what effects soundstage depth

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Putting aside the recording and room acoustics, What factors in a speaker design determine sound stage depth.

Will certain changes to specific aspects of a design make the sound stage more forward or recessed or is this just a result of everything working correctly together and reproducing whats recorded.

I ask because I have a speaker that seems to suck the life out of vocals and the singer always seems to be recessed in space. The frequency response is flat when measured close up.

That said, I don't really want to trouble shoot this particular speaker. I'm more interested in better understanding speaker imaging.


Thanks


David
 
By what I know I come to the conclusion:

Good off-axes for wide stereo image.
Good impulse behaviour so everything is very detailed to hear with out any muddy sound.

The placing is the result of the difference in loudness between left and right channel and phase.

Then the reflections of the listen room play also their roll, by interfering the stereo image.


Your speaker and listening room messes with the phase and impulse response and loudness of the original recording.


Regards Helmuth
 
It helps to understand how people mic and mix for depth effects.

Basically as a sound gets closer to you the bass gradually increases (proximity effect) and there is less reverb. As a sound gets further away it has gradually less bass and more reverb.

High end is effected as well but it's kind of tricky. But basically the same idea if there is a lot of high end it will sound close the more you roll it off it will give the illusion of depth.

So if you have a properly setup stereo system (equilateral triangle) and you take a mono signal add some reverb and roll off the bass a bit it should push back in distance wel beyond the wall behind the speakers.

If I was trying to make a playback system that would expose the depth in recordings it would probably have to de-master most CDs with some light filtering. Bring 5kHz down about 3dB with a wide band EQ and maybe bring the bass down a little. THis is just based on odds and the way I have found most CDs being mastered and mixed. It will be different if your system isn't somewhat flat.

But as far as what needs to be there I think a wide dispersion/off axis reponse is very important. Probably second would be phase relationship in the speakers and crossovers.
 
Thanks for the replies.

That bit about the recording was interesting.

As I think about it I suspect something else is at play in the recessed voices. I need to make more thorough measurements. Could be related to the transmission lines I built. I'm going to try them in ported cabinets for comparison sake.


D
 
frugal-phile™
Joined 2001
Paid Member
1st it has to be on the recording...

a number of things in no specific order...

1/ downward dynamic range. The speaker has to have the ability to reproduce -- even in the precense of a much streonger signal, all the subtle peices of information that carry imaging/soundstaging clues. Think 40 dB down. This means everthing must be quiet -- the box, the cone, diffraction, internal reflections.

2/ the harmonics need to stay within the envelope of the fundemental. Good impulse, good phase response.

3/ meantioned already. speaker placement in the room

What are your speakers?

dave
 
I would actually argue that room acoustics aren't exactly important or as important as people and measurements would lead you to believe. I think as long as you get rid of early reflections and give a decent amount of breathing room for your speakers you can effectively blank out the rooms effect with your mind. Sort of like looking directly into a flashlight - you can not see the reflected light outside of the source because your senses are being flooded by the source itself. I think surround sound is better at this for obvious coverage reasons.
 
planet10 said:
1st it has to be on the recording...

a number of things in no specific order...

1/ downward dynamic range. The speaker has to have the ability to reproduce -- even in the precense of a much streonger signal, all the subtle peices of information that carry imaging/soundstaging clues. Think 40 dB down. This means everthing must be quiet -- the box, the cone, diffraction, internal reflections.

2/ the harmonics need to stay within the envelope of the fundemental. Good impulse, good phase response.

3/ meantioned already. speaker placement in the room

What are your speakers?

dave

I agree with this, but would point more strongly at phase and downward dynamic range. I believe these are the two most important factors in getting that hair raising imaging effect from a good recording.
 
Key said:
I would actually argue that room acoustics aren't exactly important or as important as people and measurements would lead you to believe. I think as long as you get rid of early reflections and give a decent amount of breathing room for your speakers you can effectively blank out the rooms effect with your mind. Sort of like looking directly into a flashlight - you can not see the reflected light outside of the source because your senses are being flooded by the source itself. I think surround sound is better at this for obvious coverage reasons.

Getting rid of early reflections and positioning the speakers in adequate space IS the heart of room acoustics!

-David
 
Dave,

the speakers are an RB-kit in a transmission line.

I used Martin Kings alignment tables to arrive at a length of 39 inch and went with a 10 to 1 taper. The drivers are about two to four inches higher than recommended. The stuffing is somewhat on the light side since I found the sound to be a bit closed with more stuffing. This is the first TL I've built so I'm not really sure how they could effect the sound if there is a mistake in the calculations or execution.


D
 
Polar pattern, room acoustics. All else is minor, assuming (as Dave points out) that the recording is well-made in that respect.
....by SY

Simply said but more complex in application. But I think this sums up most things 'simply'.

Having spent many years chasing the Holy Grail for sound , I think there is no 'ultimate response' possible to offer to mankind in general. Why do you think so many people ( as seen on this and other forums ) keep changing their systems so many times ?
There is NO single/multiple 'performance parameter(s)' that is acceptable to all ! Can reproduced sound actually get better and better every year for over 50 years ( for example ). That means every year the reviewers still haven't figured out what IS the ultimate 'real' home performance.
Many people like to listen to what THEY think sounds real ( or just what they LIKE ). That has many interpretations. It's good , as it keeps a great hi-fi business afloat !
I know many people who will spend oodles of money on systems they 'think' sounds best to them, though it might be something I don't like at all .
The business is now more about catering to the fluctuating taste of the hi-fi public. Isn't it ?
We will never reach a state where knowledgeable people or not will find one FINAL design solution that will be deemed to be 'accurate live sound' by all. No it can't happen! We are finiky Humans after all. It's great for the Hi-fi business!
:D

Edit: Sorry my blurb is off topic !;) Didn't realise it when I typed it. I'm in a 'bubbly mood' and about to hit the sack tonight !:D
 
frugal-phile™
Joined 2001
Paid Member
kifeinthesink said:
the speakers are an RB-kit in a transmission line.

I used Martin Kings alignment tables to arrive at a length of 39 inch and went with a 10 to 1 taper. The drivers are about two to four inches higher than recommended. The stuffing is somewhat on the light side since I found the sound to be a bit closed with more stuffing. This is the first TL I've built so I'm not really sure how they could effect the sound if there is a mistake in the calculations or execution.

What did you build them out of and how well braced are they? Drivers rebated? Square edges on the front of the box.

dave
 
half inch MDF, no bracing aside from the taper of the line, the tweeter is not recessed, and there are no roundovers.

Guess that says a lot.

I didn't try to finesse these since I wasn't sure what to expect from my TL calculations and I plan to make nicer cabinets once I know I'm happy with the design.

I can try round overs and recess the tweeter, and I can fill the back chamber with sand.


D
 
frugal-phile™
Joined 2001
Paid Member
kifeinthesink said:
half inch MDF, no bracing aside from the taper of the line, the tweeter is not recessed, and there are no roundovers.

If the line is folded that is a fair amount of side-to-side bracing. 1/2" MDF is like using 6mm of plywood in terms of stiffness, and will ooze a fair amount of low-lwvwl grunge which will tend to bury imaging clues.

Get yourself a mechanics stethescope and have a listen to the box while it is playing.

dave
 
Member
Joined 2004
Paid Member
I've found there is very good imaging in rooms where there is very good L-R geometric symmetry with the speakers, room boundaries, windows (and glass doors), and large furnishings.

For good soundstage depth, I believe speakers need to come out into the room plus have no furnishings or CRT on the front wall between the speakers.

(But the image and depth need to be in the recording, and sadly the better my speakers and room have gotten, the less often it isn't there. But when it's there....!)
 
frugal-phile™
Joined 2001
Paid Member
Tosh said:
very good L-R geometric symmetry with the speakers, room boundaries, windows (and glass doors), and large furnishings.

Not my room

I believe speakers need to come out into the room

Maybe 1/2-1 m out

no furnishings or CRT on the front wall between the speakers.

nope.

Get the right speakers in and it sounds like that whole end of the room has disappeared...

dave
 
At one time I was interesting in writing a software library for simulating directional audio in games in the like. I did a survey of the literature, and while I've forgotten most of it and lost my notes, I remember there are *at least* 11 separate mechanisms by which we perceive the directionality of sound.

In the case of speakers, add to this the speakers interaction with the room, and things are very complicated indeed.

There is a really interesting technical solution however, and one that doesn't require repeated format changes like stereo to 5.1 to 7.1 and so on, but it would require the industry to change from discrete representations to spherical harmonics used in a scalable number of channels. Even if this doesn't happen, as the number of channels moves up I suspect some clever company will do some dsp processing of the 10.2 (or whatever) format signal to try to reconstruct a soundfield and then play it back on an arbitrary number of speakers with largely arbitrary location.
 
If you can remember any of the 11 I'd like to hear them. I personally think phase relation of sounds is top priority when it comes to perceived directionality.

Heh my system that I have come up with is exactly like what you are talking about. It's easier to explain visually and with an audio demonstration but basically take ideal stereo and define it as an equilateral triangle. Then if you want two more channels (quad) add two more triangles. If you want more coverage then add two more speakers making 6 triangles arranged as a hexagon within a circle. It's fairly simple actually and it's not that far off from what I have seen old pros from the 70s say they used.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.