Xjyuk

I just read about how miners have to change the nonce bit in order to find an hash below a certain result.

Granted each miner process it the same way, incrementally, adding 1 and hashing, adding one and hashing, what prevents the most powerful computer on the network to snatch all rewards?

I feel like beside the moment each miner starts mining a block, there's no room for luck and a lesser powerful machine, to ever add a block.

Because each miner may select the transactions (and the order in which they are included) in each block. Also, they do not all start on the same exact nonce, in fact they are changing more than just the nonce in many cases (e.g. block versions - see ASIC BOOST).

This means that they are not all performing the HASH256 (proof of work algorithm) on the exact same data, so the outcome is random. However, more powerful computers can perform more hashes per second, so they will find a solution on average, more often than a less powerful computer. These computers, by the way, are specialized hardware called miners, not general purpose computers.

Just adjusting the nonce is not sufficient for mining a valid Bitcoin block at the current difficulty level which is around 80 exa-hashes (80 x 10¹⁸ hashes) per second. Nonce field in the block is only 4 bytes which means it can at most provide you 2^32 different hashes(4.2x10⁹ hashes). That is the reason, miners adjust the extra nonce value in the coinbase transaction and then iterate over the entire 2³² nonce field again.

Supercomputers are way slower than ASICs for mining purposes. Supercomputers contain huge CPU power that are highly useful for general purpose computing however, for running a particular task they will be much slower compared to a hardware that is designed specifically for that purpose. That is the reason miners use ASICs where the dsha256 algorithm is printed on the circuit in the hardware. These chips can't do anything else, but run the double SHA256 hashes which provides them advantage over even a supercomputer.

Miners generally run another machine connected to these ASICs chips. While the ASIC chip is exhausting the nonce space, the other machine changes the extra nonce value and queues it up for the ASIC hardware to take it from there. So these task can be run in parallel.