What is "Sound Stage?"

[bblue]

Now it is my time to be sound like a broken record. , but I am curious.

What is the knowledge of acoustics we must have to infer height?

Reverberation characteristics.
But the height inferred is not that of a source placement, but of the performance space volume.

Right. You'd have a knowledge of what an X x Y room would sound like delay-wise, and any element longer than that could (approximately) be inferred. But depending on how close the mics are to the source, it might not even be present enough to tell.

--Bill

 

[Phelonious Ponk]

One of my vintage amp's has a very sophisticated circuit. It was released in 1967, solid state, and came with L/C/R speaker OUT. The L and R channels are completely independent of one another, and phase inverted, and the C channel is created by joining the L and R signals. Due to the design, the three-channel reproduction of two-channel is in correct phase, and adjustable through the use of a separate sound stage controller.

http://www.beolab.dk/box-uk.htm

Pretty advanced for its time, that.

Wouldn't a left and right channel, and a third, center channel that is the sum of the L an R give you ... mono?

Tim

 

[JackD201]

You'd still have the time differences from L and R

Trifield, I think that's what some call it anyway, has a cult following. I've never heard it. Closest I've experienced is plain jane Pro-Logic with the rears off. Given the quality of my C channels in the past it did more harm than good. My impressions might have been very different with identical amplification and speakers to L and R.

 

[Soundproof]

The ideal would be to have access to the center-channel that was recorded in the early decade of stereo, of course, but that wasn't made available to us, in tape or LPs. So a few companies decided to resurrect it from the L/R information on LPs and Tape, and letting their customers have the option of dialing in the weight of the center channel.
Meridian's present day Trifield system does the same, and as Jack writes it gives you the discrete information in the L and R channels (level, time) needed to create the stereo soundfield, while providing the center-grounding that Blumlein was convinced was required for solid (stereo) reproduction, but which the technology of the time wasn't capable of providing.

A few re-issues of well known recordings have provided the 3-channel information on DVD-Audio and Blu-ray, giving us an idea as to what we've been missing out on. I'm presently running that vintage amp in two-channel mode, but have had fun having three identical full-range speakers connected to it.

 

[bblue]

Wouldn't a left and right channel, and a third, center channel that is the sum of the L an R give you ... mono?

Tim

In practice the L+R center channel would be set much lower in level. Just strong enough to give you the effect of a point-source for the center. It helps keep the center information in the center as you move away from an ideal listening spot. Systems that already have a strong and well defined center might be affected negatively in the sweet spot by having the additional center.

--Bill

 

[Soundproof]

Many think stereo means two or two-channel, but it actually means solid, and became solid sound in stereophonic. The need for a center channel for solid reproduction was considered crucial by the engineers who devised stereophonic recordings, but was sacrificed because there was no way of fitting one on consumer systems. The wives didn't want three speakers, I guess.

 

[Phelonious Ponk]

I don't want to sound like a broken record (no pun) but individual microphones, though they can hear the various delays and reverberations in a room, they can NOT tell which direction the sounds come from. They hear all sound as if it comes from directly in front of them.

Additionally, because mics have no ability to identify the direction that information is coming from, they mix the sounds off of the ceiling and the floor, indiscriminately, with sounds reflecting off of the walls.

Place a single mic in a reverberant space and it will pick up direct sounds and reflected sounds and mix them all together with no regard for the direction - left, right, front, back, high, low -- from which the sounds came: Mono. The mics can, however, differentiate distance through volume and reverb content, so...

Place a pair of mics in a reverberant space and they will pick up direct sounds and reflected sounds and mix them together as well, but the sounds coming from a surface or source closer will be louder: Stereo.

Not only do mics not differentiate the direction from which vertical information comes, they mix it up with the lateral information, rendering your stereo sound stage an inaccurate reflection (pun intended) of the recording space.

The fact that all of those distance ques, all of that reverberation from the ceiling and the floor is mixed with the reverberation from the left and right wall, which is mixed with the reverberation from the front and back walls, which is mixed with the secondary and tertiary reflections...none of that prevents us from sitting back in our homes, listening to our tapes and records and CDs and files and imagining the performace space. So it should come as no surprise that we can "hear" a vertical image, even through panels and unconventional speaker arrays. Hell, a nice glass of wine, the right mood, I can lean back in my recliner and hear the performance space....in my headphones.

But I don't come on the internet and absolutely insist that its real.

Tim

 

[Phelonious Ponk]

In practice the L+R center channel would be set much lower in level. Just strong enough to give you the effect of a point-source for the center. It helps keep the center information in the center as you move away from an ideal listening spot. Systems that already have a strong and well defined center might be affected negatively in the sweet spot by having the additional center.

--Bill

I'm still not sure I follow, Bill. It does this without discrete center channel information? I'm trying to imagine that, and at any volume level, even with very directional speakers, you'd still be filling the center with the sum of the left and right. I have no idea how such a system could image at all.

Tim

 

[Soundproof]

I'm still not sure I follow, Bill. It does this without discrete center channel information? I'm trying to imagine that, and at any volume level, even with very directional speakers, you'd still be filling the center with the sum of the left and right. I have no idea how such a system could image at all.

Tim

This is why L/R needs to be phase-discrete in this amp, if not then the sound image would devolve to mono. But with phase discrete L/R, the summing for the center channel does not cancel the peripheral channels. Very neat, actually - when balancing the stereo image I can use a test function where I tweak the L/R against one another, until both are cancelled. The result being that the sound stage is then perfectly balanced when I leave test mode - and very, very wide and large, with excellent and realistic placement of sources. (Achieving this test synchronization requires a mono-signal, btw).

 

[bblue]

I'm still not sure I follow, Bill. It does this without discrete center channel information? I'm trying to imagine that, and at any volume level, even with very directional speakers, you'd still be filling the center with the sum of the left and right. I have no idea how such a system could image at all.

Tim

Well, there is no discrete center channel when your source is stereo, as I'm sure you're aware. So in that case, you still combine L+R and pipe it to the center channel speaker at a lower level. Just enough to enforce the center with some point-source. The difference can be readily be heard and the stereo effect is not significantly reduced.

Ideally, tri-channel would be best recorded as three track, L C R.

Btw, A lot of the RCA Living Stereo SACD's from classical works recorded in the 50's and 60's were originally recorded in three track in a more or less L C R format. Apparently the decision was made to evaluate each stereo mix with and without the center channel mixed in and pick the combination that sounded best. (third hand info)

--Bill

 

[Soundproof]

Lots of Jazz records were also done in L/C/R, the engineers would have it no other way, hoping a commercially viable 3-channel stereophonic format would be developed. While waiting for that, they did 2-channel mixdowns and played 3-channel dubs at home.

With the center-channel adjuster I linked to above you could narrow and widen the soundstage, by balancing the weight of the center channel.

 

[andromedaaudio]

Very interesting i didnt know that , could be great ,having more energy coming from the middle with a third speaker, i m sure there are some here with wilson centers that can comment although the recording might not exaxtly be L/C/R

Lots of Jazz records were also done in L/C/R, the engineers would have it no other way, hoping a commercially viable 3-channel stereophonic format would be developed. While waiting for that, they did 2-channel mixdowns and played 3-channel dubs at home.

With the center-channel adjuster I linked to above you could narrow and widen the soundstage, by balancing the weight of the center channel.

I am sure the salesdepartment of many speakermanufacturers would also embrace it and call it genuine

 

[microstrip]

One aspect that bothers me is that almost all works on sound localization stress the non-reliable characteristics of the localization of sound sources. See a typical statement :

In conclusion, sound source distance judgments based solely on aural cues are unreliable and must rely on a combination of context, loudness, reverberation, timbral, and visual changes that may accompany changes in distance.

This implies that most of our localization capacities in sound reproduction rely on our empirical knowledge of the real event. Can we consider that some persons will not be able to localize a source simply because they do not know how it should sound in that situation - they were not "trained" to perceive it?

Height relative "localization" seems particularly susceptible to this type of previous knowledge. As far as I remember no one ever referred to a recording having clues of incorrect height localization, only to its absence or correctness. I have also found a reference that height localization is easier for slightly moving sources than for fix positions ones.

 

[treitz3]

One aspect that bothers me is that almost all works on sound localization stress the non-reliable characteristics of the localization of sound sources. See a typical statement :

In conclusion, sound source distance judgments based solely on aural cues are unreliable and must rely on a combination of context, loudness, reverberation, timbral, and visual changes that may accompany changes in distance.

This implies that most of our localization capacities in sound reproduction rely on our empirical knowledge of the real event. Can we consider that some persons will not be able to localize a source simply because they do not know how it should sound in that situation - they were not "trained" to perceive it?

Could be.

 

[andromedaaudio]

Well there is one thing i do know about height of the soundstage/ image depth and have tested numerous times .
The serial tweeter cap has to be of high quality ,if i would switch from a good mundorf or duelund to a mkt or whatever the image is merely a blurred center, so a lot of info must be at least above 2,5 khz , it certainly isnt all there is to it but still very important.
sweetspot: tweeter off axis reponse is key to a wide sweetspot in my opinion , if you have a decent tw output on 15/30 degr or 45 degrees of axis the sweetspot will be wide, if not (a lot of folie tweeters ) then small .
a lot of electrostats also have a small sweetspot

 

[Soundproof]

What might have been interesting, if 3-channel had become standard, would have been to dedicate the center-channel to vertical information, by raising this speaker above the L/R speakers.
Depending upon what kind of performance was recorded, one could create a sound stage that had width along a lateral axis, and height in the center. The center-microphones would be dedicated to capturing a portion of the performance that realistically would emanate from this level.
As singers, violins and wind instruments commonly are worked at head-level, this might have created a rewarding reproduction of actual verticality in sound reproduction, rather than today's imagined.

 

[treitz3]

Hello, gentlemen. I have pages of posts to catch back up on but since we are on the topic of 3 channel.....I did have a chance to hear a 3-channel system a couple of years back. I believe it was called a trinaural system but could be wrong, as time takes it's toll on the ol' memory sometimes. What I do remember is that it had a blue faceplate and a bunch of knobs and/or switches and was [according to the cat who brought it] supposed to place good old 2-channel to rest. I'll try to dig up more information as time permits.

What I do distinctly remember is that at times, it sounded great but there was a lobing issue that was rather prominent. The singer, for instance, would not change location on the sound stage but her voice would change location between the three speakers. This not only happened with the vocals but sometimes the instruments as well. It left a very bad impression on these ears. I'm open to another whirl and hopefully it can be all that folks say it is but at that moment in time. With that rig and with that setup?

I wasn't impressed at all. Give me 2 channel stereo and send me on my merry way.

As for the height? Eh, the speakers were all the same height, size, type and supposedly properly placed but the lobing issue had so much of a negative influential factor, I never really paid any more attention to the rest of the aspects of the reproduction. Maybe he had it set up incorrectly, I can't confirm that but he said he did. I was just hoping for something better after all of the hype.

 

[NorthStar]

One of my vintage amp's has a very sophisticated circuit. It was released in 1967, solid state, and came with L/C/R speaker OUT. The L and R channels are completely independent of one another, and phase inverted, and the C channel is created by joining the L and R signals. Due to the design, the three-channel reproduction of two-channel is in correct phase, and adjustable through the use of a separate sound stage controller.

http://www.beolab.dk/box-uk.htm

Pretty advanced for its time, that.

Wow, quite advanced indeed for 1967.

* I first built my own Center REAR speaker back around 1970.
I was also way ahead of my time regarding surround sound envelopment from the rear soundstage.
But today I'm back to the front soundstage. And I like what's possible with Audyssey DSX (Front Width & Height), and also Trinnov Optimizer. :b

 

[RBFC]

Many of the Mercury Living Presence and RCA Living Stereo SACDs allow you to compare the two-channel and three-channel presentations of classic recordings. I have several, and the three-channel option almost always sounds more realistic.

Lee

 

[Johnny Vinyl]

Many of the Mercury Living Presence and RCA Living Stereo SACDs allow you to compare the two-channel and three-channel presentations of classic recordings. I have several, and the three-channel option almost always sounds more realistic.

Lee

Now that I never knew. Wasn't "stereo" at first a 3-channel configuration?