Posted by optout
Jun 12, 2026/07:19 UTC
The concept of utilizing larger filters instead of smaller, individual ones for each block has been proposed to streamline the process wherein a client would initially download large filters covering multiple blocks (specifically, a segment of 2016 blocks). Upon detecting a match within these large filters, the client would then proceed to fetch the corresponding smaller filters related to that specific range. This approach suggests an aggregated method of filtering, potentially simplifying the initial stages of data handling.
Upon evaluation, it appears that the size benefits of employing a single large filter per 2016 blocks, in comparison with numerous small filters for each block, are marginal. Detailed measurements indicate that the large filter achieves a mere 0.12% reduction in size relative to the aggregate of small filters for the same number of blocks. This negligible decrease in size might not sufficiently compensate for the increased complexity introduced by this method. Such a minimal reduction could be considered insufficient to justify complicating the filtering system, suggesting that the current method of using individual filters for each block may still be the more efficient option.
Thread Summary (10 replies)
Apr 19 - Jun 12, 2026
11 messages
TLDR
We’ll email you summaries of the latest discussions from high signal bitcoin sources, like bitcoin-dev, lightning-dev, and Delving Bitcoin.
We'd love to hear your feedback on this project.
Give Feedback