Question About Simultaneously Mined Blocks and the Possible Consequences

[butka]

This is a question that has been lingering in my mind for some time. I think I know the answer but maybe I'm wrong. Hopefully someone can clear this for me.

In a totally decentralized peer to peer network, two miners relatively far from each other (if you can say this) mine a block approximately at the same time.

They announce their solutions, and we have a situation where two blockchains propagate through the network. Each one of them is being accepted by their closest local nodes and continues to spread out.

Roughly speaking, after some time we should have a situation where half of the network has accepted one blockchain, and the other half of the network another.

If this logic is valid, than the first half of the network should already be mining a new block on top of their version of the blockchain, and the second half on top of theirs.

Again, if this logic holds, is it so that the one who mines the next (second) block decides the overall winner?

If the second block is found within the first half of the network, their blockchain is the ultimate winner (as the longest one), and the other half of the network has to revert to the solution of the first half and abandon their version of the blockchain.

Is this indeed so?

[1715018528]

1715018528

[1715018528]

1715018528

[Xynerise]

Nodes store both copies until one with a cumulative higher proof of work emerges.
See my post here for details on blockchain reorgs

[etherixdevs]

It can happen. you will have two competing blocks. Let's call them A and B.

Assume A was solved just a second or two before B. This means A is likely to have propagated to most of the network than block B, but let's assume it's not a huge margin.

Other miners who received the A block before receiving the B block will try to mine on top of A. Miners who received B first will try to mine on top of B.

If the next block to be found is on top of block A, the miners mining on top of B will abandon B and go back to mining the A chain. The reverse is true if a block were to be solved on top of B first.

Of course, a block could be mined on top of A and another on top of B at roughly the same time, causing the same problem again. The competition between chain A and B will just continue until it gets resolved. This is part of the reason why most people suggest waiting for multiple blocks of confirmation before assuming a transaction is non-reversible.

The block that gets abandoned is called a an orphan block and any chain built on top of it is an orphan chain.

[butka]

Nodes store both copies until one with a cumulative higher proof of work emerges.
See my post here for details on blockchain reorgs

I see, thanks for the link and for clearing this up. After all, it is a little bit more complicated than I thought.