What Makes the 'Hard Problem' of Consciousness So Hard?

Cognitive scientist Donald D. Hoffman explains that mechanistic models of cognition can't explain the "hard problem" of consciousness.

[andrii.personal]

It seems like he is not in touch with neuropsychology and modern theories of consciousness\feelings, which is strange for cognitive scientist. His vision of us as "not machines" may be nascent from our defense mechanism - to see ourselves as unique, exceptional creatures. From this point of view all our "feelings" and "experience of smelling cofee" seems so unrepeatable. But going this subjective way we are ignoring the simple thruth - all this "feelings" are special just to us because of their subjectivity.
I mean, we can understand how is it to be human, but also we cannot understand how is it to be a computer.

Our feelings is just a way our brain represents our demands or our body responses to the environment.
The biggest neuro-problem for today is to find those objective units in which we can measure and identify those conditions of our body (whether it synapse interconnections or whatever).
Thereby "our experienxe" is not something extraordinary, its just so special to us.

[theultimatereductionist]

I am ultra reductionistic. I am Mathist. To me, mathematics rules everything. NOWHERE does this imply that I think physical reality doesn't exist. OBVIOUSLY it does. Hey, perhaps OTHER physical realms exist, even after life, even before life. But, at the same time, it is super important to NOT ASSUME those alternatives, speculations as any sort of "fact", but to remain vigilant, super-critical, and continuously adjusting our view according to the evidence/knowledge at the time. And at this time, we have zero evidence of consciousness outside current physical bodies.

But, THAT does not solve the question: WHY can't consciousness exist in RANDOM or (better word) ARBITRARY collections of atoms, or even quarks or other particles?