Reading transactions-per-second claims (one school of thought)

A headline transactions-per-second figure is not comparable until it is classified. iKnow Blockchain records every speed figure as a theoretical peak, a lab benchmark or an observed window, with its hardware and bandwidth assumptions and whether a home full node could keep up, and never compares figures of different classes. The home-node question comes from one school of thought: raising base-layer throughput raises the cost of full verification, which narrows who can independently verify the chain, so a figure that ordinary nodes cannot follow is read with that cost in mind. A counter-view holds that base-layer capacity can grow with hardware and bandwidth, that a stated protocol limit can be a real design parameter, that most users verify through light clients or proofs, and that added layers bring their own trust assumptions.

Key idea. A speed number means little without who measured it, on what hardware, over what window, and how ordinary users can verify the result.

Example

A theoretical peak computed from block limits is not comparable with transactions observed over a week on another chain; block interval, finality time and verification requirements usually say more.

Keep in mind

In this school's view, theoretical peaks often ignore propagation, state growth and validation cost on ordinary hardware; in the counter-view, a peak derived from a stated protocol limit can describe a real design parameter. In either view, a peak is not comparable with observed load.
Block fullness and mempool saturation are out of scope until a reproducible, published method exists.
Observed throughput reflects demand in a window, not capacity.

Sources

Provenance

Comparison design card, profile schema Compare Contract r1.3.

Report an error in this lesson