Issue with Unlimited is that it provides the tools to kill off smaller competition. Only a very few pools, the biggest of all will survive in the long run. With that said, I do support bigger blocks, but ideally implemented in a different way.
A better L1 dynamic capacity increase is hard-coded (i.e. 5%, or flat 0.5MB annual increment). Or any other number. A hard coded dynamic block size solution is by all counts better than the one involving miner signalling. "The free market" should not decide, because majority hash rate is not the market.
- Initial consensus change will be a hard fork, but that's all. Should there be an undesirable outcome in relation to ever-increasing block size, a simple soft fork can put a full stop to growth, and another simple soft fork can resume the growth.