|
Posted by Romy the Cat on
03-27-2013
|
|
fiogf49gjkf0d A few days back I visited an audio person who demonstrated to me his experiments with extreme lateral deviation of souses. Regardless if I agree or not what he was doing or trying to do the demonstration has an unexpected result.
One day I was driving home from work and I asked myself: could the imaging that we get from out stereo to be improved but in different ways? The key for my thinking was my familiarity with Lamm L1-L2 preamps that did some amassing imaging trick the no other preamp could approach. Do not get me wrong, there are much more interesting preamps, my current is one of them but not of the preamps known to me do that imaging trick. I use to write about it in past.
So, if you remember I proposed that Lamm in his L1 ad L2 preamps does some intentional or not intentional cross-injection of phases, mixing some signal from one channel to signal with other. It is not what I know or something that Lamm told me – that is pr my speculation as it has similar effect with sound I heard when cross-phasing was done in headphone amplifiers.
Looking what my lately visited person did and considering some results I got from Lamm electronics I wonder: of t would be interesting to introduce lateral phase cross-injection in Macondo. Macondo is phenomenally easy for such experiments as it has multi-amping and to do it very simple. All that I need to do it take 4-5 itch MF driver, attach it to frame and source it from another channel. I do a pair of small monitors the I can use for that purposes.
For sure such of acoustic cross-channel injection will kill something and here is the key (I think) for the whole application. The cross-channel injection need to be very smart cross-phase injection what the injection needs to be very limited by band and very precisely dialed by amplitude. I think I will start with 300-2000z hand and with minuses 12dB. The time alignment of such injection is also might be the factor but I have no opinion about it as now.
So, I declare the Club’s open projects. I have outlined the idea, feel free to experiment and to post your results. I presume that somewhere in during experiment one of us might (or might not) come across a configuration that might take the whole notion of imaging, instruments positioning, distance to the first reflective boundaries to a very new dimensions.
The Cat
|
|
|
|
Posted by Paul S on
03-27-2013
|
|
fiogf49gjkf0d I don't think this is what you had in mind, Romy, but I do find it interesting in the context of this thread. It is the late James Bongiorno's (not un-complicated) idea for "correct" playback. I spoke with him about it a couple of times. JB said there are lots of problems with the way regular stereo is "allocated" and processed during the recording process, and he adamently insisted his device is the only way to correct the problems and correctly present the acoustic space, etc. There are "reviews" of the item that are linked from the SST site. One thing that came up (unexpectedly) in a review was that CD and LP were not affected "the same" when processed through this device. ???
http://www.ampzilla2000.com/trinaural_manual.html
Best regards,
Paul S
|
|
|
|
Posted by Paul S on
03-31-2013
|
|
fiogf49gjkf0d Here is an active (I think), "analog" (they specify) product that is said by its promoters to develop something good by way of "phase" information that otherwise remains buried in recordings. While the "specifications" and marketing are vague (to say the least), the promoters do offer some sort of "nearly no risk" guarantee (qv) to encourage curious punters. Since most specs and hyperbolic claims are bullshit to begin with, their absence alone should not troubling.
http://www.bsgt.com/wp-content/uploads/2011/05/qol_signal_completion_stage.pdf
Scant information here; mostly a sort of fishy smell.
http://www.bsgt.com/wp-content/uploads/2011/05/qøl™_owners_manual2012_oct.pdf
The owner's manual is where the phase-related idea is literally espoused.
Maybe someone who ponied up for a PP would also take a calculated risk on this and report back?
Paul S
|
|
|
|
Posted by Romy the Cat on
03-31-2013
|
|
fiogf49gjkf0d Thank, Paul, interesting links. What is necessary here is to do some critical listening with the idea proposed above. It would not e any measurements but just subjective listening. I am not a huge to of do it on line level and I think the speaker level with dedicated amplification is more elegant way to go. It might be played not only with amplitude cross-injection but with injection of difference between righ and left channel. You can bridge the plus outputs of right and left amps and drive from it the injection channels – it will have the delta signal, you can revise the polarity for another channels and you have ready to go test environment. Just add a high impedance volume control in order do not screw ups the signal to the main channels and you are in game. I wish I could experiment with it but nowadays I have too much on my plate, someday I certainly will look into it.
|
|
|
|
Posted by JJ Triode on
03-31-2013
|
|
fiogf49gjkf0d The old Dynaquad system involved using the L-R difference signal, but produced at speaker level across a pair of rear speakers (L-R in one and R-L in the other.)
Decware has one amp (the "Taboo") that can be operated in what they call "lucid mode" where there is cross-feedback between L and R channels. I guess this is not an option if you are not using feedback to begin with.
There are some tricks with room-acoustic treatments that also enable the user to send some reflected sound diagonally across the listening room to the opposite side.
All these things affect imaging or "the sense of acoustic space" but how to use them to serve the purposes of the music is of course the real and hard question.
|
|
|
|
Posted by Paul S on
03-31-2013
|
|
fiogf49gjkf0d The fact is, typical "stereo" L and R separation is mostly an engineering hodge-podge to begin with, and any further messing with it will be very much settled by ear, whether at line or at speaker level. Yet James Bongiorno set out to "solve" the problems with a practical application of his own, proprietary, theoretical solution to the mess he insisted was/is inherent in "stereo" recording and playback. And "phase issues" are what he meant to address.
The "theory" is confusing, anyway, whether adding "parts" of left to "parts" of right, and vice-versa, or "simply" adding together/blending the left and right channels to make a center channel. I have planned for some time to have a main, multi-driver center channel, and I even have a 3rd MA-9S2 sitting there, ready to drive it, and I will do this sooner or later. Meanwhile, I have been told repeatedly by folks who should know that there is more to summing left and right channels "correctly" than just tapping a channel and/or strapping leads together, whether at line or at speaker level. In fact, I first approached James Bongiorno about designing a transformer-based stereo channel sum-er. But by then B was already convinced that he had discovered the only viable answer to this in his Trinaural processor, and he literally would not "waste [my] time and money" on a "non-solution" like I wanted. He did offer to build a Trinaural processor for me, and he also offered me a money-back guarantee; but at the time it just seemed like "too much" to me.
Sadly, although I have listened intently to all the experts, I have yet to hear a simple explanation for the "channel thing" that I really understand. Apparently, generic stereo phase "issues" make any sort of summing mix-and-match a fairly wild hit-and-miss proposition, due to electrical and (ultimately) audible summing and cancellation that might or might not be wholly predicted or predictable (unless you're James Bongiorno...). Naturally, I won't let reality nor concerns about it stop me from trying a center channel, and of course I will tell all here, apropos.
Paul S
|
|
|
|
Posted by zztop7 on
03-31-2013
|
|
fiogf49gjkf0d Paul S wrote:"typical "stereo" L and R separation is mostly an engineering hodge-podge". I do not disagree with that statement. I do feel the major advantage of stereo is not "stereo", but the separation of parts of the information in the signal to the drivers. Therefore, less information is dumped on each driver. If a driver has 50 pieces of info. to handle in a given time vs. 100; the 50 pieces should be handled more cleanly and accurately. So the analogy would be MONO= 2 drivers each handling the same 100 pieces /// STEREO = 2 drivers each handling 50 pieces of the total signal of 100. [I do know there is varying degrees of overlap, and 50/50 is not a perfect real world scenario.] So much of Romy's research & work has been based on dividing up the signal to where it belongs. Each driver only doing its' specialized area with the correct pieces of info. Contributors please correct anything wrong in the above statement. Best to all, zz
|
|
|
|
Posted by Paul S on
04-01-2013
|
|
fiogf49gjkf0d ZZ, clearly, no single driver or pair of drivers can reproduce the range and power of an orchestra, and each of us deals with this according to our own personal demands and resources. The concern of this thread is trying to juggle/work with "phase" to induce/produce better playback, and for Romy, multiple, DSET "channels" are a given. Since you brought it up, it is all the more ironic in this context that multiple drivers often exacerbate "phase issues" even as they allow for frequency-specific, "specialty" drivers to do each what it does best in terms of rote sound reproduction. In this thread, the discussion includes the idea that multiple-drivers-per-channel might be further/better exploited by "cross-pollenating" parts of what is embedded as "fixed" L and R channel information in typical stereo program material. Just to be clear, Romy also uses the term "channel" to mean a certain frequency band that is separately routed at or near line level in his "multi-channel" DSET amps, which would ostensibly better "facilitate" the cross-pollination of parts of the L and R stereo channels.
Best rgeards,
Paul S
|
|
|
|
Posted by decoud on
04-01-2013
|
|
fiogf49gjkf0d Is it not strange that we are so fixated on spatial separation in sound? Instruments in an orchestra are spatially separated because you can't have one musician sitting in another's lap, not because the music demands it. Music is intrinsically time- not space-varying. So insisting on spatial separation is an insistence on reproducing an aspect of performance that is incidental to the essence of the music: in an ideal world you would not have it.
Of course there *are* known frequency/space interactions in human audio perception: for example, people in whom brain damage causes an inability to attend to the left of visual space tend also to be poor in attending to low frequencies. So perhaps this is the mechanism these devices are trying to exploit.
|
|
|
|
Posted by el`Ol on
04-05-2013
|
|
fiogf49gjkf0d I only heard such a quadrophonic system with a special recording, no idea what the conversion software can do for stereo basis broadening of conventional stereo recordings, but might be worth a try (with the Tannoys as rears).
http://www.andreavonsalis.com/
|
|
|
|
Posted by steverino on
04-05-2013
|
|
fiogf49gjkf0d I have to disagree with the notion that spatial separation is an artifact of music stand geometry and of no material importance to musical expression. Composers themselves have written for specific spatial groupings (eg, Gabrieli, Bartok, Henry Brant to name a few). The orchestra itself is arranged spatially to allow groups of instruments to be heard to best advantage. Louder instruments are placed to the rear so they don't overpower softer instruments. In addition, orchestral harmonic balances would be lost if there was random or even too narrow placement on the soundstage. It would also be a problem if the different instruments had too great a spatial separation as the harmonies would not blend appropriately.
BTW on a different topic and thread thanks for your noting of the lithium iron battery development. I wasn't aware that they had progressed as far as they had.
|
|
|
|
Posted by decoud on
04-06-2013
|
|
fiogf49gjkf0d The point is that spatial separation is not necessary in music in the way it is necessary for visual representation. A single musical instrument is -- more or less -- a point source, and it need be no less expressive for that, including in relation to harmonic blending, whereas no-one could ever do a one dimensional painting (though of course one might do a piece of one dimensional installation art). Yes, composers might use the spatial separation that an orchestra enforces on them, but they do not have a choice here, and to the extent to which they do have a choice they do not do that much with it. For example, I do not know of any significant composer who shuffles the positions of individuals in an orchestra dynamically to an expressive end, would move a single performer around a stage, etc. And imagine for a minute that someone came up with a way of generating the sound of an orchestra from a single point: would people really *necessarily* speak of this as a defect? On the contrary, I can imagine many telling us what a revolution it was, the grand unification of music, all of sound channelled through the single portal of God, etc.
If you are interested in LiFeP04 technology, I have found the kit these people http://www.servovision.com/ make cost-effective and reliable, though I am some way off building something that could run a full range melquiades. It is, nonetheless, now very possible.
|
|
|
|
Posted by Romy the Cat on
04-06-2013
|
|
fiogf49gjkf0d  decoud wrote: | | Is it not strange that we are so fixated on spatial separation in sound? Instruments in an orchestra are spatially separated because you can't have one musician sitting in another's lap, not because the music demands it. Music is intrinsically time- not space-varying. So insisting on spatial separation is an insistence on reproducing an aspect of performance that is incidental to the essence of the music: in an ideal world you would not have it. |
|
I have written a long replay to the post above but accidently deleted it from my MS Word document. That sucks and I do not feel to write it again. Still the subject is important and feels to briefly comment about it, interrupting my deck re-building project….
I disagree with everyone regarding the spatial separation. I do not think it is unimportant and I do not think that it is important as “steverino” feels. The fact that many composers and musicians do use spatial location as an expressive tool is irrelevant to me. If Mahler insisted that brass section in M3 has to sit off the stage then does it mean that if my playback can’t handle it (let pretend it is mono) then I am not able to hear the M3 at home?
The point that I am trying to make is that in audio there are two spatial separations.
Fist is the reflection of original performing event. This is a complicated subject as practically no one nowadays records properly: poly-microphone techniques and barbaric editing in most of the cases make the original spatial separation if not irrelevant but for sure less imperative. You can search for years an orchestra where first violins are able to enter with chairs escalating delay but even if you find them then rest assured that it will be killed by multiple microphones.
The second spatial separation is truly audio trick and from some perspective it shall not be relevant. The trick is that as irrelevant it is the ability of playback to make that trick says a LOT about the quality of playback. The little dirty secret of audio is that we do not truly recognize amplitude but we do recognize phases. It is not about the “recognition” but rather an impact that audio doe to us – it is not amplitude but phase centric. So, here is that purely irrelevant audio trick – the ability of playback to do spatial separation become like some kind of certification of quality. Of cause there are many other evidences of quality but surprisingly the spatial separation is very very handy and very easy to use. It is like pH test of water in spa – if pH is off then all kind of problems comes….
The Cat
|
|
|
|
Posted by steverino on
04-06-2013
|
|
fiogf49gjkf0d My point is not that separation per se must always be maintained but that the character of the music determines the degree of optimum separation or co-location. For example the first violins are grouped together for a reason; so the individual string tones blend. The location of the second violins is more discretionary but certain compositions benefit from one location or the other (the two possible sites are in back of the first violins or across from them on the right side of the stage. Two different sonic effects. The melodic, harmonic, dynamic and instrumentation attributes created by the composer always imply a certain kind of spatial arrangement in live performance to reproduce them satisfactorily. Even on a recording an artificial arrangement should be based on those factors in addition to any directions from the composer, producer etc. It's called mixing.
|
|
|
|
Posted by Romy the Cat on
04-06-2013
|
|
fiogf49gjkf0d | steverino wrote: | My point is not that separation per se must always be maintained but that the character of the music determines the degree of optimum separation or co-location. For example the first violins are grouped together for a reason so the individual string tones blend. The location of the second violins is more discretionary but certain compositions benefit from one location or the other (the two possible sites are in back of the first violins or across from them on the right side of the stage. Two different sonic effects. The melodic, harmonic, dynamic and instrumentation attributes created by the composer always imply a certain kind of spatial arrangement in live performance to reproduce them satisfactorily. Even on a recording an artificial arrangement should be based on those factors in addition to any directions from the composer, producer etc. It's called mixing.
|
|
Steverino, yes, that all understood but I was taking about slightly different things. Present that there is no spatial intend in music. Let pretend that we have a single flute plays in anechoic chamber, I know it is disgusting but let pretend it for methodological reasons. Even then we still have audio payback do phases processing and the virtual imaging of spatial reconstruction will be there. So, what I advocate is that we shall not confuse the musical or actual spatial information with spatial information as audio debugging tool. Sure they are related but I think that “art of audio thinking” is to be able to separate between importance of spatial reality and specifics of spatial reconstruction of reality.
|
|
|
|
Posted by steverino on
04-06-2013
|
|
fiogf49gjkf0d I'm sorry Romy usually I follow your train of thought but I'm lost on this one. There is a spatial arrangement that is optimal for the musical content. I think we agree on that. Then there is the actual spatial arrangement whether live or mixed. Then there is the audio system reproduction of that actual spatial information whether anechoic or reverberant. But you say that should not be used as an audio debugging tool? Because we can't map it precisely from the stage to the speakers..?? But wouldn't that issue be avoided by artificially mixed material? Or am I missing everything?
|
|
|
|
Posted by Romy the Cat on
04-07-2013
|
|
fiogf49gjkf0d | steverino wrote: | | I'm sorry Romy usually I follow your train of thought but I'm lost on this one. There is a spatial arrangement that is optimal for the musical content. I think we agree on that. Then there is the actual spatial arrangement whether live or mixed. Then there is the audio system reproduction of that actual spatial information whether anechoic or reverberant. But you say that should not be used as an audio debugging tool? Because we can't map it precisely from the stage to the speakers..?? But wouldn't that issue be avoided by artificially mixed material? Or am I missing everything? |
|
Well, what was saying that spatial capacity of playback has nothing to do with spatial arrangement that is optimal for the given musical content. You see, playback is brainless dead substance that has no knowledge or understanding of musical content or spatial need for played musical content. Playback has however own spatial capacity that has absolutely nothing to do with music you play. Pretend that we play on our playback ONLY test signals then you will see that different inhalations would demonstrate completely different spatial capacity. In my view when we talk/think about spatial capacity of playback we need to forget about spatial load of music and let the playback-only to do the spatial tricks only for sake of playback. We can talk to each other stories how clarinet could magnificently bind string sections located at opposite sides of orchestra but it will not help us to figure out why for instance Convergent Audio Technology preamplifiers, being fine preamplifier has problem to position vertical image properly. CAT preamps have that vertical parabolic curve which position violin section on right and left but mid image elevates vertically, and it look like it is not depending from acoustic system setup. Operating by musical categories it does not make sense. So I do feel that a playback has to be given some strictly audio test where spatial capacity would be validated. It might be playing some spatial y charged music where playback would demonstrate its ability to deal with it. Then the most interesting things happen: each playback has own way to deal with space. Pay attention: no one would argue with musical spatial information but we all would have different opinion and different experiences how our current or imaginary playbacks would deal with space. The interpolations by playback of the spatial factor has nothing to do with spatial intend in music and therefore I do not feel that it is possible to prove the importance of spatial character in audio by stating that live music has a lot of spatial intend.
|
|
|
|
Posted by steverino on
04-07-2013
|
|
fiogf49gjkf0d I think my statement " Because we can't map it precisely from the stage to the speakers..??" was on the right track. Yes there is a distortion of spatial information from performance to speaker in the same manner that a curved shiny ball distorts a painting reflected on it. I think the only slight exception would be a recording of musicians playing in the same room between where the playback speakers stand. Playback then should have a fairly close fidelity of spatial information on a high fidelity system.
|
|
|
|
Posted by Romy the Cat on
04-07-2013
|
|
fiogf49gjkf0d | steverino wrote: | I think my statement " Because we can't map it precisely from the stage to the speakers..??" was on the right track. Yes there is a distortion of spatial information from performance to speaker in the same manner that a curved shiny ball distorts a painting reflected on it. I think the only slight exception would be a recording of musicians playing in the same room between where the playback speakers stand. Playback then should have a fairly close fidelity of spatial information on a high fidelity system.
|
|
Well, frankly I am less concern about spatial information as imaging of instruments and sections. Sure it is important but I’ll rather interested in different way to deal with that information. It is hard to explain but if you try the bypass test with Lamm L1/L2 you might understand what I mean as it is VERY different from anything else out there. Still, I consider the most interesting not the spatial information from imaging perspective but pure spatial information without imaging. I mean let say a sting quarter is playing then the locations of both violins, alto and cello would be imaging but the virtual proximity between the players and the boundary of the performing space is pure no-imaging space. Lamms preamps did nothing in that direction and I would like to experiments with it worth my idea or cross phase injection. You see, I think that if we use some kind of smart cross-phase injection and turn the injection channel toward to the wall then we mish create some kind of condition where something might happen. It would be nice to play with delay channels why we do it but they we would end ups with those 200 drivers… :-)
|
|
|
|
Posted by steverino on
04-07-2013
|
|
fiogf49gjkf0d I don't have Lamm, but I think conceptually there is no real difference between imaging the location of the players and imaging their non location. Sound waves generated by the musicians are traveling in the space between them and even if it weren't (as in a pause), some sound information however low level should be present at every point of the performance space. How we (or the mics) hear the space between the players or between them and the room boundaries, if any, all qualifies as imaging doesn't it? It's just that people tend to use other terms such as ambiance or hall reflections to discuss the imaging Not associated with the location of the performers. That's why mic'ing of each player can sound like simultaneous mono recordings arranged in stereo space.
|
|
|
|
Posted by rowuk on
10-20-2018
|
When listening to stereo, we initially have 4 images to deal with. Th right speaker to the right ear, the right speaker to the left ear, the left speaker to the right ear and the left speaker to the left ear. Broader images can be made when we “cancel” the cross talk. Most of the attempts to do so had such serious problems elsewhere that the experiment could not be considered valid.
We can at least partly “cancel” cross talk. Polk Audio did this acoustically many years ago. Bob Carver made his contribution electrically with “sonic holography”.
I do not know of any real high quality experiments but rereading this thread gives me some ambition to try. The initial L-R and R-L will be created acoustically. The speakers will be located about an ears width from the main speakers and I will play with level. Perhaps there is something to learn here and it is certainly cheap enough.
|
|
|
|
Posted by Paul S on
10-20-2018
|
Robin, there have been many and varied attempts to present a "realistic", "3-D" version of recorded music. The late Jame Bongiorno adamantly insisted that it could not be done with two channels, and he came up with his Trinaural Processor to "correct" the problems. This device takes into account just what you mentioned about the R/L/L/R thing, with an algorithm that directs a "scaled" combination of R & L to a center channel and "scaled" amounts of R to L & L to R. There isn't much of worth out there about this device, but I cynically suppose anything gained is offset by electronic "issues". Another "problem" is that original recordings might be... anything, so any given fixed "treatment" might have any sort of final outcome one could imagine, even without other variables.
Paul S
|
|
|
|
Posted by Paul S on
10-20-2018
|
(with apologies to Giordano Bruno)
OK, had I re-read this entire thread from the top before posting again (above) I would likely not have repeated myself. However, IMO, this subject is not thought about nearly enough going in, at the recording end, to make for a simple, one-size-fits-all fix. I'm thinking as I sit here typing that "Sound" and "Presentation" are juxtaposed.
Paul S
|
|
|
|
Posted by rowuk on
10-20-2018
|
Paul,at least from my moronic standpoint, using a couple of closed boxes with wide band drivers (I have some JBL LE-8T2 without the aluminum domes) in a L-R/R-L connection, should give me the power to widen the stereo basis. Adjustable level could make the amount of widening fine tunable based on the recording. I have pretty much an equilateral triangle between the listening spot and the speakers so I get somewhere around 60° now.
If it works as intended, this could be my injection channel...
|
|
|
|
Posted by Paul S on
10-20-2018
|
then it works! Of course, nobody really wants a 12' piano keyboard, or 2' bells on the French horns, so (obviously) we want both more and less than rote "widening" of the event. Likewise with solo performers; we want a "more accurate" recreation of the "venue", and some of this also has to do with "subtle" sound cues that can actually be blurred by combing, etc..
Best regards,
Paul
|
|