V3 transaction policy for anti-pinning

V3 transaction policy for anti-pinning

Original Postby sdaftuar

Posted on: February 8, 2024 13:51 UTC

Recent research conducted on the effects of the v3 rules on transactions observed by a datalogging node in the network has been published, providing new insights into the dynamics of transaction spends within the Lightning Network.

The study, accessible at, includes a histogram that illustrates an intriguing characteristic: most anchor spends are of small size.

This finding prompts a reassessment of the maximum child size for v3 transactions. Currently standing at 1000 virtual bytes, there's a suggestion to consider lowering this threshold to minimize the extra transaction size that might incur costs when replacing by fee (RBF) another v3 child. This proposal is particularly relevant to the Lightning Network wallet use case, as it necessitates a balance between the utility of creating larger Child Pays for Parent (CPFP) transactions and the associated pinning costs.

The choice of the max v3 child size parameter appears to be a tradeoff, weighing the benefits against the potential expenses. For other applications of v3 that do not involve the creation of children, abstaining from having any children may be advantageous. However, for the LN wallet context, the decision hinges on optimizing transactional efficiency and cost-effectiveness. The dialogue is open for further discussion and input from experts like @t-bast and @MattCorallo to refine these parameters and enhance the transaction process within the network.