Why Block Propagation Time Matters

A block produced in one place must reach nodes everywhere before it can be built upon. The time that takes constrains everything else.

The problem

If the next block is produced before the previous one has propagated, some nodes build on the older block. Two competing chains briefly exist, and one is discarded.

Discarded blocks waste the work that went into them and, more importantly, weaken security: an attacker competes against the effective work of the network rather than its total work.

What determines propagation time

Block size. Larger blocks take longer to transmit.

Network topology. How many hops between distant nodes.

Validation time. Nodes typically verify before relaying, so verification cost adds to propagation.

Geographic distribution. A globally distributed network has an irreducible latency floor.

The consequences for design

Block time cannot be much shorter than propagation time without producing frequent forks.

Block size cannot grow without lengthening propagation, which either lengthens block time or increases fork rate.

This is the underlying constraint behind the block size debates. Larger blocks mean either slower blocks or more forks, and the third option, requiring better connectivity from node operators, reduces how many people can run one.

The mitigations

Compact block relay. Rather than sending a whole block, a node sends identifiers for transactions the recipient likely already has, and the recipient reconstructs it. Dramatically reduces bytes transmitted.

Relay networks. Dedicated infrastructure optimised for fast block distribution among producers.

Validation before relay, or after. Relaying before full validation is faster and risks propagating invalid data. Implementations vary.

Why this explains throughput claims

A chain advertising very high throughput has made a choice about this constraint.

Usually the choice is higher node requirements: faster connections, better hardware, fewer participants able to keep up. That is a legitimate trade and the honest comparison is not transactions per second but what it costs to run a node that validates them.

A chain where validation requires a data centre connection has purchased its throughput by reducing the number of independent verifiers, which is the property most people think they are getting from a blockchain in the first place.

The user-facing consequence

Confirmation times and the number of confirmations required before a transaction is considered settled both derive from this.

A chain with short block times but frequent reorganisations requires more confirmations to reach the same assurance. That is why venues wait different numbers of blocks on different networks, and those thresholds, published by platforms such as a platform with real on-chain withdrawals, are the market’s practical assessment of each chain’s settlement quality.