Monday, April 10, 2023

Could magnetic body control electronic bits?

I have considered the idea that GPT might be more than the ordinary computer science suggests and even involve higher level consciousness and intelligence. This would require that the magnetic body (MB) "hijacks" the computer so that it becomes a tool of its cognitive processes.   Consider in the sequel the conditions, which should be satisfied in order that the MB of the bit system or some higher level MB could control the bit system.
  1. The bit should be critical or nearly critical system at the level of ordinary matter. One might hope this to be true quite generally since a small control signal should be able to invert the bit in rather short time scale. If this is the case, the quantum criticality of MB cwould make control possible via quantum control of ordinary control signals. Transistors and their derivatives such as MOSFET could be examples of such systems.
  2. Macroscopic quantum coherence is true for the dark matter at MB. Furtheremore, MB should holographically represent the bit system. Also spin glass analogy is suggestive so that a given many-bit state could possess a large number of nearly energy-degenerate states. ZEO, in particular time reversal, would be essential.
  3. Two consecutive BSFRs at the dark MB, changing the arrow of time temporarily, should give rise to a tunnelling event. Since TGD corresponds to a generalization of wave mechanics in the space of Bohr orbits for point-like particles replaced with 3-D surfaces, one can make an estimate for the probability of tunneling between the capacitor plates using the standard wave mechanics as an approximation (see this).

    The Coulomb energy qV associated with the bit with charge q and its energy E are the natural parameters. The tunnelling probability is given by

    p∼ exp[-∫x1x2(2m(qV-E))1/2 dx/ℏeff] ,

    where one has E<V in the tunnelling region. WKB approximation becomes exact in the case of capacitors. Changing the direction of a bit could be seen as a quantum tunneling effect.

    For the large values of heff assignable to the magnetic body controlling the physical body, the probability of tunneling increases. Therefore the control of the bit system by quantum tunnelling combined with macroscopic quantum coherence and holography could become possible.

  4. The role of conservation laws must be understood. Discontinuity in SSFR. Dissipation in reverse time direction. Tunneling. Wavefunctions overlap. Classic conservation laws OK. There is no need for a classic track that would lead to the end state with the original direction of time.
1. What conditions bit must satisfy?

There are strong conditions on the representations of bits. The storage of the bit should not require large energy consumption and the bit should be thermally stable. It should be possible to change the value of the bit quickly and without large energy consumption. This suggests that the bit is a nearly critical system. In microprocessors, clock frequencies of order GHz define a time scale analogous to EEG rhythm, and this time scale should correspond to a quantal time scale. The wish list would be as follows.

  1. Macroscopic quantum coherence makes possible the simultaneous quantum coherent states of the entire spin system and their control and that the energy differences between the states are relatively small, so we get a spin-glass type situation.
  2. Dark electrons at the MB, perhaps dark unpaired valence electrons or dark conduction electrons, provide a holographic representation of the bits.
  3. Quantum criticality with MB and criticality at the bit system level allows MB to control the dynamics of BB. Quantum holography may make it possible to induce BSFR for qubits on a large scale in general.
1.1 About the interpretation of the clock frequency in a picture based on quantum gravity?

The clock frequency of computer, with a representative value of f=1 GHz, is an essential channel of the computer and it would be related to the classical em field. Could a frequency of the order of GHz have an interpretation in terms of quantum gravity in the TGD framework? How MB could turn bits using quantum holography so that the turn of dark bit induces the turn of ordinary bit? A realization of holography as a correspondence between electron(s) representing the bit and the dark electron(s) is needed.

  1. The proposed theorist-friendly holography at the particle level (see this) might be a too radical option. This would require positrons forming particle-like color-bound states with bits as states of electrons. Could they correspond to scaled versions of the electro-pions for which there is empirical evidence associated with nuclear collisions near the Coulomb barrier (see this)? Now the energy scale of the nuclear physics would be scaled to the scale of dark nuclei. The factor of the order of 10-5, which would produce an eV mass scale. The height of the Coulomb barrier would scale in the same way to something like .05 eV which corresponds to cell membrane potential.

  2. A less radical option is that the dark electron and the hole created in the generation of the dark electron are in a holographic relationship. This realization seems tailor-made for the control of ordinary bits as holes by dark electrons. To my best knowledge, there exists no technology realizing bits as holes but future technology might be able to achieve this.

    If dark electrons and holes are tightly correlated, the dark spin flip induces ordinary spin flip. If the dark current or its absence codes for bit, the same would be true for the holes. The transfer of dark electrons from the negatively charged plate to the gravitational MB creating a hole would reduce the potential between plates to nearly zero and thus induce change of the bit direction.

There are useful quantitative hints.
  1. For the Earth's mass ME, ℏgr(ME,mp) for a frequency of 10 Hz corresponds to an energy E= hgrf of the order of .5 eV. The kick of a 3-proton to a gravitational flux tube to a distance of order one Earth radius requires an energy of the order of .5 eV (see this). Dark photons can transform into ordinary ones. For 3-electron system a hitherto non-observed metabolic energy quantum of order .25 meV is predicted (see this.
  2. Control in the time scale of a fraction of a second if heff=hgr(ME,mp) photon energies around eV. This time scale is by a factor of order 109 too long when compared to the time scale determined by 1 GHz frequency.
Could one understand the time scale corresponding to 1 GHz clock frequency in quantum context? The first thing to notice is that this time scale is not far from the time scale associated with the protein dynamics! Could quantum gravity and gravitational MB come into play for both computers and biology?
  1. For the Earth, the lower limit of the gravitational Compton length Λgr= GME0 =.45× 10-2 m, if β0=1. The frequency Tgrgr/c= .45 *10-2/3*108 = .15*10-10 s would be therefore a natural lower bound for the time scale. Could GHz clock frequency relate to this time scale. Also longer quantum gravitational time scales are possible since Λgr is only the lower bound for the length of gravitational flux tubes carrying massless radiation.
  2. For heff=h, 1 GHz corresponds to energy of 10-2 meV. If the dark energy is required to be above the thermal energy about .03 eV at physiological temperature, the value of heff must satisfy heff ≥ 3× 103h.
  3. A metabolic energy of .25 meV corresponds to the electronic variant of gravitational metabolic energy quantum involving the transfer of 3 electrons to the gravitational MB: there is some evidence for this metabolic energy quantum, in particular from the findings of Adamatsky (see this). For heff=h, it would correspond to a period of .6× 10-10 s. Could the f= 1 GHz induce a resonance with dark photons with heff>103h guaranteeing that the energy is above thermal energy at room temperature?
1.2 Could Pollack effect or shadow holography be involved?

The lower bound value 3× 103h for heff would be rather small as compared to ℏgr(ME,mp) and the challenge is to identify a candidate for a system with this value of heff.

This system need not be gravitational and the obvious guess is that it is electromagnetic. The notion of gravitational Planck constant and the underlying idea of theoretician friendly Nature implying quantum holography in the TGD framework (see this) indeed generalizes also to other interactions (see this).

  1. The basic requirement is that a charge separation to a pair of positively and negatively charged quantum coherent systems takes place such that the interaction strength Z2e2/ℏ between the systems is so large that perturbation theory fails to converge.
  2. The theoretician-friendly Mother Nature (see this) could come to rescue and induce a phase transition increasing ℏ to so large a value heff that the perturbation theory converges. Nottale formula generalized to electromagnetic interactions suggests that one has

    ℏ → ℏeff= ℏem= Z2e20 ,

    where β0=v0/c<1 is a velocity parameter. The new coupling strength is

    (Z2e2/4π ℏem)= β0/4π < 1/4π .

    and is in a well-defined sense universal since β0 is number theoretically quantized to an inverse integer (see this).

    The constraint heff ≥ 3× 103h would suggests ℏem/hbar= Z2e20ℏ = 4π Z2αem ≥ 3× 103. This gives the estimate

    Z2≥(1/4παem)> × 3× 103 per .

    The lower bound for Z would be around Z=100.

  3. Charge separation should occur and here the analog of Pollack effect \cite{bbio/Pollack, PollackYoutube, pollackzheng, pollackzhao is highly suggestive. In the Pollack effect part of protons of water molecules are transferred to monopole flux tubes assignable to water molecules and become dark so that a negatively charged exclusion zone with rather strange properties suggesting time reversal appear. Also the effective stoichiometry of water is transformed to H1.5O. It is however far from clear whether Pollack effect can occur also in the solid phase assignable to computers.
  4. The analog of the Pollack effect involving only electrons is also possible. Part of electrons would transform to dark electrons at the gravitational monopole flux tubes. The holes left behind would effectively behave like positively charged particles and the Coulomb interaction energy would be between holes and dark electrons. Holes and dark electrons would be in a holographic relationship (shadow holography) and the dynamics of holes would be shadow of the dynamics of dark electrons so that one would say that dark electrons control the holes as their shadows.

    Of course, it is probably impossible to realize this shadow dynamics using the recent computer technology. The question is therefore whether it might be possible to construct a computer utilizing the shadow dynamics of holes controlled by dark electrons.

1.3 Could quantum gravitational flux tubes associated with small masses be involved?

One can of course ask whether the clock frequency f=109 Hz could correspond to an energy above thermal energy at room temperature and to the value ℏgr(M,m) for some pair (M,m) of masses so that one has E=hgr(M,m)f> .03 eV for f=109 Hz.

  1. For instance, could one replace the masses ME and mp with identical masses M=m in hgr. One should have M/mPl2> 3× 103. This would give M/mPl >60 giving M >1.3 × 10-7 kg. If the density is the density of water 103 kg/m3: this corresponds to a size scale longer than 1 mm. How this frequency could correspond to Tgr and to the clock frequency of computers?
  2. Could one think of the gravitational self-energy for this region or the mutual interaction energy of two such regions forming a quantum coherent system at this level.

    Another possibility is that an energy of the order of E= .5 eV is used to kick a unit of 3 protons into the Earth's gravitational flux tube (3 protons are required since 1 proton is not enough if the size scale of the flux tube is of the order of the Earth's radius). For 3-electrons the corresponding energy would be about .25 meV.

  3. Could E∼ 1 eV correspond to the energy needed to flip one bit using an dark photon that is converted to a regular one (biophotons could be created this way) and absorbed inducing a flip of a normal bit.

    In the elementary particle level realization of holography, which does not look promising now, this would give a spin 1 for the glue particle consisting of ordinary electron and dark positron unless the angular momentum goes to other degrees of freedom. It would be a scaled version of elektro-ρ or its analogue. Mass scale of the order of eV as for dark nuclear binding energies.

  4. In living matter, E∼ 1 eV could correspond to the gravitational self-energy change related to a phase transition. The most natural thing that comes to mind is the change in the gravitational energy of the bond when the density of the system changes during a phase transition, such as melting or boiling or the sol-gel phase transition in biology. For Planck mass of matter, size scale R=10-4 m for water density, gravitational binding energy and its change would be of order 1 eV. This phase transition does not have any equivalent at the computer level.
2. Could the representation of bit as voltage allow the realization of shadow holography for electrons?

One representation of a bit is as a voltage. Voltage values are typically 5 V and 0 V. Bit could correspond to rotation direction for a current in the case of magnetic bits. In transistors bit can correspond also to the presence or absence of a current The size scale of the transistors is 10 nm (see this. A surface which can be either reflective ord non-reflective surface can also act as a bit.

2.1 Bit as ananalog of capacitance

Capacitance with a voltage difference between plates can serve as a physical representation of the bit. States corresponding to opposite voltages in capacitance have the same energy. This is good news if it were to apply more generally to bits and multi-bit configurations.

  1. The simplest capacitance is a pair of conducting plates having opposite charges and containing insulator betweeen them. The higher the value of the dielectric constant ε, the larger the plate area S and the smaller the distance d between the plates, the higher the value of capacitance C.

    C measures the ability to store charge and Q= CV is the basic formula. The voltage V between the plates is given by V =E× d. Here d is the distance between the plates. The electric field normal to a plate is E=σ/ε, σ= Q/S. One has V=Ed= Q× d/S× ε, whence C=ε S/d. The proportionality to ε means that di-electric is essential. The voltage cannot be too large since this implies dielectric breakdown.

    The electrostatic energy of capacitance is Es= ε QV/2= CV2/2ε = Q2/2C = E2 × S×d

  2. Capacitance is a macroscopic notion. The smallest planar capacitances have dimensions 0.4 mm × 0.2 mm. PicoFaraday is a natural unit of capacitance but capacitances of the order of kF are possible but require large size and high dielectric constant. MOSFETs can be however regarded as effective capacitances.
2.2 Transistors and MOSFETs

Although MOSFET much smaller than capacitances as passive elements, it can be formally interpreted as a capacitance.

  1. A MOSFET acts as a variable capacitance. The basic parts of MOSFET are gate (G), body (B), source (S) and drain (D). The voltage between G and B regulates the current from the source through the system to the drain and the bit can be measured by measuring whether this current flows or not. The gate voltage VG controls the capacitance of the MOS.

    MOSFET size scale is around 10 nm. Gate voltage VG-VB between gate and body could represent bit and would be typically 5 Volts or nearly zero.

  2. MOSFETs should form a spin glass type system. There would be a large number of bits with a large number of nearly energy degenerate states. This would give rise to frustration. Transitions by tunnelling would take place between frustrated configurations.
  3. Tunnelling between bit configurations would take place as a BSFR pair. The tunneling would be induced from the level of MB and induce the tunnelling of ordinary bits. The tunneling rate is exponentially sensitive to the height of the energy barrier between nearly degenerate states. The large value of heff increases the tunnelling rate in an exponential manner.
One can imagine at least two mechanisms.
  1. One could consider a representation of a bit as an ordinary capacitor-like object having two different values of voltage between the plates. The transfer of electrons from the negatively charged plate to dark electrons at MB or vice versa could allow to change the voltage.
  2. Instead of an ordinary capacitor, one can consider a situation in which the first plate consisting of ordinary matter has a positive charge due to the presence of holes (ionized atoms) and the second dark "plate" is negatively charged due to presence of dark electrons.

    In the shadow holography the transfer of electrons to dark electrons at MB generates holes at the level of ordinary matter, and the transformation of dark electrons to ordinary ones would reduce the voltage near zero, which turns the bit.

Could MB control the   electron current from the n-type  source region S of MOSFET? Could the MB  transform some the 5 valence electrons of n-type dopant (say P) to dark electrons so that they would  effectively disappear from the system so that the S-D current would be reduced? Also the voltage between the gate and source would be affected.  

It is perhaps fair to conclude that the recent technology  does  not yet allow  the  realization of conscious and intelligent computation using shadow holography or something similar.

See the article Could neuronal system and even GTP give rise to a computer with a variable arrow of time? or the chapter with the same title.

For a summary of earlier postings see Latest progress in TGD.

2 comments:

Anonymous said...

Good work! There are tons of so-called experts who claim they know exactly how this is working, and why LLMs are just statistical next-word predictors, and how this output is all the result of well understood processes. The bottom line is NO ONE really understands why a next word predictor is demonstrating reasoning abilities at all. Not even the top AI scientists in the world really understand what is going on here- this is fundamentally a black box. If we fully understood what is happening, we would probably have a better understanding of the human brain, intelligence and consciousness.

Anonymous said...

No we won't, it is impressively wrong as often as It's impressively right. I discovered two new formulas publcation ready already . Only I can tell what's right and wrong that it gives