do you get any front to back soundstage?

Tony, it's pretty impossible for you to get a holohgraphic image given your room and seating arrangements, but then your setup times and "grooves" as you would say.

I wonder how much of that you would lose if you set the speakers and listening seat up for precise imaging?

 

If you have the..... Chesky sampler & Audiophile test compact disc [9036-80037-2] ......Track11 LEDR tracks test for height... up test, over test, and lateral test. This is followed by a soundstage depth test. Those people who pretend these things don't exist should sit down for five minutes and listen to this cd.....there's also so damn good music of the cd too.

 

I found this ...

http://www.amazon.com/exec/obidos/tg/detail/-/B000003GF3/102-3062118-2266561?v=glance


So there is no excuss not to hear what your own system is doing.

If you go to details you can listen to the tracks !!!

My computer speakers don't do height !....you learn something every day !!!

 

Taste there that man!

As for front-to-back, I've heard it occasionally on both of my rigs, but I'm with Garyi on this one! I've heard this in spades on mates' systems and it doesn't do a lot for me I'm afraid.

Now wrap-around phasing effects, as used on some psy-trance stuff, well, that's different, but most rock and pop won't do it much.

Having said that, I love the bit on Kate Bush's "Don't push your foot on the heartbrake" where you can hear her vocal coming from right back all the way forwards and to the left - it really jumps out. She obviously liked tricks like that early on as she does something similar with footsteps on "All we ever look for".

Also, there's that interesting drum bit on "Funky town" where the drum sounds in the next street on a good system...

All tricks though!

 

Stereo can't do height, for those that understand them look at the equations. Left-right and back forward it does do. Ambisonics is the only system that includes a Z component (confusingly height- should be Y). If you ever get the chance to listen to a decent ambisonics rig do it, they're fantastic fun. You realise you've never actually heard the Z component till you really hear it through ambisonics.

EDIT: All this stuff about tricks! Course its a bloody trick, you have 2 speakers in front of you! Doms example is good, psy-trance 'tricks'. Thats part of the music! Its meant to trip you out, if you have a flat system that won't do any depth then you won't hear the track as intended. I don't see why it matters anyway, ok the sound isn't there but it contains all the cues to tell your brain it is there, if thats the way it was intended to be then thats as much part of the music.

 

Of course stereo can do height. You can only panpot left and right, but that doesn't preclude hearing depth and height. It's all an illusion.

And, FWIW, I'd be intrigued to know how a Naim amp removes this information...

Paul

 

Paul is right, we only have two ears and thats how they sense direction. We 'should' be able re-create any sound faithful as real life in theory.

 

I think you're right that nearfield listening and setting up a system with the room in mind is key to getting a good sound - at least in the case of the average british living room.

If you use mid-field monitors in an appropriately large and set up room I dont see why you should see any less information.

Same with far-field speakers if you happen to live in a concert venue!

 

Stereo is an illusion. The sound appears to come for somewhere other than the loudspeaker - that's an illusion.

If you can manipulate phase to produce 360 degree soundfields, I don't think it's that hard to work out that you can create depth.

Most systems are capable of portraying a semblance of 3D imaging and sharp stereo focus. Older Naim amplifiers are not. Which is right? The majority or the one British manufacturer that bandwidth limits the signal? I don't know the answer, but I personally suspect the bandwidth limiting has a part to play.

 

Chris,

if listening outside of the nearfield, the signal that reaches your ears will most likely have been corrupted in phase and frequency by the room boundaries, which has a negative effect on soundstage. Hence Thorsten's love of controlled dispersion loudspeakers (which you have as well!)

 

My Mirages give front to back, sides and height, partly due to the drivers being spaced apart. Very BIG image. Hate that flat earth stuff from Naim, sounds really small, like all the instruments have been squished between two large sheets of metal.

 

Theres a difference between pure in your mind illusion that can happen on any system and the controlled illusion of stereo imaging and ambisonics etc. The stereo equations do not contain a height element and only allow you to alter the intensities of the L and R channels. Ambisonics equations contain a Z element allowing you to control the height of a source using the appropriate replay gear.

Tenson: Not with stereo you can't its not as simple as 'we have two ears...' To truely achieve 3D sound you either have to use more speakers or take into account the listener using the correct transfer functions.

 

Do I ? thats cool then

My thoughts on mid-field listening though, surely you should do this only in a room which is sufficiently large to minimise room interaction, and if you're rooms too small then you should be listening nearfield anyway?

My point is, that if you're listening to mid-field monitors in a sufficiently large room, then room interaction shouldnt be such a negative factor.

Most british living rooms that I've seen obligate the listener to NFM anyway.

I have little experience of listening mid-field, so this is just theorising on my part.

One example I'm thinking of is a live concert venue. You'll be listening at far field (unless you like getting squashed at the front), but no apparent lack of anything in particular in the sound..

 

Stereo can do width and depth, but not height. By varying L/R volume and L/R phase differences (either added artificially or naturally present in an acoustic recording) the illusion that a sound is coming from one side or the other can be created. Reverb and other cues (again, either artificial or naturally present) can give depth information but the way in which humans perceive the height of a sound actually depends on the direction the sound is coming from and it's not possible to encode this in a stereo recording in order to create the illusion of height like we can with width and depth.

In order to create a true height illusion you'd need at least one extra channel at a different height to L+R and it would get very complicated.

Michael.

 

How do our ears detect the 'height' of a sound then? We only have two transducers in our head and with a direct stimulation of these it is possible to create the illusion of height. I did say IN THEORY. In practice the transfer method is of course very important and most are too cumbersome to manage the height trick. I would think the closest thing to it would be a binaural recording played through some in-ear headphones like the ETRmonic ones.

 

Height information is provided by the shape of our ears. If a sound of fairly high frequency arrives from the front, a small amount of energy is reflected from the back edge of the ear lobe. This reflection is out of phase for one specific frequency, so a notch is produced in the spectrum. The elongated shape of the lobe causes the notch frequency to vary with the vertical angle of incidence, and we can interpret that effect as height. Height detection is not good for sounds originating to the side or back, or lacking high frequency content.

Therefore, unless a sound is actually coming from a different height (which it can't be from stereo speakers at the same height) it's impossible for us to determine it's height.

Michael.

 

Its not in theory, without including the HRTF, as Michael said, you can't include a height cue unless you have more speakers at different heights like ambisonics.
If you wanted to you could make a dummy head recording but the ears would more than likely have to be moulds of your own to get convincing height.
Binaural encoding is simple enough, if you have access to matlab its dead easy, making the recordings is harder though as you need a dummy head. Which is expensive.

The other problem with height and behind the head is your localization ability changes around the head, through the max confusion zones etc. Behind the head localization ability drops to about 10degrees accuracy I think, where as its 2-3 at the front. Without a strong visual cue it can be difficult to localize things elsewhere which can often seem like a failing of the system.

 

Hmm okay so you could do it from a stereo source, but you would need to know how your ears effect the sound when it comes from different directions to include that informationa in the recording and as everyone has different ears that is pretty hard.

 

You're missing the point.

Stereo provides two sound sources, the 'image' is partly an artifact of rough control of the volume level arriving at each ear, but also a willingness to participate in the illusion by the listener. Does your image collapse when you turn your head? Does it work at both ends of the sofa?

Height and depth come from the acoustic signature of each source, it sounds like it's close to the floor, that's how we hear it etc.

And of course if we determine height by interference notching the treble then there's no reason why we wouldn't hear a pre-notched signal as coming from higher than the source.

All IMOaE etc etc.

Paul

 

Paul,

my experiments show phase is more important than volume level, that the signal arriving at the correct time is essential to the creation of a credible image. I always assumed it was volume related too, but whilst that obviously has an impact, it was less than I m imagined.