I also can t understand why hashrates accumulate.
I guess it will be the same result when several computer calculate with difference addresses.
Or even that find a block have higher probability on differences address because it is possible that
on the differences machine there calculate overlapping hashes. Something wrong in this thinking?
This is the super simplified version.
Mining is the action where your computer guesses for a very low number, the computer randomly guesses for these numbers.
Eventually your computer may guess the right number. Once your computer does that you earn some reward. (Solving a block).
Because the search for these super low numbers is random, you can have any number of computers all searching together, they don't need to be linked together or anything. They stack in the sense that you have more computing power going towards the search.
The address you set as your mining address in the mining program does not matter, you can have each machine mine for a different address or the same address, it makes no difference at all.