3

During the discussion of Bitcoin’s scalability, I’ve read the statement that

"Large miners benefit from larger block sizes, because it makes it more difficult for small miners to compete." [1]

In the linked comment this is used as fact without explanation, but it isn’t obviously true at first glance. Please explain: Do larger blocks make it harder for smaller miners to compete? Why?

  • I’m aware that this may produce answers that are opinion based, yet believe that answers well supported by explanation could add progress to the community’s discussion. Answers without explanation will be removed. – Murch Jan 3 '16 at 16:45
2

Yes. There are reasons at multiple different levels.

  1. Propagation related.

It takes time for a new found block to propagate around the world to all other miners. Bandwidth, network latency and validation times play a roll. It's also important to keep not just worst case numbers in mind, but even adversarial case. The average case isn't very relevant.

So assume propagation takes 1 minute (bit extreme but not unheard of), that means that miners that hear about the block 1 minute after it was created, have just wasted 1 minute of hashing work on the wrong block.

Assuming all else equal this will happen to large and small miners. However a 30% large miner will NOT have that problem for 30% of the blocks (their own blocks). Small miners will only NOT have it for their own 1%.

Similar logic applies to orphan blocks: 2 new blocks found by different miners at the same time. A 30% miner has a good chance of finding another block on top of their own, while for a smaller miner (all else equal) that chance is negligible. Meaning smaller miners will have a higher orphan rate.

  1. Economies of scale related.

Almost by definition economies of scale are favouring larger players.

All kinds of discounts and better deals you can make when buying in bulk: equipment, electricity, accommodation, staff, security, DDOS protection, fast redundant internet connections... Pretty much everything you can get faster (latest technology) and cheaper when you're bigger.

Economies of scale are sort of naturally limited by diminishing returns though. At some point electricity doesn't become cheaper anymore if you buy more and at some point fixed overhead costs become insignificant anyway. It's hard to say when the combined benefits to growing further will stop being significant. But it can certainly push small miners out of business and make it that much harder and riskier to start a new mining company.

One economy of scale that sort of works as a disadvantage to larger miners is heat dissipation: it's a easier to get rid of waste heat (or even to use the heat productively) at small to medium scale. Large concentrated miners need special equipment to get rid of heat or it prevents them from operating in hot climates.

1

Larger blocks take longer to propagate. However, when you have solved a block yourself, there is no delay (regardless of the size), because you already have all of the block data and have already checked it.

So, let's say a miner has 40% of the network hashing power. They solve 40% of the blocks. That means 40% of the time, they have essentially NO latency in starting to work on the next block, regardless of block size. Smaller miners who might mine 1% of the blocks have to wait for the latest block to be propagated to them 99% of the time. So the larger miner will be affected by block propagation delays 60% of the time, whereas the smaller miner is affected by block propagation delays 99% of the time.

Since mining operations with a smaller portion of the hash power are disadvantaged in this way, it can lead to decreased profits and encourages centralization into a few large pools.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.