Originally Posted by wogoga
All of them.
You seem to imagine that photons emitted at discrete times can have a single frequency. But this is not the case. In order for a photon to be emitted within a specified time window, its spatial extent must also be finite, which means (ala Heisenberg) that its momentum will have a minimum uncertainty as well, which in turn means an energy (and frequency) uncertainty.
You can consider a monochromatic beam of light to be a superposition of lots of individual photons, each of which is has some spread in momentum and frequency and hence is spatially localized. But that superposition, in order to form a monochromatic beam, will end up cancelling components of each photon that vary from the frequency of the beam itself.
If you construct your monochromatic beam where the phase shifts for each successive photon you're building your beam from, then the non-cancelled frequency will be shifted. Your continuous phase shift will still be the same thing as a frequency shift.
No. What you mean by "phase jump" doesn't correspond to what the authors of your source mean by "phase jump".
Sure, but they all still involve conditions under which stimulated emission dominates spontaneous emission, as it does in a man-made laser. None of it is applicable to your theory.