The skills of engineers and producers of multi-channel program appear to be about at the level of early "ping-pong" stereo, but even at its best, the limitations inherent in the technology itself preclude realization of the objective of moving a virtual performance into our listening spaces.
Whatever spatial cues are present in two-channel recordings do not rely upon sidewall reflections for their effective reproduction. To the extent that having them "enhances" the apparent source width, the effect is pleasant, but also deleterious to imaging and spectral quality.
Studies indicate that given the ability to adjust reflections to increase spaciousness, listeners will dial them to extremes far in excess of any natural occurrence, and call the result "Good," artificial and inaccurate as it may be. Think Bose 901.
In this sense, soundstage and imaging are somewhat mutually exclusive, and I believe Geddes in his Summa paper, and Toole in his new book outline the psychoacoustics well. Cue extraction shows potential for providing us with the best of both....
Whatever spatial cues are present in two-channel recordings do not rely upon sidewall reflections for their effective reproduction. To the extent that having them "enhances" the apparent source width, the effect is pleasant, but also deleterious to imaging and spectral quality.
Studies indicate that given the ability to adjust reflections to increase spaciousness, listeners will dial them to extremes far in excess of any natural occurrence, and call the result "Good," artificial and inaccurate as it may be. Think Bose 901.
In this sense, soundstage and imaging are somewhat mutually exclusive, and I believe Geddes in his Summa paper, and Toole in his new book outline the psychoacoustics well. Cue extraction shows potential for providing us with the best of both....
markus76 said:
Ascertaining those soundfield cues which are responsible for the perception, isolate them, and manage them. Research indicates what cues relate to enhanced ASW, apparent source width, and suggests that LEV, listener envelopment, cannot be achieved in small spaces such as our listening rooms except via surrounds implementing the appropriate delays. Please read Toole's book; I can't reproduce the illustrations without violating copyright.
The alignment Earl recommends incorporates some of these teachings: minimize the early first ipsilateral (same side) reflection via constant directivity, and emphasize the later low IACC, interaural cross-correlation, contralateral one. The text and diagram in his Summa paper contrasts the behavior of omni vs. CD, and you'll also find the 22.5° rationale presented here:
http://www.gedlee.com/downloads/Cum laude.pdf
ZilchLab said:The skills of engineers and producers of multi-channel program appear to be about at the level of early "ping-pong" stereo, but even at its best, the limitations inherent in the technology itself preclude realization of the objective of moving a virtual performance into our listening spaces.
Whatever spatial cues are present in two-channel recordings do not rely upon sidewall reflections for their effective reproduction. To the extent that having them "enhances" the apparent source width, the effect is pleasant, but also deleterious to imaging and spectral quality.
Studies indicate that given the ability to adjust reflections to increase spaciousness, listeners will dial them to extremes far in excess of any natural occurrence, and call the result "Good," artificial and inaccurate as it may be. Think Bose 901.
In this sense, soundstage and imaging are somewhat mutually exclusive, and I believe Geddes in his Summa paper, and Toole in his new book outline the psychoacoustics well. Cue extraction shows potential for providing us with the best of both....
What you are discussing here is rather different then what I was.
Note that I'm not suggesting that you should increase *listener* room reflections in an attempt to achieve better soundstaging or imaging. (..that was more the adding "ambiance channels camp".) But I *would* suggest than an increase to listener room reflections *may* not be quite as harmful to either imaging or soundstaging as others would argue.
Soundstaging is not an artifact of the listener's room. It's an artifact of the venue where the recording takes place (or is processed). It is integral to the performance, both artistically and in respect to image localization and as such is not at all mutually exclusive to imaging.
ZilchLab said:
..and it's quite incorrect.
Anyone who has heard good stereo reproduction could tell you the same.
Listener envelopment *can*, and often is achieved via good *2* channel reproduction. It's part of the reason people pay so much money for audio equipment or spend so much time in forums like these (..trying to find a way to do it for less).
You guys are mixing things up. LEV can't be achieved with 2 channel stereo. LEV is a property of a diffuse (not a property of acoustically small rooms!) soundfield >80ms. There's no way to recreate these properties with 2 speakers. ASW can be created with strong lateral reflections but you sacrifice localisation by doing so. Furthermore the increase in ASW is not controlled by the recording. It's a incalculable property of the speaker–room interaction. In the end it's like sounding a speaker because "I like it better". That's no longer accurate sound reproduction that is true to the original (the recording).
Extreme toe-in of speakers has nothing to do with LEV or ASW (indirectly it can). It introduces trading effects when the listener moves to the left or the the right. But this effect is too incalculable because trading is signal dependent. In addition I suspect thresholds vary a great deal from person to person too.
Best, Markus
Extreme toe-in of speakers has nothing to do with LEV or ASW (indirectly it can). It introduces trading effects when the listener moves to the left or the the right. But this effect is too incalculable because trading is signal dependent. In addition I suspect thresholds vary a great deal from person to person too.
Best, Markus
Well, imho, if the speakers fire straight out, you usually get a not very good stereo image... and you are way off axis for the tweeter (which on some speakers might be a design benefit?). So that brings us to some sort of triangulation set up. The next question is if you are on axis, or alternately the speakers cross in front or behind you. In my room you can choose to sit in all three positions if you wish... different presentation of the soundfield.
Sort of like being on tracks and you can roll toward or away from the stage... that's all.
The issue of off center line imaging for your friend(s) who are visiting is tougher.
We should consider the Beveridge setup and what the Ohm F speakers do about this. (Heresy, both).
As far as the room, I'd say that 99.9% of us are in extremely handicapped rooms - way way too small to do justice. But then there is the converse problem that of the speakers being way way too small to deal with a way way bigger room. So, that leaves us with absorbing and diffusing reflections, which is not the same at all as a much longer natural return time (absorbed or diffused or not). If you want to hear things to good advantage I am afraid you and I need to build or buy a room the size of a small barn, literally.
Pick your compromises, optimize as much as possible and then you have to live with it...
_-_-bear
PS. the last guy to do something genuinely interesting with stereo was probably Jimi Hendrix on Electric Ladyland...
Sort of like being on tracks and you can roll toward or away from the stage... that's all.
The issue of off center line imaging for your friend(s) who are visiting is tougher.
We should consider the Beveridge setup and what the Ohm F speakers do about this. (Heresy, both).
As far as the room, I'd say that 99.9% of us are in extremely handicapped rooms - way way too small to do justice. But then there is the converse problem that of the speakers being way way too small to deal with a way way bigger room. So, that leaves us with absorbing and diffusing reflections, which is not the same at all as a much longer natural return time (absorbed or diffused or not). If you want to hear things to good advantage I am afraid you and I need to build or buy a room the size of a small barn, literally.
Pick your compromises, optimize as much as possible and then you have to live with it...
_-_-bear
PS. the last guy to do something genuinely interesting with stereo was probably Jimi Hendrix on Electric Ladyland...
markus76 said:
Listener envelopment *can* be achieved in a small room with regard to *reproduction*. It can't be with regard to *creation*.
If late arrival reflections are part of the soundtrack and your playback system is capable of reproducing them well, then you can be "enveloped" in sound.
In the very best systems the small-room acoustic of the listener just seems to be "dissolved", with the recorded venue in it's place. Even merely good systems can minimize the perception of the listener's room, achieving something in-between the listener's room and the recorded venue.
ScottG said:
That's not true for 2 channel reproduction. Only multichannel reproduction is capable of providing LEV in a psychoacoustically correct way. The LEV you might experience in your listening room is that of 2 sound sources playing in your room. That is different from the LEV you would perceive in the recording space. You might want to take a look at http://www.aes.org/e-lib/download.cfm?ID=13686&name=harman
Best, Markus
markus76 said:
An interesting paper, thanks!
However, if you are going to suggest a reference for your position - please describe *where* support can be found within that reference.
Moreover your argument is quite different from my own. Mine specifically is:
"If late arrival reflections are part of the soundtrack and your playback system is capable of reproducing them well, then you can be "enveloped" in sound."
This statement *is* true.
I have not stated that the reproduction is necessarily "psycho-acoustically correct". Hell, you won't even find anything on that paper that claims multi-channel, even at its very best, will be "psycho-acoustically correct". More pleasurable perhaps, but not "psycho-acoustically correct".
In fact typically recorded mult-channel is a disaster. At its very best it is preferably to NOT have multi-channel *encoded* and then reproduced, but rather to have an acoustic model for the exact space where the recoding was recorded and then derive the sound via multi-channel from that model.
The LEV you obtain from a pair of loudspeakers is most assuredly different than what you would obtain by actually being in the space during the recording (..if you could).
This does *NOT* however preclude either:
1. That LEV isn't obtained to some degree with a "good" recording and a decent domestic system.
2. That you can't, in the very best systems, come relatively close to what perhaps would be considered "psycho-acoustically correct".*
*(..note of course that the perspective is always shifted, so technically as far as I know there is really no such thing as "psycho-acoustically correct".)
btw, this is a tangent to this thread - I'll not reply here again.
ScottG,
wrong claims and incorrect conclusions in your last post. I suggest you to read Toole's "Sound Reproduction". There you'll find an extensive list of references with relevant studies. Please read up what LEV is and what it is not. LEV is a property of the late arriving soundfield. The soundfield in acoustically small rooms with 2 speakers is completely different. Only delayed sounds from additional loudspeakers are capable of introducing the sensation of LEV. If its "real" or "good" enough is another question. But at least we should use the word LEV the way it's defined. There's no second meaning of LEV that would support your claims.
Best, Markus
wrong claims and incorrect conclusions in your last post. I suggest you to read Toole's "Sound Reproduction". There you'll find an extensive list of references with relevant studies. Please read up what LEV is and what it is not. LEV is a property of the late arriving soundfield. The soundfield in acoustically small rooms with 2 speakers is completely different. Only delayed sounds from additional loudspeakers are capable of introducing the sensation of LEV. If its "real" or "good" enough is another question. But at least we should use the word LEV the way it's defined. There's no second meaning of LEV that would support your claims.
Best, Markus
ScottG said:
From my very limited experience, the old Technics and Yamaha ambience processors, that predated Dolby Digital, are *much* more believable sounding than any multi-channel recording I've heard. Especially when mated with ambiophonic decorrelated front channels. That experience is the by far and away the most realistic sounding reproduction I've heard.
If I could practically setup a similar listening environment in my home, I would. The increase in believability is an order of magnitude greater than the differences in good and better stereo reproduction IMHO.
The Yamaha DSP-1 is what I like. They tend to go on eBay for about $100US. They come up often. Be SURE to get one with the remote - they are useless without it. I even bought a spare remote, just in case.
You will need 4 channels of amplification to use it properly. You place 2 speakers behind your mains and two behind you. These add a huge amount of depth and ambiance tho they play at low levels. For most of the normal hall effects, bandwidth is limited to 7 KHz, so no need for super tweetes here! Some of the more crazy effects go higher.
There were some later models that had amplification built in. I don't know these, so can't comment. Don't know if thay had all the bells & whistles of the DSP-1.
If you have some extra speakers and amp channels, it's a great boost to the acoustics and sense of being there. Very natural sounding - Yamaha did a great job. And you can always turn it off if you don't want it.
Lexicon also made one, but I have not used it. I believe Mr. Linkwitz does.
You will need 4 channels of amplification to use it properly. You place 2 speakers behind your mains and two behind you. These add a huge amount of depth and ambiance tho they play at low levels. For most of the normal hall effects, bandwidth is limited to 7 KHz, so no need for super tweetes here! Some of the more crazy effects go higher.
There were some later models that had amplification built in. I don't know these, so can't comment. Don't know if thay had all the bells & whistles of the DSP-1.
If you have some extra speakers and amp channels, it's a great boost to the acoustics and sense of being there. Very natural sounding - Yamaha did a great job. And you can always turn it off if you don't want it.
Lexicon also made one, but I have not used it. I believe Mr. Linkwitz does.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.