Quantum superposition is the crucial component of Quantum Mechanics believed to open a new world of quantum computing on quantum computers.
An $n$-qubit quantum computer with $n=2,3,..,$ is supposed to operate in parallel on $2^n$ states in superposition, to be compared with a classical digital computer operating on $n$ bits at a time, where a bit can take the value 0 or 1. Quantum computing thus gives promise of exponential speed-up vs digital computing.
The idea of quantum computing was first presented by Richard Feynman in an attempt to get around the apparent exponential complexity of Quantum Mechanics QM based on Schrödinger's multi-dimensional wave equation making digital simulation impossible by demanding computational work scaling with $2^N$ for a system with $N$ electrons. The idea was to replace digital simulation with some form of analog quantum computation where the simulation of a quantum system would be performed on a quantum system. An intriguing idea, but could it work? The apparent exponential complexity would then be met with an exponential computational power making simulation possible. To meet a perceived difficulty by ramping up the capacity or simply to "fight fire with fire".
Let us then consider the basic idea of a quantum computing, which is
- Simultaneous operation on states in superposition.
What then is "superposition"? The mathematical answer is the following: Consider a linear algebraic or differential equation with solutions $\psi_1$ and $\psi_2$. Then the algebraic sum $\psi =\psi_1+\psi_2$ is also a solution and $\psi$ is viewed to be the superposition $\psi_1$ and $\psi_2$ with the + sign signifying algebraic sum.
As concerns classical physical wave mechanics, the algebraic superposition can take two forms with physicality of (i) the algebraic sum $\psi$ or (ii) of the individual terms $\psi_1$ and $\psi_2$. Case (i) represents the interference pattern seen of the surface of one pond, while (ii) represents the beat interference generated by two vibrating strings with nearby frequencies.
As concerns QM the physicality is supposed to be displayed in the double-slit experiment: Let a single photon/electron pass a double-slit and then be detected on a screen behind a as a spot. Repeat the experiment many times and notice a fringe pattern appearing on the screen, which is the same interference pattern developed by a macroscopic wave passing through both slits. The photon/electron after passing the double split is representing the quantum superposition supposed to carry quantum computing. This is not a superposition of realities as in (ii) above, but a superposition of possibilities made real by repetition of the one photon/electron experiment which represents the physics of a 1-qubit: The single photon/electron is by passing through the double slits put into a superposition of passing the left slit and passing the right slit not as realities but as possibilities. It is here essential that the superposition only concerns one photon/electron in superposition of $2^1=2$ states. This is supposed to generalise to $n$ entangled photons/electrons in superposition of $2^n$ possible states.
The evidence that quantum computing is possible thus boils down to the double-slit experiment. If one photon/electron can be put in superposition of two states which appears to support interference with fringe pattern, then constructing of a $n$-qubit computer may be possible. If two photons/electrons are needed for two states then we are back to classical computing with bits.
The crucial question is now: Is it sure that because there is one click on the screen at a time, the input is one photon/electron at a time?
Think of this, and return to see a more precise analysis.
QM in its standard multi-dimensional form has exponential complexity, which requires exponential computing power. RealQM is an alternative with polynomial complexity which can be met by classical computing.
Maybe quantum computing is neither possible nor needed?
