Light and its accuracy

So why is the relative sample rate of light not constant?

The accuracy of the null bubbles, as powers of 3/2,  matching powers of Phi stop, somewhere around Phi^107.  Mainly because, I think, the ratio that matches Phi better comes from:

Phi^n = a * Phi + b

At some point, that ratio gets better faster than the error term from (3/2)^m can cycle.  But does that mean the light is not constant? Light rate is based on the exchange rate, and those are Null bubbles, so I would not expect light to be more accurate.

There is a point where the rational approximation of Phi, at Phi^17, does indeed drop to the approximation error of Nulls at (3/2)^127.  So there are packings too small and packings too large. Three points seem to match.  Phi, estimating itself, needs a quant of (3/2)^17, which more or less matches the electron matching Phi with (3/2)^89, which matches the error at Phi^107, and in between is the nearly perfect match at the proton peak. So, (3/2)^17 seems to be the smallest thing Higgs can handle. So, something like 1/(3/2)^17 looks like the smallest fraction. The number I throw out are seriously rounded, I have not worked the spread sheet in this.  But (2^17-1) makes a 16 bit system at the peak of the proton, so I am merely guessing that that is the number. The smallest thing is packed nulls, that does not include wave kinetic energy.

Bubble size?
There must be a difference, but the two wave bubbles would not change size to allow better packing a low quants. Trying to make light rate variable would remove all the stability of packing. I would guess that the phase error in exchanges corresponds to the minimum packing ratio, simply on minimum redundancy grounds of how the bubbles are set in the quasar.