First we know that photons are elementary particles that function both as a wave or a particle at a certain time though not both at once.
We know that we want to use either or for the production of bits of information. Particles for individual computations and waves for larger computations if enough bits of information can be read correctly at the same time waves may be superior at the cost of space used to propagate an answer. A single wave offering massive functionality within a larger area though they can be condensed if you refract the waves up one the nuclei of the atoms within each other without cross sectioning each other—though that may also be useful as an additional functionary device within the function schema. Perhaps if they cross they mean they multiple or if they don’t they remain untouched. If they interfere with an electron they divide. Something like that as a simple example. Individual particles get us exact answers over and over if read repeatedly. Very quickly and take little energy to wave reading unless you collapse a wave function into a rotational amount feasibly readable as though they were particles. You don’t need waves to read waves if you’re looking to collect specific bits of data Just sets of particles to oscillate and deform the readable wave function. You do need if you want a single read of an entire function in a go in one mass deformation. So you build a function map using all types of functions you might possibly want around in such a way that the photons moving both backwards and forwards/sideways in time have a chance to interact with the system. What you may be reading isn’t in fact the current set of photons but photons from the past passing through the system in reverse but since the male is segmented so that it repeats itself at various angles/degrees it may not matter. But you would know this by first placing it in the center of a room and bombarding it with photons and seeing which side lit up first. My understanding is that there are four positions from Feynman’s research a photon travels based on their delivery.
So eventually we get good enough at bouncing these waves and particles of photons throughout the system of lattices to get multiple answers at once Multiplied by the number of deformation levels we want (each wave particle is computer generated to deform through the run systematically causing little to no error outside of the lasers need to be on the best gimbal possible). The thing about setting up a function map physically is that you can pick and choose which bits you want read from the read direction. But anyway we get good enough we get down to the gluon method using QCD which has eight methods of propagation. And that’s when we start bouncing gluons off of quarks in much the same method reducing the size exponentially further but in much the same manner. Then the real fun begins.