05/017079

01/09/1973

03/06/1970

Download PDF 3710034       PDF help

Click for automatic bibliography generation

Fibra-Sonics, Inc. (Chicago, IL)

360/7

369/60.010, 381/19, 369/89

G11B5/00; G11B20/00; H04S3/00; G11B23/18; H04H5/00; G11B21/00

179/1.1TD,1.2RE,1G,15BT,1.2MD,1.3B 181/31B

3538265

INSTANT REPLAY SYSTEM FOR RADIOS AND THE LIKE

November 1970

Keeler

Urynowicz Jr., Stanley M.

Cardillo Jr., Raymond F.

I claim

1. A system and apparatus for recording and playing back sound which duplicates the volume of the original recording space and wherein the volume of the original recording space differs from the volume of the playback space comprising:

2. A system and apparatus according to claim 1, wherein said means for recording comprises four microphones and said sound reproducing means comprises four speakers respectively mounted in the upper corners of the recording space and playback space.

3. A system and apparatus according to claim 2 wherein said means for recording includes a phonographic recording means formed with a plurality of recording channels.

4. A system and apparatus according to claim 3 wherein said phonographic recording means is a disc with two separate recording grooves each of which contain two channels of information.

5. A system and apparatus according to claim 4 wherein said means for reproducing includes a plurality of styluses receivable in said separate recording grooves to pick up the various channels of information and means for adjusting said styluses relative to each other.

6. A system and apparatus according to claim 2 wherein said means for recording comprises a magnetic recorder with a plurality of recording heads for recording said different recording channels.

7. A system and apparatus according to claim 6 wherein said magnetic recorder is a magnetic tape device and said sound reproducing means is a magnetic tape device with a plurality of playback heads and said plurality of time delay means comprises means for adjusting the positions of said playback heads.

8. A system and apparatus according to claim 6 wherein said magnetic recorder is a flat magnetic disc device with a plurality of recording heads for recording different recording channels and said sound reproducing means receives said flat magnetic disc and has a plurality of playback heads and said plurality of time delay means comprises means for adjusting the angular positions of said playback heads relative to said disc.

9. A system and apparatus according to claim 6 wherein said magnetic recorder is a magnetic drum device with a plurality of recording heads for recording different recording channels and said sound reproducing means receives said drum and has a plurality of playback heads and said time delay means comprises means for adjusting the positions of said playback heads about said drum.

10. A system and apparatus according to claim 9 comprising a plurality of arms upon which said playback heads are mounted and means for adjusting and holding said arms in various positions.

CROSS REFERENCE TO RELATED INVENTIONS

My prior U.S. Pat. No. 3,360,073 entitled LOUD SPEAKER ENCLOSURE which issued on Dec. 16, 1967, discloses a speaker which when mounted in a corner of a room formed, by example, by the ceiling and two adjoining walls creates the effect on a listener in a room of being surrounded by a sound emanating from the speaker. This patent discloses an apparatus for the projection of sound but can be modified and used in the present invention for recording as well as the projection of sound.

BACKGROUND OF THE INVENTION

1. Field of the Invention

This invention relates in general to sound recording and reproduction. Numerous patents and articles have disclosed devices and systems of speakers, or of speakers/amplifiers microphones in combination which have as their objectives the faithful reproduction of the original sonic materials at a location other than the recording location. Such multi-dimensional sound has been called "stereo" (meaning solid), "high-fi," or 3-D sound. Yet, most of these systems approached the problem as if it were merely 2-dimensional in nature and as if it were taking place in an area, e.g. "the listening area." Actually all sound takes place in volumes and not in areas of space; and these all have 3-dimensional characteristics of resonance, reverberation, absorption, and many other characteristics, which not only can, but do change the sound that comes from any radiator, or which arrives at any receptor, most drastically. My prior U.S. Pat. No. 3,360,073 describes a method and apparatus whereby it is possible to achieve 3-dimensional coupling to a 3-dimensional space with effectiveness, efficiency and fidelity which was superior to prior methods. My prior patent however, did not give the mathematical basis for a complete understanding of the theoretical work resulting in that invention. In addition, quantitative reduction to practice of that invention has resulted in a superior understanding of the mechanisms which are at work and of further applications of the principles involved which have resulted in further inventions and applications; one of which is the use of the principles in a superior microphone encasement and sonic input-conditioner.

2. Description of the Prior Art

There exist several severe nomenclatural errors in sound production fields regarding pickups and loud speakers (not the least of which is the latter word, i.e., loud speaker, which should be called at the least, a "loud-sounder"). One prevalent error has been in the almost complete use of the narrow 2-dimensional concept of the "listening-area" when what is actually meant is the "listening-volume," since sound is reflected from all available surfaces including floors and ceilings. Another error, due to the failure to appreciate and use the volume concept rather than the area concept, is the present idea of "stereo" sound, wherein two sources of essentially point-source sound are supposed to create a sonic solidity (stereo), in 3-dimensions. As a matter of fact this idea has been proved erroneous time after time by several acoustic engineers. Also, in existing systems many compromises are made between the accuracy of the left-to-right spacing and nothing whatsoever is done about the front-to-back spacing, or to the ever present third dimensional "modeling" of this type of sound which is urgently required.

For many years it was felt that the objective was to create a "wall of sound" based on the concept of 2-dimensionality. The erroneous mixing of sound components in a temporal/spatial hodgepodge of relationships, from left-to-right, and from front-to-back, even by some advocates of so-called surrounding stereo, is crude, inadequate and not, in fact, true sculpturing of the sound to a solid sound (stereo). One of the weakest links in the entire science has been the consistent failure to admit that sound is truly a 3-dimensional, or true volume phenomenon although authorities such as Olson clearly states and shows, as in Acoustic Engineering of October, 1964 at page 4, that "the general case of sound propagation involves three dimensions."

SUMMARY OF THE INVENTION

The present invention allows the design and construction of a recording and reproduction system which implicitly obeys the general equations of sonic propagation in 3-dimensional space. A novel recording means for similar reproduction of sonic material is provided and a system and apparatus is provided for accurately achieving a multi-dimensional playback sonic effect highly similar to the original material experienced in the playing volume such as an auditorium, studio or theater and to accomplish this regardless of the dimensions of the playing-back volume. Various apparatuses for recording and playback in which multitracks, that may be individually adjusted, are provided so that the phasing of various signals may be accomplished in order to provide a replica of the original sound in a volume which differs from the volume of the recording location.

Other objects, features and advantages of the invention will become readily apparent from the following description of certain preferred embodiments thereof taken in conjunction with the accompanying drawings, although variations and modifications may be effected without departing from the spirit and scope of the novel concepts of the disclosure and in which:

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a plan view of a typical small room provided with the present invention;

FIG. 2 is a sectional view of a novel pickup microphone according to my invention;

FIG. 3 is a plan view of a recording chamber illustrating the invention;

FIG. 4 illustrates a recorder for reproducing sound according to this invention;

FIG. 5 illustrates a playback mechanism according to the same invention;

FIG. 6 illustrates a recording room;

FIG. 7 illustrates the base relation among various signals in the recording room of FIG. 6;

FIG. 8 illustrates a playback volume;

FIG. 9 illustrates a phase relationship in the playback volume of FIG. 8;

FIG. 10 illustrates a special illustrative recording system of the invention;

FIG. 11 illustrates a similar playback system of the invention;

FIG. 12 illustrates a special tape recorder for recording and playback;

FIG. 13 illustrates a sideview of the tape to be used with the machine of FIG. 12 to show the 4-tracks used;

FIG. 14 illustrates a magnetic recording disc with recording and playback mechanism;

FIG. 15 illustrates a magnetic recording and play back drum;

FIG. 16 illustrates a generally pie-shaped recording auditorium to demonstrate a complex situation;

FIG. 17 illustrates a typical listening volume converted to an equivalent pie-shape;

FIG. 18 illustrates the invention as applied to regular dual-track disc type recording;

FIG. 19 is a multiple stylus holder for the disc of FIG. 18;

FIG. 20 is a schematic view of the microphone or loud-speaker enclosure of this invention or of U.S. Pat. No. 3,360,073;

FIg. 21 is a side view of the microphone or loudspeaker enclosure;

FIG. 22 is a plot of sound waves versus time; and

FIG. 23 is another plot of sound waves versus time.

DETAILED DESCRIPTION OF THE INVENTION

My prior U.S. Pat. No. 3,360,073 discloses a 3-surface corner sound radiator for mounting at the ceiling at a junction of two walls and which has adjustable control vents at each of its three corners. FIGS. 20 and 21 illustrate such a radiator 10 with corner vents a, b and c. Such a 3-surface corner radiator, built as shown obeys the classical 3-dimensional space matrix configuration in what I designate "ω" (omega) space and "π" (pi) space inter-operations.

The equation of continuity states that: "The amount of matter entering "ω" space equals the increase in matter inside this contained volume of space."

It can be assumed that the influx and efflux through the ports a, b and c is analogous to that which would enter from the x, y, and z dimensions. Hence, since, Δ x, Δ y, Δ z equals the permitted Δ a, Δ b, Δ c, then the difference between the influx matter and the efflux matter would be:

where a, b, c, equals the coordinates of a particle in the common medium; u, v, w, equals the component velocities of the particle in the medium; and ε' equals the static density of the medium.

The rate of growth of the gas mass in the "ω" space volume δε'/δt Δ a Δ b Δ c, is equal to the above expression, or in time units:

and it can be seen that the three dimensionality of space is obeyed, as is the conservation of matter, hence this system will react to these facts implicitly, and all the acoustic wave equations are valid since the revent ports (a, b, c) are oriented in such a manner as to be aligned in x, y, and z dimensions, and the system obeys implicitly the conservation of matter law, and the 3-dimensionality of space energy propagation laws.

The important point is that energy transfer from "D" (the energy generator) to "π" space (which generator also occupies "ω" space) can only be done with proper 3-dimensional obedience if "π" space and "ω" space are coupled in such a manner as to preserve their correct interrelation. Ports a, b, and c being present, and properly oriented do this, while all other systems cannot do this completely since usually only one of the dimensional components (x, y, or z) are coupled from their "ω" space to their "π" space, or even no coupling is used.

Since motion (referring to Δx, Δ y, Δ z) changes "ω" space, it is found that the acceleration of momentum parallel to x, is ε' Δ x Δ y Δ z δ u/δ t (as is the acceleration of momentum parallel to y and z with simply a change in symbols).

Then the mean pressure on the face perpendicular to the x (or y or z) dimension is:

and for the y dimension:

and for the z dimension:

(when ε o' = the pressure in the media.)

The difference in ε o' being a force δε o/δ x Δ x Δ y Δ z, or δ ε o/δ y Δ x Δ y Δ z, or δ ε o/δ z Δ x Δ y Δ z in the increasing x dimension (or in the increasing y and z dimensions) and in the direction of increasing x (or y or z). The equation of motion for a three dimension space then becomes:

which can be simplified to: dVuvw/dt + 1/ε Grad. ε o' = 0.

Obviously, sound pressure (εo') can only obey the conservation of momentum and act in 3-dimension space, if the x, y and z components co-act between "ω" space and "π" space.

Since the media of propagation is a gas under rapid compression and expansion due to the sound waves present, the temperature of the media varies adiabatically, and it has been shown elsewhere that:

where εo = static pressure (no sound present)

ε = static or original density

ε o' = total pressure (excess plus static)

ε ' = instantaneous density (excess plus change)

(for air γ = 1.4); γ = ratio of specific heat at constant pressure to that at constant volume: the ratio of the increment of Δ ε ' to the original static density is: ratio = ε '-ε/ε;

combining this with the preceding equation of compression we have:

or: ε o' = ε o + ε oγ ^. ratio, and the excess (instantaneous pressure) sound pressure P equals ε o γ ratio, or P = ε oγ ε'- ε/ε.

All the above applies for a perturbation of the media by sound waves of any amplitude and it is to be noted that the rate of growth of the mass, and the equation of motion for the mass, is non-linear, if the amplitudes are large; hence, must be non-linearly provided for, while the compressibility of the gas equation is linear and can be readily handled regardless of amplitude.

The saving factor for acoustic waves are their comparatively small pressure amplitude (compared to the pressure of the atmosphere), and the wavelength being relatively large, hence, u, v, w, and the ratio ε'-ε/ε change very little in their x, y and z travels.

Since this is true it follows (from D'Alemberton's wave equations) that the equation for the velocity of propagation of sound in a media is: γεo/ε = C ² = velocity squared where,

γ = 1.4 (for air)

ε o = the static pressure with no sound present (in dynes/cm ² )

ε = the density of the medium of transmission.

and we must critically note that for sound in air γ is fixed at 1.4, as is the value of ε o, hence, the velocity of propagation, and all other parameter related backwardly with the above, is a function of the density of the media, which is varying directly in front of the diaphragm in "π" space and behind the diaphragm in "ω" space. Thus, the velocity of sound is different at all times to the front and rear of our speaker diaphragm (and in all moving cone speakers) and compensation for this fact must be provided in some manner.

The problem therefore becomes simply a case of the proper control of the pressure to the rear, and to the front of an electrically driven mechanically oscillated diaphragm, which produces wave motion in the chosen medium in which it is operated.

In the actual operations of the tri-surface, tri-portal sound-conditioner enclosure, we observe that the volume to the rear of the sonic generator (the diaphragm 11) is exactly prescribed, and is exactly determined due to the air being sealed into the rear at each edge by a rubber seal; however, the revent ports a, b and c provide escape for the air and are of a definite length and cross sectional area, both of which are variable, and are carefully oriented in the x, y and z axis. These are made of acoustically "dead" material, so as not to "add-to" or "subtract-from" the frequency components of the sound that is desired, and which is present in the feedback wave sample as it passes thru these ports.

When the diaphragm 11 moves forward momentarily under the influence of an input of electrical energy into the voice coil of the speaker, a slight vacuum is created to the rear of it, while a positive pressure appears at its front surface. This latter is not permitted to escape instantly, but is critically controlled in its behavior by the sound conditioning cone in its pathway, which is so designed of selected laminating materials of various sonic absorbing or reflective abilities so as to provide conditioning of the sound on a frequency selective basis. Further, it provides a precisely calculated loading to the pressure front. As the wavefront moves outward, part of its x, y and z components find pressure relief ports properly oriented in their pathways (note that I purposely feedback components of the x and y, the y and z and the z and the x complex wave). Controlled samples of the x and y, the y and z and the z and x components are permitted to return to the rear of the diaphragm to alleviate the vacuum there present. The phase/time variations of this unloading wave may be controlled by the length of the tubes a, b and c. By controlling the cross sectional area of these same tubes, the amount or quantity of the unloading is controlled. The total amount of unloading experienced is of course equal to the sum of the three port areas, times the size of these areas, which are usually varied by selective insertion of various tubes of decreasing diameters.

Thus, we have devised an acoustic analog of negative-feedback, so important to amplifier linearity, and as might be expected this negative acoustic feedback also provides increased sonic linearity, equivalent to that of the many RC circuits used in various amplifiers; the tube length being equivalent to the resistance (R), while the tube diameter is equivalent to the capacity (C), of such type circuits.

In the above, we can see that the loading/unloading cycle behaves similarly if the diaphragm moves backward, wherein the positive pressure appears at the rear of the diaphragm, and an instantaneous vacuum appears at the front. As these variations take place, it is to be noted that the greater that the "to" and "fro" movement becomes, the greater is the pressure (or vacuum) generated and that it will follow the non-linear compressibility of the transient air cushion prevailing at any given instant of time. This is done dynamically by a static system of acoustic negative feedback built up of static parts. It provides for severe electrical driving of the one moving part (the diaphragm 11) without distortion appearing in the air wave or without the so-called "breakup" of the cone appearing. One can operate the diaphragm air-driver to the limits of the available power of the associated amplifier without distortion appearing, since it always looks into the proper loading at all times. In this manner, one may use all of the power available and have no need for purchasing excessive amounts.

Referring to the previous equations for a moment, the significance can be seen of the prescribed volume concept being to the rear of the cone driver, to the conditioning cone concept (for selective absorption and reflection of various frequencies), to the vital principle of sampling the x and y, the y and z, and the z and x components, then using these as acoustic negative feedback for the automatic variations in time of the instantaneous pressures and vacuums which appear so transiently at any given moment.

It should be noted that while the ear can detect and instantly notice the increased fidelity, clarity and intelligibility of the eminent sound from this system, when properly designed, it is quite difficult to measure this on most electronic instruments, unless one knows what is being done, and uses special methods, based on the transient responses of the system.

As seen from above, the simple seeming enclosure, shown in my U.S. Pat. No. 3,360,073, is not simple and is indeed the only speaker which implicitly recognizes the three dimensionality of the sonic problem, and by its proper placement, its conditioning effect, and most of all, by its use of the revent ports a, b and c (which are really negative-pressure feedback devices, designed to control the efflux and influx of pressure variations in the particle flow from the front to back of the speaker, thereby controlling the loading and unloading of the piston) operates in such a manner as to automatically control its own excursions, precisely and at all levels of sonic drive, and that its unique tri-surface design insures the rapid formation of a true spherical-segment wavefront. As this is formed, it spreads out from its upper-corner location, maintaining all its spherical 3-dimensional aspects, into the listening volume. Even this wavefront is somewhat degraded by the varying refraction, diffraction and absorption near its edge, but not anywhere near to that of all other systems. A moment's visualization will show why this is true and of importance. With sonic wave lengths present (in small chambers) of 55 feet, in 1 cycle of length (20 cycles per second) which takes but 1/100th of a second to travel across a 10 foot room, we can see immediately that stereo-modeling of the sound by a single point source is most difficult. The addition of another radiator according to my patent in one of the other upper corners, will behave similar to the number one speaker; but, in addition, it will provide a most excellent stereo effect (a simulated spatial effect since no part of the wave will be out of phased, 3-dimensional relationship). Even when very large separations are used, due to the tight coupling to the walls and the ceiling, the stereo effect is not "choppy," but startlingly lifelike. Measurements which have been made, have shown the sound characteristics to be about 30 percent similar to that of recorded tests made in the same room with a live source of sound present. By that is meant that the phased components of the various complex segments are present in proper spatial relationships. Tests on other systems showed less than 15 percent spatial preservation of the characteristics of the complex sonic-components of the outgoing sound.

Although this figure of merit is somewhat arbitrary, it is the only known method of checking the "modeling" of the sonic profile, generated by a speaker, or a speaker system. Even with the previous high figure of merit, it is entirely unsatisfactory, and this present invention allows modeling of the spatial/sonic distribution to be increased to a figure of merit of 75 percent, by means of multi-dimensional, total immersion stereo sound.

GENERAL DESCRIPTION

In FIG. 1 is shown a typical small room 12 (e.g., 15' × 10' × 9'), equipped with four sound-conditioners 13, 14, 15 and 16, according to U.S. Pat. No. 3,360,073, mounted in the four upper corners as shown and connected, respectively, to four separate amplifiers 17, 18, 19 and 20. A multi-mixer preamplifier 21 is connected to amplifiers 17-20 and a four channel tape machine 22 is connected to preamplifier 21. Machine 22 includes four channel play back tape playback capability. Now provided channel No. 1, channel No. 2, channel No. 3, and channel No. 4 are transmitting true spatial/multi-dimensional data from its four separate recorded channels, which were properly related each to the other at the time of recording, true stereo determined sound will be heard anywhere in the room, and as the listener 23 moves around, the sound will vary similar to that of a holosonigraph with a three-dimensional effect. If it is not correctly recorded, this will not be true. To achieve correct recording, I propose the following:

Replace each of the sound-conditioner speakers located in the four corners U.S. Pat. No. 3,360,073) by four pickup microphones located inside specially modified tri-surface corner enclosures, as shown in FIG. 2, using this room, or another studio of its same inside dimensions (15' × 10' × 9'), being sure to use a sound-conditioner device for the microphone (to preserve the vital "X:Y:Z" components needed to preserve the 3-dimensionality of the sound during recording), we would record a spatially determined multi-dimension stereo model of any sound appearing in that chamber. Then, as shown in FIG. 3, all discrete sonic components, from wherever derived either directly or secondarily in the recording chamber, will be recorded on the four-channel recorder as 1st order 4-dimensionality data along with higher order n ^th -dimensionality data (due to the multitude of reflections, absorptions, etc., in the chamber).

FIG. 2 is a sectional view through the pickup 27 mounted to adjacent walls 24 and 25. The microphone 26 is mounted on a support 28 behind conditioning cone 29.

FIG. 3, we see in room 30 that sound from the sonic point "A" travels distances a ₁ , a 2,a ₃ and a ₄ , so as to arrive at microphones 31, 32, 33 and 34, respectively. In the examples (sound from emitter point "A"), these distances are 61/2 ', 111/2 ', 111/2 ', and 61/2 '. In a like manner, sound from point "B" could be coded as b ₁ - 8', b ₂ - 10', b ₃ - 10', and b ₄ - 8'. The variation in distance from the sound source to the several microphones 31-34 are even more diversified in the case of a doubly offset sound source, such as that of point "C." Point "C" sounds, spatial distance wise, may be coded as follows: c ₁ - 101/4 ', c ₂ - 9', c ₃ - 8'. Then if we take, as but one example, a sound wave at a low frequency of 32.7 c.p.s., which will have a wave length of 34.5 feet, then the condition illustrated in FIG. 22 would take place, if this wave started from point "C."

The amount of "phase variation" as shown in FIG. 22 is readily detected, and it is this which gives the various homes, studios, theaters, and auditoriums their coloration and atonality; since we must remember that these chambers will be constantly reflecting, absorbing, and reinforcing the various components of the complex sounds that impinge on their walls, ceilings, and floors, and onto and from the materials with which they are covered. Notice that the wave, when it is long, as in this case, is linearly "sampled" at four distinct points by the four microphones.

The situation becomes even more complex as the frequency of transmission increases somewhat. For example, if the wave has a frequency of 112.8 c.p.s., then its wave length would be 10 feet and we would have the situation illustrated in FIG. 23.

We see that with microphone 31, we will have a zero amplitude input for recording on our channel No. 1 (at this time), while with microphones 32, 33, and 34, we are at a somewhat less than maximum pressure amplitude (at the same time), which will be recorded on our channels No. 2, No. 3, and No. 4 recordings. All of which implies, that we can quite precisely provide n ^th -dimensional modeling of the sonic energy inside a chamber, if we use the corner locations as microphone pickup regions, and also as the playback points, and if the room in which the playback is made is identical in volume dimensionality to that of the recording room. In this case, we will achieve a true holostereogram (entirely solid record) reproduction of the original (a replicate) with all the overtones, second order reflections and reverberations, and higher order effects. It is assumed, of course, that the input sensors, the transfer audio system, the recorders, the amplifiers, and the playback reproducers that are used, preserve the many other important aspects of the sonic n ^th -dimensioned matrix, such as frequency, fidelity, balance, level, etc.

It further should be obvious that if microphones are placed elsewhere, than in the corners of the recording volume, even if the speakers are placed in the identical locations in the reproduction room, we cannot achieve true stereo, or fidelity, since the n ^th -ordered components will not be permitted to obey the three-dimensionality of space, as set forth earlier. If the speakers are not located in the same place as the microphones, then of course all effort to reproduce the original sound is futile; as is also the case when the microphones are indiscriminately placed, but the speakers are correctly located. Some results in the past have been achieved with attempts at sonic modeling of the recording chamber; but, as can be anticipated, use was made only of the first order dimensionality sonic energy.

Thus, no combination of microphones, placed in any manner, in any studio, theater, or auditorium, can be used, which will capture the n ^th -dimensionality spatial relationships of the recording-volume made use of, for the recording and later reproduction of the sound in it, except when we place them in an upper corner and use a geometrically determined identical room, which we provide elsewhere. Thus, any effort to maintain the spatial relationship in an oblong room when the original recording was made in a truncated chamber is hopeless. Even if the corner microphone technique is used, there is no guarantee that the home listener will place his speakers in an identical position, and in fact, and in most cases, it is not possible; nor, can one pick the proper, or "best" place, since furniture and other bric-a-brac will invariably intervene. With the use of the upper microphone/speaker corner technique, true modeled stereo, which duplicates the original, is now possible.

I shall now show how this is indeed possible, and how we additionally can "time-model" small listening rooms in which most recordings are played back, in such a manner so as to duplicate the original studio or other recording location. Indeed, it is possible not only to compress and expand the volume of a small room at will, by time-modeling, but to achieve purposely distorted and highly interesting effects by this technique.

FIG. 4 illustrates four microphone sound-conditioners 31-34 located in the corners of the recording studio 30, auditorium or hall similar to that in FIG. 3. A recorder 36 has 4 distinct and separate recording systems which will produce four separate tracks of taped materials 37-40, one from each channel which are respectively connected to microphones 31-34.

FIG. 5 illustrates separate amplifiers 41-44 for playback of the recordings, each of which will make use of one of the tapes 37-40 made on recorder 36. Each amplifier 41-44 will have its own volume, balance, high/low boast controls, etc. We will identify channel No. 1, as RT _o , or the "Record at Time Zero" channel, and channel No. 1 _a , the PT ₀ , or the "Playback at Time Zero" channel, etc. Amplifiers 41-44 are respectively connected to corner speakers 46-49 according to U.S. Pat. No. 3,360,073 mounted in room 50.

Now since from any point P ₁ , P ₂ , P ₃ ... P _n , (FIG. 6), in the recording volume different arrival times of the sound from any point will be recorded by the pickups, if we call one channel RT ₀ and assume that the sequence of events start at T ₀ ; then RT ₁ , RT ₂ , and RT ₃ , will have a spatial/time/displacement equivalent to the physical distance that the sonic waves travel at a velocity of approximately 1,128 feet per second. Let us assume that the studio of FIG. 6 is 30' × 20' × 9' in its dimensions and we have a point sound source at P ₁ , then for sound to go from RT ₀ (P ₁ ) to RT ₁ (P ₂ ) would take 26.4 milli-seconds to arrive at point RT ₁ ; 31.7 milli-seconds to arrive at RT ₂ ; and 17.7 multi-seconds to arrive at RT ₃ . This is shown in FIG. 7. But, since our replay room FIG. 8 is only 15' × 10' × 9', or one-half of these dimensions in the left-to-right, and front-to-back dimensions, the sound would be projected back to any listener much too fast (see FIG. 9), and, as is actually observed, would be unsatisfactory to a great extent, since we have completely destroyed the second order time variant reverberant characteristics of the original sound. If, however, we make arrangements to insure that we delay each of the non-PT ₀ channels, i.e., PT ₁ , PT ₂ , and PT ₃ , by an amount equal in delay, in this case, to one-half of the spatial time factor, we will recreate the original dimensions, as far as an observer in that particular volume ("ω"), is concerned. We simply place each tape 37-40 on the channel 1, 2, 3, and 4 machines 41-44 in such a manner that their separate sounds are slightly late in being replayed, by specific time amounts. That is by 13.2 milli-seconds for PT ₁ , by 15.8 milli-seconds for channel PT ₂ , and by 8.8 milli-seconds for channel PT ₃ ; then as far as the observer is concerned, he will perceive the spatial relationship of the sound as if from the original studio, i.e. 30' × 20' × 9'.

This can be readily accomplished by simply turning each tape 37-40 on its axis by an amount equal to the tape's passage time past the pickup head by an amount equal to the desired equivalent transit time. This will, of course, depend on the tape speed used by the replay channel. For example:

At a tape speed of 7.5 inches per second, we have

A plus displacement of PT ₁ of 0.099 inch on a linear line

A plus displacement of PT ₂ of 0.119 inch on a linear line

A plus displacement of PT ₃ of 0.066 inch on a linear line.

While these amounts of rotary displacement may seem small to attempt to physically displace an 8 inch diameter reel, for example, they could be achieved if necessary; however, better and more adequate methods are available for actual production of usable devices. Nevertheless, the entire principle can be demonstrated as shown, with no special equipment being necessary. In the milli-second delay region that is required, a normal type "set-delay" knob, (in FIG. 5) marked 51, 52, 53 and 54 could be used to actually rotate the four reel holders on machines 41-44 by a very slight amount if required and the effect would be achieved. Alternatively, the various playback heads, 55-58 of the machines could be easily displaced along the tape tracking path by these amounts to achieve the same effect.

If the recording studio had been an auditorium of say, 180' × 120', then since these dimensions are far larger (about 12 times as large as those of the playback chamber) we must increase the playback room (by means of our sonic space "Compander") by an amount equal to the increased linear delay time that would be required to sonically change the dimensions of the listening room. This will require the following changes in the relative displacement of the tape heads (referred to PT ₀ , FIG. 8):

A plus displacement of PT ₂ by 0.495inch

A plus displacement of PT ₃ by 0.714 inch

A plus displacement of PT ₄ by 0.396 inch.

Thus, we can make our small room of 20' × 10' × 9' into a 180' × 120' auditorium, or indeed, into any size we choose, simply by turning the tapes, or moving the playback heads relative to a common reference point PR ₀ . As proposed herein, there may be a slight error in the vertical top-to-bottom dimension, which can be corrected on the tape directly at the time of recording with little error resulting, due to the fact that most playback rooms will be very close to 9 feet in height (7 feet to 12 feet on the average). It is of course possible to compromise quite severely, and "build-in" the delays on a somewhat average basis for the various recording chambers, and assume that the listening volume will have an average spatial size of say 15' × 12' × 9', as was done for the vertical dimension and still obtain good results.

FIGS. 10 and 11 illustrate a system for directly and instantly achieving the modeling-effect, wherein a single four-track tape recorder 55 is used to assure rotational-time synchronization at the recording studio. This means that our four track recording will have all the n ^th -dimensional spatial data on it, keyed to the dimensions of the recording studio's volume, and provided we use our corner microphone pickups 56-59 and our corner playback speakers, 60-63 (see FIG. 11), we can "scale" the n ^th -dimensionality factors down or up, as we desire, by delaying the various channels relative to each other. We show four control knobs 65-68 on the playback machine 64 whose attached mechanisms move the playback heads along each separate track relative to each other so as to achieve the appropriate and required delay, thus accomplishing the necessary time sculpturing of the original sonic data.

This method can be installed on existing-type tape recorders with ease and slight modifications; however, a separate auxiliary system can be used to advantage with a resulting increase in the ease and precision of sculpturing the sound in the listening volume to that of the original recording volume.

FIG. 12 illustrates a modification in which a tape machine 70 has magnetic tape 71 carried by rollers 72, 73, 74 and 75. An erase head 76 and spaced record heads 77, 78, 79 and 80 are mounted on the machine 70. Playback heads 81, 82, 83 and 84 are adjustably mounted on shaft 86 relative to the tape 71 to obtain variable delay. To obtain easier precise control of the displacement, we use tape at band speeds up to 15 inches per second. This will give a displacement variation of from 0.15 inch to about 1.5 inches for movement of the heads. The replay heads 81, 82, 83 and 84 may be adjusted on shaft 86 by knobs 87, 88, 89 and 90. As shown in FIG. 13 the tape 71 has four bands 91, 92, 93 and 94 and a pair of record and playback heads are associated with the track.

FIG. 14 illustrates a modification in which we use a flat magnetizable disc 95, containing four bands of imbedded magnetic material 96, 97 98 and 99. Erase heads 100-103 are mounted on fixed arm 104 and record heads 105-108 are mounted on fixed arm 109. Movable arms 110, 111, 112 and 113 respectively carry playback heads 114-117 which are aligned with bands 96-99. The position of arms 110-113 may be selectively adjusted to obtain the advantages of the invention. It should be noted that by judicious selection of the tracks on the various bands, we can decrease the necessity to allow for the difference in true velocity pass the head of the peripheral bands to some extent. Note this in the ordering of the replay bands, wherein band 97 may be taken as PT ₀ , band 96 is taken as PT ₂ , band 98 is taken as PT ₃ , and finally band 99 as PT ₄ .

FIG. 15 illustrates another modification in which a magnetic drum 115 has bands 116-119 of imbedded magnetizable ferro-material on the cylindrical surface. Record heads 120-123 are mounted over bands 116-119 and permanent magnet erase head 124 erases all tracks. Playback heads 126-129 are mounted on movable brackets 130-133. The drum 115 driven by belt 136 which passes over pulleys 134 and 137. All heads are in the same curved plane, relative to the axis 138 of rotation. Various delays can be obtained by simple rotation of the pickup heads by moving the brackets 130-133 back and forth relative to each other. Note that a "negative" time, can theoretically be placed into the system if desired.

The three systems illustrated in FIGS. 12-15 will have the same need for the same band speeds past the pickup heads of the recorded material and of the same order of magnitude, and for the same amount of displacement each from the other, depending on the control of that speed, to effect the necessary precise sculpturing of the eminent sound.

Thus, it is seen that we can literally "carve-out" a solidity (stereo) of sound, to the degree we wish and that we can shrink, expand, or even distort the resulting amount of sculpturing as desired. We can also, actually move the location of the original sonic source around our listening volume to any place that we so choose. It is necessary to "delimit" the sound reception of any specific volume by use of the "at its limits" limiting sonic-pickups, and do the same for the playback room volume.

It is difficult to duplicate the sound components of an auditorium in a small room if the primary recording volume is highly non-structured in the first place. However, if recording is done in a pie-shaped auditorium as shown in FIG. 16, compensation and sculpturing of the listening volume by a mere adjustment of the delay controls may be accomplished since distances d ₁ and d ₂ will be determined by D ₂ -- D ₁ which can be placed into the delay channels as a negative override, if required. FIG. 17 illustrates an actual listening volume 140 in the corners of which the sounders have been mounted which appears as the larger pie-shaped area.

In FIGS. 18 and 19 an embodiment of the principle of time/spatial sonic-modeling as applied to a regular dual disc-type recording is shown. In this case, the record 141 has a standard bi-lateral, stereo groove 142 in which record channels 1 and 2 are recorded, and an adjacent stereo groove, also of bilateral stereo, for channels 3 and 4 are recorded. The tone arm 143 instead of using a single needle for picking up the first and second channels of stereo, (since this would provide only a zero time displacement at that point), uses a double needle in a single or double cartridge. Each of these may then be displaced relative to each other and to the other double-needled cartridge in the adjacent channel. Needles 144 and 147 rest in one groove 142 of the record disc while needles 146 and 146 rest in the adjacent groove. Therefore, bilateral, adjacent groove, horizontal recording allows achievement of the four channel effects desired. The needles 144-147 may be respectively adjusted by screws 148-151 for the delay setting.

Thus, this invention results from the realization that sound is a 3-dimensional phenomenon, which when recorded in any chamber becomes an n ^th -dimensioned resultant (due to the multiple additions and reflections, as well as selective absorptions). By providing recording and playback equipment as shown which takes these facts severely into account, true holostereo (solidity) sculpturing of the sonic materials far superior to any stereo system previously in existence is obtained. The invention also allows replication or duplication of the original time/spatial-dimensions of the original recording chamber, thereby truly achieving a faithful duplicate of the sonic matrix created in that original recording chamber. Indeed, if desired, even non-faithful, but nevertheless, intriguing effects can be obtained wherein one literally rotates or moves the orchestra or singers around at will.

It should be understood that various modifications to the method and techniques as herein shown can be made without departing from the spirit or intent of the invention's novelty or scope.