[YTCrazyTieGuy]

Sorry if there is already a thread about this, but the first 10 that popped up in my search didn't really answer my question and I didn't feel like going through dozens of threads.
It isn't very hard for me to understand how width works - amplitude difference between left & right, though if that was it then why can't low end headphones sound wide?
The thing that I'm most interested in is depth, and how you can perceive it.
From what I read on headroom it has to do with phase, but if it does, how would you explain this?:
 
"You see, the Zana has some wonderful qualities as a top-class tube amp, but also the common weaknesses of tube amps such as resolution and control. The Zana has a very magical midrange tone and a superb three dimensional soundstage with very good depth. The Zana’s soundstage imaging is the best I’ve heard so far among all the headphone amplifiers I’ve tried, though the width of the soundstage is much narrower than the Beta22′s. At the other end, the Beta does have its own strengths as a solid state. Resolution, control, impact, soundstage width, are among the strengths of the Beta22 amp. When I hook up the Beta22 as a power amp after the Zana’s pre-out, what I get is the three dimensional soundstage of the Zana, a little bit of its midrange sweetness, added with the superb Beta22 resolution, technicalities, impact, and control. The soundstage is even more three dimensional than if I had used the Zana alone. The depth is increased, and you also get the wide Beta22 soundstage, while instrument separation is far more distinct. With the Sennheiser HD800 headphone, the soundstage pretty heavenly." (http://www.headfonia.com/pre-and-power-amps-for-headphones/)
 
Thanks!

 

[Uncle Erik]

I'm not sure how much an amp affects soundstage.

You primarily get these depth cues from a good recording. If you haven't heard of binaural recordings, read up. A good binaural recording will do more for soundstage than any amp. Second most important would be the headphones or speakers you use. Driver design and placement are important with headphones. Check out the AKG K-1000 to see how headphones can contibute a lot to soundstage. For speakers, the room itself and speaker placement change the soundstage. If you want to have fun with soundstage, check out dipole speakers, like Magnepans and Quads, and omnipoles like MBLs. They radiate sound very differently from traditional box speakers. Set them up correctly and you can make your listening room sound like the real deal.

Another way to get at soundstage is with multiple speakers. Surround is one way to do it. I've also seen an obscure system that places a speaker on the ceiling to create a vertical soundstage, too.

If you want great soundstage, look for recordings with great mastering. Recordings are all artifical to an extent, but you'll find the well-made ones to sound natural. Next, get headphones or speakers that do soundstage well. I'd put those much farther ahead of the amp. I've run the K-1000 and Quads with good recordings off mediocre/consumer grade amps. They still sounded terrific.

 

[YTCrazyTieGuy]

Thanks for the input, though the thing I want to understand most with this thread is how soundstage works, and not necessarily how to get great soundstage (for now)

 

[eucariote]

Here is a decent introduction to the construction of space & soundstage from interaural phase difference which conveys information about l/r localization along with loudness, spectral information which changes depending on distance traveled, and delay information which gives information about distance in a room.

 

[LFF]

Quote:

Here is a decent introduction to the construction of space & soundstage from interaural phase difference which conveys information about l/r localization along with loudness, spectral information which changes depending on distance traveled, and delay information which gives information about distance in a room.

Beat me to it!

 

[Tyll Hertsens]

I'll add that I think the cues needed for depth are quite subtle, and that I think it takes a good deal of finesse on the part of the amp and cans to get these cues in proper time alignment. Should there be any smearing of phase (group delay, etc) I feel these cues get muddled up and prevent good perception of depth.
 
Total speculation on my part ... backed up with a liberal dose of good sense and experience ... I hope.

 

[YTCrazyTieGuy]

Thanks guys it helped a lot! (sorry I was out for a while).
Now I get that Depth and width are distortion and a small soundstage is the accurate one. (Otherwise how would you explain the Zana+B22 soundstage?)
 

 

[revolink24]

I wouldn't go that far. Sometimes soundstage is the accurate representation of spatial cues, which would definitely be an admirable trait, not distortion.

 

[jcx]

stereo only encodes one spatial dimension - any height/depth in "soundstage" is entirely your brain's internal construction possibly using clues like relative volume, frequency response, inferred notches from floor bounce comb filtering - all given expectations of (recorded) voice, instrument sounds
 
"soundstage" is painted on from (the near universal) close miced multitracked stubs in the mixing room - I really don't think even with today's DSP horsepower that any particular attention is paid to phase/propagation delay effects when "positioning" an instrument feed in the mix - it is primarily amplitude/frequency shaping between the R/L channels
 
pleasing musical experience from 2 channel recordings involves many sonic illusions and some cultural conventions that are learned - we don't really have anything objectively close to live musical event wave fronts in our listening rooms, or from headphones - although Binaural recordings for headphone listening are very interestingly different

 

[YTCrazyTieGuy]

 
Quote:

stereo only encodes one spatial dimension - any height/depth in "soundstage" is entirely your brain's internal construction possibly using clues like relative volume, frequency response, inferred notches from floor bounce comb filtering - all given expectations of (recorded) voice, instrument sounds
 
"soundstage" is painted on from (the near universal) close miced multitracked stubs in the mixing room - I really don't think even with today's DSP horsepower that any particular attention is paid to phase/propagation delay effects when "positioning" an instrument feed in the mix - it is primarily amplitude/frequency shaping between the R/L channels
 
pleasing musical experience from 2 channel recordings involves many sonic illusions and some cultural conventions that are learned - we don't really have anything objectively close to live musical event wave fronts in our listening rooms, or from headphones - although Binaural recordings for headphone listening are very interestingly different

Thanks!
 
Quote:

I wouldn't go that far. Sometimes soundstage is the accurate representation of spatial cues, which would definitely be an admirable trait, not distortion.

your right, I should of said coloration, though I still think soundstage depth is not the accurate presentation. if it was, and the B22 ruins depth with bad phase, than the accurate depth of the Zana wouldn't have been heard through the headphones. Instead, mike (the author of the article I quoted in the OP) heard the Zana depth through the B22, even though he didn't with the B22 alone. This shows that the Zana colored the sound and added depth. (I'm not saying that depth is bad, just that it is coloration.)
 

 

[IpodHappy]

 
Quote:

stereo only encodes one spatial dimension - any height/depth in "soundstage" is entirely your brain's internal construction possibly using clues like relative volume, frequency response, inferred notches from floor bounce comb filtering - all given expectations of (recorded) voice, instrument sounds
 
"soundstage" is painted on from (the near universal) close miced multitracked stubs in the mixing room - I really don't think even with today's DSP horsepower that any particular attention is paid to phase/propagation delay effects when "positioning" an instrument feed in the mix - it is primarily amplitude/frequency shaping between the R/L channels
 
pleasing musical experience from 2 channel recordings involves many sonic illusions and some cultural conventions that are learned - we don't really have anything objectively close to live musical event wave fronts in our listening rooms, or from headphones - although Binaural recordings for headphone listening are very interestingly different

I would add to this that usually a closed mic'ed mix is fed through reverberation effects which simulate the ensemble being in a room of some sort.  The digital signal processing engine of the reverb system feeds the signal is split into multiple delay lines, each representing a virtual "wall" from which echos rebound to the listening position.  Diffusion is a parameter used to adjust the number of reflecting surfaces, their relative left/right position, and their amplitude.  The hall size is set by adjusting the relative length of delay.  The longer the delay, the larger the effective room size.  Parametric eq adjusts the brightness of the room, giving it a darkness or airiness.  Feedback gives the sense of multiple echos, creating a very live or damped room.  The amount of reverb fed back into the mix adjusts how "wet" or "dry" the room seems.  It's really easy to make Grand Canyon effects with five to ten second delays. 
 
The symphony hall in the city where I life adjusts many of these parameters by using curtains to change the dampening and frequency response of the room, and has echo chambers in the ceilings with baffles which can be opened or closed to adjust the decay and delay of the hall.  Ceiling reflectors also move up and down to adjust the apparent space of the hall.  A chamber piece can sound like it's happening in a chamber.  An organ piece can sound like it's in a cathedral.  (During one symphony, a cable in the ceiling piece settled with a firecracker like pop causing the startled orchestra musicians to suddenly look up.)
 
When the dynamics of the reproduction system aren't right, the reverb can seem to fall off too fast, reducing the size of the room.   When a low bit rate conversion have been made, the tails of the reverb can be dropped, again swallowing the depth of the reverb.
 
There's also been a long running (over a hundred years) discussion regarding how high a frequency the ears can hear phase differences from one ear to another.  I sometimes think this controversy has to do with genetics; some can hear the difference while some can't.
 
A simple an easy way to make a mono source sound more 3D is to take the signal and add a slight delay between the left and right channels. 
 
An interesting 3D effect which doesn't work with headphones is to put two slightly detuned signals hard left and hard right in the mix.  With a stereo speaker system, the sound will seem to wash back and forth between the speakers as peaks of constructive and destructive interference at the listener position move back and forth.  This effect has often been used with low frequency square waves to give a "space ship taking off" effect--as you increase the detuning between the left and right channels, the swirling gets faster and faster.  The effect doesn't work with headphones (without a little cross mixing) since each ear hears a separate signal without the constructive and destructive interference.
 

 

[ZeNmAc]

Quote:

 
Thanks!
 
your right, I should of said coloration, though I still think soundstage depth is not the accurate presentation. if it was, and the B22 ruins depth with bad phase, than the accurate depth of the Zana wouldn't have been heard through the headphones. Instead, mike (the author of the article I quoted in the OP) heard the Zana depth through the B22, even though he didn't with the B22 alone. This shows that the Zana colored the sound and added depth. (I'm not saying that depth is bad, just that it is coloration.)
 

Technically that would depend on how it was produced.  I guess you could say that it is coloration (not really the right word) if it has a bigger/smaller soundstage than it was produced to have.
 
I don't see how that is a bad thing though.  I think the people doing the recording expect your speakers/headphones will reproduce the soundstage closer to what it should be.  A tiny soundstage might be true to the recording itself, but a tiny soundstage definitely is not true to what was recorded.
 
Before you go believing some audio review, look at how soundstage actually works.  Soundstage is one of the biggest marketing stunts in audio.  Since no one can really easily measure it, people can get away with all sorts of misinformation (essentially lies), whether deliberate or not, without being accused of false advertising.  Thus 'audiophiles' get a bad reputation for spending a boatload of money chasing after gold placebo equipment that sounds no different in actuality.  The amp is just about the last thing I'd look to for soundstage.  The speakers/headphones themselves are a hundred times more important since soundstage is a physical aspect of sound.  Sometimes you get the impression the soundstage has changed just from a change in frequency response or loudness which is why amps can seem to affect soundstage.  There is also a large placebo factor you have to be careful of.  Of course there are legitimate reasons as other people already said.
 
Anyway, enough ranting, here's what I remember from high school physics...pity none of my college physics classes went into this

.  Hopefully this is accurate since I haven't seen it explained this way yet...
 
I believe depth/distance is perceived by the angle at which the sound waves hit your outer ear (not exactly true, keep reading).  That is very ambiguous, let me explain (it's hard to do without diagrams

).  Consider a small piece of vibrating material.  It creates sound waves that radiate in all directions.
 
The concept is a bit complicated to explain in words, so a simpler example is the outward spreading of a ripple in a pond when you drop a rock in it.  If you're close to the rock, the waves are still spreading out, so the part of the wave that hits you looks like an arc, or a curve.  When you are standing farther away, the arc has spread out more (essentially an arc with a smaller angle).  As you get farther and farther away, eventually that arc starts to look like a straight line.
 
A similar situation occurs with sound waves, only sound waves are longitudinal and they radiate in three dimensions instead of two.  Now go back to the vibrating piece of material.  When you are close, the waves reaching you haven't spread out much yet.  When this hits your outer ear (basically a curved plane, think of it like a small piece of a sphere) it creates reflections which your ear interprets.  Similarly, when you are far away, the sound waves reaching you have spread out more, and are closer to a flat plane hitting your ear.  The reflections in your outer ear are different so they are interpreted by your ear differently.  This is how your ear perceives distance.  Actually, to be more accurate your brain is interpreting and perceiving what is heard by the eardrum.
 
That's essentially what you're looking for as far as soundstage.  In order for your headphones to trick your brain into thinking the sound is coming from farther away, the sound waves have to be essentially flat planes.  In terms of the pond example, if you drew a line where the ripples were, when they hit you the lines should be flat and parallel.  This is why it is so hard to achieve a good soundstage with headphones compared to speakers.  With speakers you have a physical distance and thus a 'real' soundstage that is difficult for headphones to fake because of physical limitations.
 
Of course the delays and such have already been discussed

.

 

[deadlylover]

This is a great read on another take at explaining depth and what not, can't believe no one has linked it yet =P.
darth nut's review on the Stax SR-007

 

[revolink24]

Funnily enough, I was about to come in here and link that.

An interesting experiment on how the pinna (outer ear) affects our perception of soundstage, and also why some headphones have "tilted" drivers, or drivers at different distances. The pinna, or outer ear, is shaped such that when sound enters the ear, small distortions in phase occur which result in audio "cues" which cause us to perceive sound as being spatial. If you have headphones like Grados that can pivot, try tilting them, pointing them at the front and back of your ear, while maintaining the same distance. Also try stretching the headband, moving the drivers farther and closer to you ears. The soundstage changes, does it not? Our ears perceive changes in angle via the pinna as auditory cues, and we perceive distance via changes in the intensity of frequencies (much like a headphone has a frequency response.) Since some frequencies have greater intensity at certain distances than others, it is easy for the mind to "decode" these clues.

 

[matbhuvi]

This is an excellent and long read