![[H]ard|Forum](/styles/hardforum/xenforo/logo_dark.png){kind=logo}
https://phys.org/news/2020-01-computing-with-spins-of-light.html
Allow me to mislead with my own incomplete grasp. I get the impression this might
be describing a kilometer long optical fiber parametrically amplified memory loop.
Holding, amplifying, phase locking, possibly doing logic on many bits round-robin.
"Non-degenerate" could mean not the usual case of locking into only two states.
Any system parametrically excited, can only amplify two opposing phase states.
Those statements are not in conflict if the path is long enough to contain more
than one state. Long enough to contain more than one wavelet. Solitons?
One weird (or non-weird) thing about optical parametric amplification is that the
exciting clock waveform travels through non-linear media in the same direction
and speed with the 1/2 frequency wave being amplified. Its not at right angles.
Doesn't need to prevent mixing, cause a prism can separate clock from data...
So, these amplifiers are uni-directional. Unlike the parametrons we might have
discussed before, that hold one bit, amplify in one spot, and don't distinguish a
direction of travel. These amplify as they go, together with a clock.
A laser might bounce back and forth between two mirrors as it gets amplified.
It amplifies in both directions simultaneously, which of course has its own uses.
But these get to bounce around several mirrors, or take the path of a fiber and
come back through in the same direction each pass. The optical fiber itself
might provide the required non-linearity. I think thats also a requirement for
travelling solitons to hold thier shape.
But I describe two linear polarized phases, not two spins, or how would that
work exactly? Much here I cannot yet resolve and most likely got wrong.
I suppose if you superimpose another wave at 90 degree polarization and
phase lock, it becomes circular. Is a circular polarization considered spin?
These articles are dumbed down to clickbait badly enough that my take is
probably no worse. The intro seems to describe an ising machine, which
is a 2D array of frustrated qubits and something else entirely. I don't know
of anyone solving the travelling salesman with a 1D loop of parametrons,
an ising machine, or anything else.
Editors conflate things they don't understand as-if doing so was explaning.
Edsplaining. Let me edsplain you how this works, see...
Allow me to mislead with my own incomplete grasp. I get the impression this might
be describing a kilometer long optical fiber parametrically amplified memory loop.
Holding, amplifying, phase locking, possibly doing logic on many bits round-robin.
"Non-degenerate" could mean not the usual case of locking into only two states.
Any system parametrically excited, can only amplify two opposing phase states.
Those statements are not in conflict if the path is long enough to contain more
than one state. Long enough to contain more than one wavelet. Solitons?
One weird (or non-weird) thing about optical parametric amplification is that the
exciting clock waveform travels through non-linear media in the same direction
and speed with the 1/2 frequency wave being amplified. Its not at right angles.
Doesn't need to prevent mixing, cause a prism can separate clock from data...
So, these amplifiers are uni-directional. Unlike the parametrons we might have
discussed before, that hold one bit, amplify in one spot, and don't distinguish a
direction of travel. These amplify as they go, together with a clock.
A laser might bounce back and forth between two mirrors as it gets amplified.
It amplifies in both directions simultaneously, which of course has its own uses.
But these get to bounce around several mirrors, or take the path of a fiber and
come back through in the same direction each pass. The optical fiber itself
might provide the required non-linearity. I think thats also a requirement for
travelling solitons to hold thier shape.
But I describe two linear polarized phases, not two spins, or how would that
work exactly? Much here I cannot yet resolve and most likely got wrong.
I suppose if you superimpose another wave at 90 degree polarization and
phase lock, it becomes circular. Is a circular polarization considered spin?
These articles are dumbed down to clickbait badly enough that my take is
probably no worse. The intro seems to describe an ising machine, which
is a 2D array of frustrated qubits and something else entirely. I don't know
of anyone solving the travelling salesman with a 1D loop of parametrons,
an ising machine, or anything else.
Editors conflate things they don't understand as-if doing so was explaning.
Edsplaining. Let me edsplain you how this works, see...
Last edited: