Author: Karl-Johan Alm 2018-05-18 06:28:39
Published on: 2018-05-18T06:28:39+00:00
In a recent bitcoin-dev email thread, Matt Corallo expressed concerns about the size of filters making existing Simplified Payment Verification (SPV) clients reluctant to adopt BIP 158. He suggested exploring ways to reduce filter sizes such as providing filters for certain script templates, e.g., segwit version X outputs only. Additionally, Corallo proposed multi-block filters where light clients download pairs of filters for blocks X..X+255 and X+256..X+511. They would check for matches and grab pairs for any that matched. The added benefit of this approach is you can accept a slightly higher false positive rate for bigger ranges, because the probability of a specific entry having a false positive in each filter is independent. Corallo wrote about this approach in his blog post, but the filter type differs from his experiments.
Updated on: 2023-06-13T02:27:13.422194+00:00