take longer to arrive at the listener’s position are perceived as echoes of the original. Minimally delayed sounds, on the other hand, blend with the direct sounds and allow the brain to process localisation cues.

A well-designed DAC transmits the most accurate signal possible. It’s important to remember that a device like the MOON280D comprises not only a DA stage to convert a digital audio signal to an analog one, but an analog stage also, which delivers the analog audio signal to the outputs of the device. The designs of both these sections influence the performance of the device.

Most the imaging cues that enable the listener to appreciate the soundstage are concealed in low-level information in the audio signal. Most DACs, perform well enough with high-amplitude audio signals, generating relatively little distortion. However, low-amplitude signals are reproduced more precisely by electronic circuits capable of extracting greater resolution from such small signals, and that’s precisely how a better DAC can distinguish itself. It maintains greater linearity and generates less noise, revealing details often buried under the noise floor in some other DACs. The result is that a well-designed DAC can make a difference because, amongst other reasons, it better resolves sound localisation cues and permits a greater appreciation of their relationships.

Of course, no electronic device could ever be perfect, and there will always be a certain amount of signal leaking between the channels. As more of the audio signal becomes common to both left and right channels, the width of the soundstage appears narrower than it would be under more ideal conditions. Better quality parts and clever circuit design can significantly reduce the crosstalk, therefore expanding the perceived width of the soundstage.

I hope that provides at least a basic explanation of some of the reasons why a well-engineered DAC can provide a more elaborate soundstage.

In any case, enjoy!

Ok, so I'll summarize, without belabouring this further:

1. The Moon DAC gives me, subjectively, a bigger soundstage and more detail, which is verifiable, and not merely suggestion.
2. From this forum, it appears there's no known technical reason why it should do so.

I think this is analogous to the bumblebee problem. Bumblebee flight was initially incompatible with known aerodynamics, but they obviously fly. In the early 2000s, however, modern computational fluid dynamics solved bumblebee flight (Young et al., 2009, Science 325:1549-52). I think we're waiting for the science to catch up on this one. The proof of the pudding is in the eating. The Moon unit gives me notably more pleasure from listening to music, which is why I have the sound system in the first place, and so I'm going to shell out the bucks for it.

I am going to contact the folks at Moon to see if they have a technical explanation, and if they do, I'll post it here.

Thanks again to all for the discussion.

 

[tvrgeek]

OK, just did a test. My little desktops are FE-85's with some cheapo domes. I did a quick and dirty crossover years ago so decided to play and learn on them before I redo my HT and stereo crossovers.

So, I modeled a crossover that was very flat, accepting the basket reflections of the Fountek of course. At crossover, the phase miss-match was about 100 degrees. I then tweaked them with APO in situ to be flat with in 1 dB from about 200 up. Sounded pretty good as far as tonal response, but not stereo. Clear right and left. Not stereo. No imaging.

Did another similar, but playing with Q and a few tricks, were within 5 degrees at crossover and not deviating much through the mids.
Again, tweak with APO. ( Peace using TrueRTA) Poof! Stereo imaging. Same bad location. Same too close reflections. Same desktop. My conclusion is crossover phase does matter to out brains.

One can do the research, but back in the 70's I think, AR did extensive testing to see how many sources it took to generate true 3D sound. Answer was 16. The other fun test is with a very very pure tone in a chamber you can't tell if it is in front of you, behind, or anywhere in that plane.

PS: Wish I still had a set of the old Stax Nearphones. Stereo in front, not inside your head.