The ironic thing is that these forks will actually drive companies like bitmain to make the miners and never sell them to the public, thus causing real centralization. Most of the people who complain are those who spent a lot on GPUs and refuse to adapt to the future.
Bingo! You speak the truth brother.
here's the rub; if the average joe miner wont make a profit; it will be down to the major manufacturers only; thus driving the value down. IF they are the only ones mining and selling the vast majority; well... they could hold out for an epic dump in the future... but I doubt companies like BMT can work in such a way.
notice how mineable coins always raise in value up until asics? Theres a reason behind this.
Instead; if they take their heads out of their asses, like SHA256 manufacturers behaved early on; things would be more similar to the BTC side of things on the mining side with all the newer asics. If they operated in a forsenCD (transparent) manner about their projects; they would be trusted more and less of an overall problem.
Theres always room to argue more points further; but this is the root of what I see that matters most out of the rest of the details.
Again to comment about my re-quote previously:
we need to change the meta.
purely calculating power was the meta in the past. ASICS are best at calculating and you want to limit their effectiveness.
Don't focus on the calculation side. focus on changing the meta surrounding the calcs so that the speed of the calculation itself, is less of a concern.
We can also do things like penalties in the blockchain code for 'n' blocks reported by single sources within 'x' time.
Whittle it down small enough, and you end up with only being able to mine a single GPU from a single wallet if you so desired. Balance 'n' and 'x' based on the current hardware that is known to be available.
Such a simple and elegant solution to large hashrates.
Just think outside the box guys.
Change the meta.